definitions of recklessness known to English law

1. Introduction:

There are two definitions of recklessness known to English law: subjective /adevertent/Cunningham recklessness, and objective/inadvertent/Caldwell recklessness. Of the two, subjective recklessness is more important, and can be regarded as the standard variety of recklessness. Some crimes use one and some the other.

2. Advertent Recklessness:

It was in Cunningham(1957) that the Court of Criminal Appeal held that, in a statute, the term ‘malicious’ denotes intention or recklessness, and that recklessness means that ‘the accused has foreseen that the particular kind of harm might be done and yet has gone on to take the risk of it’. There are essentially three elements in this definition, and they are the same ones found in the Model Penal Code’s definition of recklessness as ‘the conscious taking of an unjustified risk’. First, it requires D’s actual awareness of the risk; this is why it is referred to as advertent recklessness, and it is regarded as the key element in bringing recklessness within the concept of mens rea. The second element is that a person may be held to have been reckless if he or she was aware of any degree of risk. The third element is that the risk which D believes to be present must be an unjustified or unreasonable one. This is an objective element.

3. What are the problems with this approach?

There are at least two types of awkward case for a test of liability which requires the court to be satisfied that the defendant actually saw the risk, however briefly. One is where a person acts impulsively in the heat of the moment. This is often expressed in ordinary speech by saying ‘I acted without thinking’, or ‘I just didn’t think’. D denies that he or she was aware of the risk at the time of acting. In Parker (1977) D tried unsuccessfully to make a telephone call from a kiosk; in his frustration he slammed down the receiver and broke it. The Court of Appeal upheld his conviction for causing criminal damage recklessly, despite his defence that it did not occur to him that he might damage the telephone. The Court held that he must have known that he was dealing with breakable material, even if that fact was no at the forefront of his mind when he slammed the receiver down.

4. The second problem:

The second problem is the ‘couldn’t care less’ attitude: D might not have thought about a particular consequence, because it was irrelevant to his interests. If this version of events is accepted, D must be acquitted on the advertent definition of recklessness. Antony Duff has argued that these cases can and should be included within the meaning of recklessness, by invoking the concept of ‘practical indifference’. This is ‘ a matter, not of feeling as distinct from action, but of the practical attitude which the action itself displays’. Moreover, it may include cases in which D fails to advert to certain aspects of the situation: ‘what I notice or attend to reflects what I care about; and my very failure to notice something can display my utter indifference to it’. Duff’s argument is that requiring practical indifference is just as subjective, and just as respectful of individual autonomy, as requiring awareness of risk. The question is what D’s attitude was towards the victim’s interests: D’s words and acts provide the basis for determining that attitude.

5. The third possibility:

The third possibility would be where D states that he was so preoccupied with other aspects of what he was doing as to give no thought to a particular consequence (although the courts might be reluctant to accept such a defence)

6. Caldwell Recklessness:

It is important to note that Caldwell recklessness does not in any sense replace Cunningham recklessness. It is rather that some crimes use Caldwell recklessness and some crimes use Cunningham recklessness. In fact, Caldwell applies only to criminal damage and a few statutory offences. Lord Diplock in Caldwell (1982) formulated the following model direction: a person is guilty of causing damage recklessly if:

a)      he does an act which in fact creates an obvious risk that property would be destroyed or damaged and

b)      when he does the act he either has not given any thought to the possibility of there being any such risk or has recognized that there was some risk involved and has nonetheless gone on to do it.

7. Definition includes the advertent element:

It will be noticed that this definition includes the advertent element by referring to the person who recognizes the risk and takes it, but then goes further, extending to all those who fail to give any thought to the possibility of a risk which may be described as obvious. The formulation in the reckless driving case of Lawrence goes a little further by requiring proof that the risk was obvious and serious, the latter term implying that the ordinary prudent person would not have considered the risk negligible. Returning to the Caldwell formulation quote above, it appears that the requirement that the risk be obvious has to be satisfied independently, no matter which of the two conditions in b) is relied upon. However, the House of Lords held in Reid (1993) that the obviousness requirement applies only when the failing to give thought condition is relied upon.

8. What are the problems with this approach?

a) Ignored the intention of Parliament:

There are four basic objections to Caldwell approach. The first concerns the manner in which the law was changed. Lord Diplock, who delivered the judgment in both Caldwell and Lawrence, was said to have ignored the intention of Parliament whose reform of the law of criminal damage provided the opportunity for the case to be decided. Lord Edmund-Davies, in his dissenting judgment, pointed out that the Criminal Damage Act 1971, the statute which gave rise to Caldwell, was in the main the work of the Law Commission, who in arguing for a modernisation of the law had specifically defined recklessness in its subjective form. The decision used a change in the form of the law to bring about a fundamental change in its substance, and one that went against the proposal of the initiating reform body. It has therefore been argued within the liberal discourse of the rule of law that Lord Diplock stepped outside the canons of legitimate judicial interpretation and did not properly attend to the ‘rules of the legal game’.

b. Interpretation makes easier to convict an individual:

The second objection concerns the effect of interpretation makes it easier to convict an individual because a jury will not, under the objective limb of the test, have to look into the accused’s mind in order to decide whether a risk was foreseen. Simply jury will consider whether a reasonable person would have foreseen the risk. Considerably more acts of damage potentially come within the purview of the criminal justice system, and it becomes easier for the prosecution to gain convictions.

The objection to this is again a liberal one. Not only does Caldwell open up the field of liability, but also, in conjunction with later decisions (Elliot v. C; 1983); Stephen Malcolm R (1984), it permits conviction where principles of subjective individual responsibility would otherwise deny it. Caldwell criminalises people unjustly, and offends against the principles of individual justice upon which the criminal law is supposedly based. Under the current law, the schizophrenic tramp who burns down a haystack (Stephenson (1979), and the fourteen year-old educationally subnormal girl who sets fire to a shed (Elliott v. C), can be convicted of acts of criminal damage. This is regardless of their inability, through no fault of their own, to match the standard of the reasonable person. Although there have been indications of a change of mind in the House of Lords (R v. Reid 1992), it is sufficient that there is a risk of damage obvious to the reasonable person. That the accused could not attain the necessary standard of care in his conduct is rough luck.

c. Caldwell Recklessness can not be termed mens rea:

Third objection is that Caldwell recklessness cannot properly be termed mens rea, because it is not a state of mind. A person who fails to give thought to a consequence does not have a state of mind in relation to that consequence. But this presupposes that the only proper ground for ascribing blame for serious offences is advertence-in other words, that the minimum requirements for criminal culpability should be that the harmful consequence passed through D’s mind.

d. Two competing tests of recklessness:

The fourth objection is that as a result of Caldwell there now exist two competing tests of recklessness within the criminal law. For criminal damage offences, Caldwell makes it clear that it will suffice if the accused has failed to advert to an obvious risk. Later cases have established that it is to the reasonable person, sharing none of the characteristics of the accused, that the risk must be obvious (Elliott v. C); Stephen Malcolm R. In the area of non-fatal offences against the person, the older test of advertent recklessness espoused in the case of Cunningham still has force (Savage and Parmenter (1991)). Other areas of law such as rape and involuntary manslaughter have also undergone periods of upheaval as a result of Caldwell. However, the existence of these two tests has no rationale.

e. Contradiction within Caldwell approach:

The difficulty with Caldwell is that the model direction makes no mention of the person who recognizes the risk but believes that it can be eliminated. This has led commentators to argue that there is a gap or loophole which may be exploited to gain an acquittal in appropriate cases: if D has recognized the risk, he is not in Lord Diplock’s category of failing to give thought to it; and if D believes the risk has been eliminated, he has not gone on to take the risk. The argument was run before the Divisional Court in Chief Constable of Avon and Somerset v. Shimmen (1987The Divisional Court did not gainsay the loophole argument, but they directed that D should be convicted because he had said in evidence that he thought he had eliminated as much risk as possible. This means that he did not think he had eliminated all risk, which in turn means that he knew there was a slight risk, and that is enough for subjective recklessness. True cases of the Caldwell loophole will be rare and, as Lord Goff observed in Reid, a defendant would in practice have to point to some specific fact as to which he was mistaken and which, if true, would have excluded the possibility of risk. If the loophole argument is assessed morally can it be maintained that there is less culpability in erroneously believing that one has eliminated all risk than in failing to give thought to the risk at all? The difference is merely that between mistake and accident: in the former case D is under a misunderstanding which leads to harm; in the latter case D fails to consider the consequences of actions and causes harm. The former amounts to an omission to discover the true nature of the risk; the latter amounts to an omission to think about the existence of the risk. It is by no means clear that the Caldwell test is logically consistent in excluding the former and including the latter.

9. The right solution:

a. Incompatible approaches:

Here is the dilemma for the present law of recklessness. It consists of two incompatible approaches where neither by itself seems right. If the orthodox subjectivist approach is too narrow, ignoring the ‘capacity’ form of inadvertent subjectivity, the objectivist approach is too broad and indeterminate. It leaves the boundaries of the substantive law to be established in individual cases on the basis of value judgments about right and wrong. Here, opinions may differ and different social and political perceptions will be brought to bear.

b. The tendency of many text-writers and judges:

However, the tendency of many text-writers and judges has been to accept the advertent or subjective approach, thus excluding from conviction certain people who may be no less culpable than those who are convicted. The decision in Caldwell might be seen as a reaction to this narrow subjectivism, but there are various objections to that development in the law. Successful as it may be in convicting unthinking, bad-tempered, and callously indifferent defendants, it does not yet have an established exception for people who lack the capacity to attain its objective standard.

c. Duff’s test of practical indifference:

The Caldwell test is not the only way of supplementing the narrow conception of moral fault embodied in advertent recklessness. An alternative is Duff’s test of practical indifference, which relies considerably on objective judgements as evidence of a person’s attitude when behaving in a particular way. A further alternative would be to introduce more offences of negligence. It is evident that an approach that focuses solely on advertence fails to capture moral distinctions and to satisfy social expectations. Subjective tests heighten the protection of individual autonomy, but they typically make no concession to the principle of welfare and the associated notion of duties to take care and to avoid harming the interests of fellow citizens.

d. Greater reliance on objective standards:

However, if we are to move towards greater reliance on objective standards, at least two points must be confronted. First, objective tests must be applied subject to capacity-based exceptions. This preserves the principle of individual autonomy by ensuring that no person is convicted who lacked the capacity to conform his or her behaviour to the standard required. Failure to recognize this in the Caldwell decision was a major fault that drew subsequent discussions away from the moral basis of the test itself. Secondly, any improved moral ‘fit’ obtained by moving more towards objective standards must be weighed against the greater detraction from the principle of maximum certainty that is likely to result. Objective standards inevitably rely on terms such as reasonable, ordinary, and prudent. They appear much more compliant and unpredictable than subjective tests that ask whether or not a defendant was aware of a given risk, and they explicitly leave room for courts and even prosecutors to make social judgements about the limits of the criminal sanction.


The Caldwell definition is now of little practical significance and occupies a somewhat isolated position in English law. It applies only to the crime of criminal damage (including arson) and to a few other statutory offences. It formerly applied to reckless driving, as the decisions in Lawrence and Reid demonstrate, but that offence was abolished in 1991 and replaced with dangerous driving. It is no longer applicable to manslaughter. Caldwell is unlikely to be extended to other crimes, and so advertent recklessness remains the dominant test. The approach to recklessness in rape continues to be the ‘couldn’t care less’ test.

11. R v G and another

House of Lords quashed the appellants’ convictions in holding that a person acted recklessly within the meaning of s 1 of the 1971 Act with respect to (i) a circumstance when he was aware of a risk that it existed or would exist; (ii) a result when he was aware of a risk that it would occur; and it was, in the circumstances known to him, unreasonable to take the risk.

Lord Bingham rejected to follow the Caldwell reasoning in this case for four reasons, taken together.

First, it is a salutary principle that conviction of serious crime should depend on proof not simply that the defendant caused an injurious result to another but that his state of mind when so acting was culpable.  This is the meaning of the familiar rule actus non facit reum nisi mens sit rea.  The most obviously culpable state of mind is no doubt an intention to cause the injurious result, but knowing disregard of an appreciated and unacceptable risk of causing an injurious result or a deliberate closing of the mind to such risk would be readily accepted as culpable also.  It is clearly blameworthy to take an obvious and significant risk of causing injury to another.  But it is not clearly blameworthy to do something involving a risk of injury to another if one genuinely does not perceive the risk.  Such a person may fairly be accused of stupidity or lack of imagination, but neither of those failings should expose him to conviction of serious crime or the risk of punishment.

Secondly, the present case shows, more clearly than any other reported case since R v Caldwell that the model direction formulated by Lord Diplock is capable of leading to obvious unfairness.  As the trial judge regretted the direction he felt compelled to give, and it is evident that this direction offended the jury’s sense of fairness.  A law which runs counter to that sense must cause concern.  Here, the appellants could have been charged under s 1(1) of the 1971 Act with recklessly damaging one or both of the wheelie-bins, and they would have had little defence.  As it was, the jury might have inferred that boys of the appellants’ age would have appreciated the risk to the building of what they did, but it seems clear that such was not their conclusion.  On that basis the jury thought it unfair to convict them.  I share their sense of unease.  It is neither moral nor just to convict a defendant (least of all a child) on the strength of what someone else would have held if the defendant himself had no such apprehension.  Nor, the defendant having been convicted is the problem cured by imposition of a nominal penalty.

Thirdly, I do not think the criticism of R v Caldwell expressed by academics, judges and practitioners should be ignored.  A decision is not, of course, to be overruled or departed from simply because it meets with disfavour in the learned journals.  But a decision which attracts reasoned and outspoken criticism by the leading scholars of the day, respected as authorities in the field, must command attention.

Fourthly, the majority’s interpretation of ‘reckless’ in s 1 of the 1971 Act was, as already shown, a misinterpretation.  If it were a misinterpretation that offended no principle and gave rise to no injustice there would be strong grounds for holding to the misinterpretation and leaving Parliament to correct it if it chose.  But this misinterpretation is offensive to principle and is apt to cause injustice.  That being so, the need to correct the misinterpretation is compelling

For the decision in R v Caldwell was made more than 20 years ago.  Its essential reasoning was unanimously approved by the House in R v Lawrence.  Invitations to reconsider that reasoning have been rejected.  The principles laid down have been applied on many occasions, by Crown Court judges and, even more frequently, by justices.  In the submission of the Crown, the ruling of the House works well and causes no injustice in practice.  If Parliament had wished to give effect to the intention of the Law Commission it has had many opportunities, which it has not taken, to do so.  Despite its power under Practice Statement (Judicial Precedent) to depart from its earlier decisions, the House should be very slow to do so, not least in a context such as this.


In my view the very high threshold for departing from a previous decision of the House has been satisfied in this particular case.  In summary I would reduce my reasons to three propositions.  First, in R v Caldwell the majority should have accepted without equivocation that before the passing of the 1971 Act foresight of consequences was an essential element in recklessness in the context of damage to property under s 51 of the Malicious Damage Act 1861.  Secondly, the matrix of the immediately preceding Law Commission recommendations shows convincingly that the purpose of s 1 of the 1971 Act was to replace the out of date language of ‘maliciously’ causing damage by more modern language while not changing the substance of the mental element in any way.  Foresight of consequences was to remain an ingredient of recklessness in regard to damage to property.  Thirdly, experience has shown that by bringing within the reach of s 1(1) cases of inadvertent recklessness the decision in R v Caldwell became a source of serious potential injustice which cannot possibly be justified on policy grounds.


In the case before the House the two boys were 11 and 12 respectively.  Their adventure of camping overnight without their parents’ permission was something that many children have undertaken.  But by throwing lit newspapers under a plastic wheelie-bin they caused £ 1m of damage to a shop.  It is, however, an agreed fact on this appeal that the boys thought there was no risk of the fire spreading in the way it eventually did.  What happened at trial is highly significant.  The jury were confused by the Caldwell directions which compelled them to treat the boys as adults and to convict them.  The judge plainly thought this approach was contrary to common sense but loyally applied the law as laid down in R v Caldwell.  The view of the jurors and the judge would be widely shared by reasonable people who pause to consider the matter.  The only answer of the Crown is that where unjust convictions occur the judge can impose a lenient sentence. This will not do in a modern criminal justice system.  Parliament certainly did not authorise such a cynical strategy.

These developments show that what Lord Diplock described in R v Caldwell as a mysterious meaning of recklessness was also consistent with the general trend of the criminal law.


The surest test of a new legal rule is not whether it satisfies a team of logicians but how it performs in the real world.  With the benefit of perception the verdict must be that the rule laid down by the majority in R v Caldwell failed this test.  It was severely criticised by academic lawyers of distinction.  It did not command respect among practitioners and judges.  Jurors found it difficult to understand: it also sometimes offended their sense of justice.  Experience suggests that in R v Caldwell the law took a wrong turn.

That brings the question whether the subjective interpretation of recklessness might allow wrongdoers who ought to be convicted of serious crime to escape conviction.  Experience before R v Caldwell did not warrant such a conclusion.  In any event, as Lord Edmund-Davies explained, a defendant closes his mind to a risk he must realise that there is a risk and, on the evidence, that will usually be decisive.  One can trust the realism of trial judges, who direct juries, to guide juries to sensible verdicts and juries can in turn be relied on to apply robust common sense to the evaluation of ridiculous defences.  Moreover, the endorsement by Parliament of the Law Commission proposals could not seriously have been regarded as a charter for the acquittal of wrongdoers.

i. What are the justifications for this approach?

The justifications for the advertent definition or recklessness are grounded in the principle of individual autonomy and the importance of respecting choice. The distinction between recklessness and negligence turns on D’s awareness or unawareness of the risk. In both cases there is an unreasonable risk taken, but D should only be held to have been reckless if he or she was aware of the risk. A person who is aware of the risk usually chooses to create it or to run it, and therefore chooses to place his or her interests above the well-being of those who may suffer if the risk materializes. Choosing to create a risk of harmful consequences is generally much worse than creating the same risk without realizing it. Moreover, holding a person reckless despite unawareness of the risk would result in a conviction in a case like Stephenson (1979). D, a schizophrenic, made a hollow in a haystack in order to sleep there; he felt cold, and so lit a small fire, causing the whole haystack to go up in flames, and resulting in damage of some £3, 500. The defence relied on medical evidence that D may not have had the same ability to foresee the risk as a mentally normal person. The Court of Appeal, quashing D’s conviction, held that the definition of recklessness clearly turned on what this defendant actually foresaw, and the medical evidence should have been taken into account on this point. This decision, then, strongly affirms the element of individual fairness in the advertent or subjective definition. An entirely objective test would exclude this.

i. What are the reasons to extend recklessness beyond the advertent definition?

Lord Diplock offered three reasons. One was that the dividing line between awareness and unawareness of risk is no narrow and so difficult to prove that juries and magistrates should not be required to labour over it. The factual basis of this must be accepted: it may be virtually impossible to know whether or not a person was fleetingly aware of the consequences of his action. A second reason given by Lord Diplock was that the term reckless is wider than awareness of risk, and includes lack of care and lack of thought. That may be true, but it does not solve the problem of how wide the law’s definition should be. The third reason given by Lord Diplock is the only one worthy of being called a justification: that it may be no less blameworthy for a person to fail to foresee an obvious risk than it is to see the risk and knowingly to take it. In other words, Lord Diplock challenged the common law distinction between recklessness and negligence on the ground that it fails to draw the line in the right place. He further attacked fundamental conceptions of responsibility by arguing that the idea of mens rea, as encompassing intention and subjective recklessness, is unsatisfactory because it omits some equally culpable cases. As I have noted above, some of the supporters of advertent recklessness accept that it is under-inclusive, in that they have attempted to stretch it to include actions during fits of temper and actions out of indifference.