Abstract
There are current arguments in criminal-law literature for alternative approaches to causation to be used in English criminal law. The most recent proposal has come in the form of replacing the standard tests of causation with a single test, known as ‘INUS’ (or ‘NS’) causation – where a cause is a
Introduction
For a defendant (D) to be criminally liable for result crimes
1
in England and Wales, it must be shown that D's conduct caused the prohibited outcome. In English criminal law, there are standard tests for causation – legal doctrines that govern the causal connection between D's conduct and prohibited outcome. Many debates in criminal-law literature have focused on the problems to which the standard tests for causation give rise.
2
This has resulted in some legal scholars proposing alternative accounts of causation to be used in English criminal law.
3
The most recent argument has come in the form of proposing to replace the standard tests of causation with a single test, known as ‘INUS’ (or ‘NS’) causation – where a cause is a
This article discusses one of the potentially problematic implications of adopting the INUS account of causation in English criminal law. It analyses the implications INUS would have for strict liability 6 result crimes involving injury or death by driving. In particular, s.3ZB of the Road Traffic Act (RTA) 1988 and s.12A(2)(b) of the Theft Act (TA) 1968, by way of case studies. It focuses on these driving offences because they are the two main stigmatic strict liability offences in English criminal law that appear to have occupied the appellate courts in considering causation, and which raise causal problems. It argues that INUS would affect the liability outcomes in cases involving s.3ZB RTA 1988 (hereinafter, s.3ZB) and s.12A(2)(b) TA 1968 (hereinafter, s.12A) so that instead of being acquitted, the defendants would be criminally liable for their respective offences. This is because INUS removes the blameworthy requirement which has been injected into the standard tests of causation. However, such liability outcomes would yield fair-labelling concerns for these crimes. This highlights the fair-labelling concerns with imposing strict liability in stigmatic offences more broadly. It concludes that the fair labelling issues in strict liability offences involving injury or death by driving could be addressed by importing a blameworthy requirement for the aggravated actus reus elements of the offences generally provided by D's careless driving. This would ensure that these offences appropriately capture the wrongdoing that they aim to target.
This article starts, in the ‘Strict Liability Offences Involving Injury or Death by Driving’ section, by canvassing the strict liability offences involving injury or death by driving, namely s.3ZB and s.12A. The ‘INUS Problematising Strict Liability Offences Involving Injury or Death by Driving’ section then examines the implications of INUS for s.3ZB and s.12A and uses cases drawn from these offences. It contends that, according to INUS, the liability outcomes in cases involving s.3ZB and s.12A would alter so that the defendants would be criminally liable for their respective offences. This is because INUS does not have a blameworthy requirement, unlike the standard tests of causation. However, such liability outcomes would be problematic in terms of fair-labelling.
Then, the ‘Is Strict Liability in Stigmatic Offences Ever Justified?’ section discusses the arguments for the imposition of strict liability in stigmatic offences such as s.3ZB and s.12A. It argues that strict liability is generally beneficial in minor non-stigmatic crimes for practical reasons. However, in more serious stigmatic crimes, like s.3ZB and s.12A, practical reasons are outweighed by the need for fair labelling. Therefore, due to fair-labelling concerns, strict liability is not appropriate in these types of offences.
Subsequently, the ‘Remedying the Concerns: Exiling Strict Liability in Stigmatic Offences’ section, discusses the potential remedies for the fair-labelling concerns that INUS causation would produce for offences such as s.3ZB and s.12A. It argues that the issue can be addressed by importing a blameworthy element into these types of offences, specifically by importing a blameworthy requirement for the aggravated actus reus elements of the offences generally provided by D's careless driving.
Strict Liability Offences Involving Injury or Death by Driving
This section will discuss the common-law developments of strict liability result crimes involving injury or death by driving, namely s.3ZB and s.12A.
Section 3ZB Road Traffic Act 1988
Pursuant to s.3ZB RTA 1988: [a] person is guilty of an offence under this section if he causes the death of another person by driving a motor vehicle on a road and, at the time when he is driving, the circumstances are such that he is committing an offence under — (a) section 87(1) of this Act (driving otherwise than in accordance with a licence), (b) … or (c) section 143 of this Act (using motor vehicle while uninsured).
The base offence under s.3ZB of driving a car uninsured or unlicensed is an offence of strict liability since it can be committed without any fault or blame on the part of the driver. The aggravating element of causing the death of another can also be constituted by an event for which D is not culpable. As Sullivan and Simester remark, s.3ZB ‘is a stark example of strict liability. No culpability of any kind on the part of D is required in the matter of causing V's death’. 7 This echoed the view of the Court of Appeal in Williams 8 and Hughes. 9 In both these cases, the defendants killed their respective victims (V) with their vehicles whilst driving faultlessly, but uninsured and unlicensed. In Williams, V stepped out in front of D's vehicle, and in Hughes, D was involved in a fatal road collision with V who was oncoming in the opposite lane. As an ordinary matter of statutory interpretation, the Court of Appeal on these cases ruled that blame on behalf of the defendants in respect of their victim's death was irrelevant and causation could be established by the mere presence of the defendants being on the road and being involved in a road traffic collision. The defendants were therefore found to have caused the deaths within the terms of s.3ZB.
The Government intended s.3ZB to be a strict liability offence. In the consultation process of s.3ZB, the Government stated that: [t]he standard of driving could be perfectly acceptable. For example, this offence could bite on a driver who was driving very carefully, but a child ran out into the road and was killed. The offence will apply where “but for” D's car being on the road the person would not have been killed.
10
However, the Supreme Court in Hughes 11 did not want s.3ZB to impose a conviction upon a driver who, other than lacking a valid insurance or driving licence, was entirely blameless for V's death. 12 The Court worried about: (1) the probability of harsh liability findings occurring 13 and (2) the severity of sentence imposed (the sentence may be a term of imprisonment). 14 To use the Supreme Court's example, if D was driving faultlessly but unlicensed owning, say, to some sort of clerical error, D would be liable under s.3ZB if V, a pedestrian intent on suicide, stepped into the path of D's vehicle and died, and thus at risk of imprisonment for a substantial term. 15 In effect, the Court was concerned about convicting and sentencing individuals of a serious offence on account of an aggravating element for which they bear no culpability.
In order to deal with such concerns, the Supreme Court interpreted the causal component in s.3ZB to require an element of fault on behalf of D's driving, thereby imputing a blameworthy requirement into the causal analysis and thus circumventing the Government's intention to create a strict liability offence. The Supreme Court held that s.3ZB requires the Crown to prove ‘something open to proper criticism in the driving of the defendant, beyond the mere presence of the vehicle on the road’.
16
In reversing and overruling the Court of Appeal in Hughes and Williams, respectively, the causal component in s.3ZB was interpreted to require: at least some act or omission in the control of the car, which involves some element of fault, whether amounting to careless/inconsiderate driving or not, and which contributes in some more than minimal way to the death.
17
The Supreme Court's decision on s.3ZB in Hughes has been confirmed in several cases where D was involved in a fatal accident whilst driving uninsured and/or unlicensed but driving faultlessly, including McGuffog
18
and the conjoined case of Uthayakumar.
19
In the former case, M was involved in a fatal collision with a cyclist, who failed to stop at a roundabout. In the latter case, U was involved in a collision with V, a pedestrian, who had walked out in front of U, and C was involved in a road collision with V who had veered over the centre line of the road into the oncoming lane. In Uthayakumar, the court noted that: [t]he judgment of the Supreme Court in Hughes is clear. The use of the phrase ‘causing death by driving’ in section 3ZB, taken in context, means the Crown must prove ‘something open to proper criticism in the driving of the defendant, beyond the mere presence of the vehicle on the road, and which contributed in some more than minimal way to the death’ (see paragraph 33). Section 3ZB requires ‘at least some act or omission in the control of the car, which involves some element of fault, whether amounting to careless/inconsiderate driving or not, and which contributes in some more than minimal way to the death’. It is not necessary that such act or omission be the principal cause of death.
20
Section 12A Theft Act 1968
In accordance with s.12(1) TA 1968: … a person shall be guilty of an offence if, without having the consent of the owner or other lawful authority, he takes any conveyance for his own or another's use or, knowing that any conveyance has been taken without such authority, drives it or allows himself to be carried in or on it. (1) … a person is guilty of aggravated taking of a vehicle if — (a) he commits an offence under section 12(1) above (in this section referred to as a “basic offence”) in relation to a mechanically propelled vehicle; and (b) it is proved that, at any time after the vehicle was unlawfully taken (whether by him or another) and before it was recovered, the vehicle was driven, or injury or damage was caused, in one or more of the circumstances set out in paragraphs (a) to (d) of subsection (2) below. (2) The circumstances referred to in subsection (1)(b) above are — … (b) that, owing to the driving of the vehicle, an accident occurred by which injury was caused to any person.
For this offence - aggravated vehicle taking - the base offence of knowingly taking a motor vehicle without consent is a fault-based offence since it can only be committed when D knows that he does not have consent to take the vehicle. As s.12(6) TA 1968 states: [a] person does not commit an offence under this section by anything done in the belief that he has lawful authority to do it or that he would have the owner's consent if the owner knew of his doing it and the circumstances of it.
This was the view of the Court of Appeal in Marsh
21
and Taylor.
22
In Marsh, D had given a barmaid a lift home to change her clothing using her landlord's car without permission. Through no fault of D, a woman crossed the road and was knocked to the ground resulting in injury. The Court held that to establish an offence under s.12A, there was no requirement that the driver of a vehicle, which had been taken without its owner's consent and was then involved in an accident resulting in personal injury, should be at fault for his driving. The Court stated: [a]pplying ordinary canons of statutory construction, it is impossible to say that the words of section 12A(2)(b) import a requirement of fault in the driving of the vehicle. No word suggesting fault appears in the statutory language. It seems to us that the ordinary meaning of the words used is simply to point to a requirement that there be a causal connection between the moving of the vehicle on the road and an accident which follows.
Like s.3ZB, the Government also intended s.12A to be a strict liability offence. In debating the Aggravated Vehicle-Taking Bill on the floor of the House of Lords, Ferrers, the Minister of State for the Home Office at the time, stated that: [i]n order to obtain a conviction for aggravated vehicle-taking, the prosecution will have to prove both that the accused committed the basic Theft Act offence, either as a driver or as a passenger, and that an aggravating event occurred during the time when the car was taken. …It is true that the prosecution need not prove that a particular defendant was personally responsible for a particular aggravating event.
23
However, the Supreme Court in Taylor,
24
like in Hughes, was also concerned about the possibility of harsh liability findings occurring if a blameworthy element on behalf of D's driving was not imposed in such cases. The Court worried that: [i]f the requirement of causation is satisfied by the mere fact that the taking of the vehicle accounted for its being in the place where the accident occurred, then all of the anomalous consequences which this court regarded as extraordinary in R v Hughes apply equally to the offence under section 12A. It means that the defendant is liable to be convicted and sentenced to a long period of imprisonment on account of an aggravating factor for which he bears no responsibility.
25
at least some act or omission in the control of the car, which involves some element of fault, whether amounting to careless/inconsiderate driving or not, and which contributes in some more than minimal way to the death.
27
INUS Problematising Strict Liability Offences Involving Injury or Death by Driving
The cases discussed above are, in effect, cases which involve accidents whilst driving unlawfully. In these cases, the standard tests of causation in English criminal law do not render the conduct of the defendants a cause of the death of or injury to the victims. The standard tests therefore do not impose liability in those cases. This section will analyse the implications adoption of the INUS approach to causation will have on these cases. It argues that the conduct of the defendants is a cause of death or injury of the victims on the INUS account. This is essentially because INUS removes the requirement of blameworthiness that is currently injected into causation in English criminal law and causation can be established by the mere existence of the vehicles on the road. Therefore, according to INUS, the liability outcomes would change so that instead of being acquitted, the defendants would be criminally liable for their respective offences. However, such liability outcomes would be problematic due to fair-labelling concerns.
The Implications of INUS
According to INUS causation, a cause is an NS condition: A necessary part of a sufficient ‘condition’ (i.e. a set of factors) for the effect. For efficiency reasons, it helps to apply INUS causation in reverse. In other words, we establish the sufficient condition first, and then consider the necessary factors for that condition. For example, where an electrical short-circuit triggered a house fire. In this situation, the condition sufficient for the house fire is the conjunction of the short-circuit, oxygen and nearby flammable material. In this sufficient condition, the short-circuit is necessary for the fire as the short-circuit was indispensable: The other factors, conjoined with one another in the absence of the short-circuit, would not have produced the fire. In other words, had the short-circuit not occurred, the fire would not have occurred in the manner that it did. Therefore, the short-circuit was a cause of the fire – it was an NS condition of the fire.
In the cases where there has been a collision between the vehicles of D and V, such as Hughes, McGuffog, Taylor and U in Uthayakumar, a condition sufficient to bring about the death of V is the collision which involves V driving on the road and D driving on the road. In this condition, D's driving on the road is necessary: if D had not driven the car and had not been in that position at the time, V's death would not have occurred. Hence, D's driving is an NS condition of V's death.
The same analysis is applied in the cases where there has been a collision where V has stepped out in front of D's vehicle, such as Williams, Marsh and C in Uthayakumar. In these cases, a condition sufficient to bring about V's death or injury is the collision which involves V stepping out into the road and D's driving on the road. In this condition, D's driving on the road is necessary: if D had not driven the car and had not been in that position at the time, V's death or injury would not have occurred. Hence D's driving is an NS condition of V's death or injury.
Because causation would be established in these cases, INUS would therefore impose liability on the defendants for their respective offences. It changes all liability outcomes in these cases back to the Court of Appeal decisions which supports the Government's aspirations for the offence but goes against the Supreme Court's position. However, as will be discussed next, this is problematic in terms of fair labelling.
The Issue of Fair Labelling
The principle of ‘fair labelling’ was originally recognised by Ashworth.
28
The concern of fair labelling, Ashworth states: is to see that widely felt distinctions between kinds of offences and degrees of wrongdoing are respected and signalled by the law, and that offences are subdivided and labelled so as to represent fairly the nature and magnitude of the law-breaking.
29
Chalmers and Leverick offer two justifications for this principle. The first concerns the right against wrongful condemnation and stigma: offence names must not ‘create a false or misleading impression of the nature or magnitude of the offender's wrongdoing or encourage an inaccurate conclusion to be drawn’. 31 The second relates to the communicative function of offence names: they ‘communicate information about the offender to a number of different bodies … [who] may form opinions or make decisions about the offender that turn on the information received’. 32 Fair labelling is therefore important because there are far-reaching labelling consequences for defendants that follow conviction. Within the criminal justice system, these consequences relate to conviction and sentencing; outside it, such consequences include stigma and public condemnation which can have negative social effects. 33 The need for offences to fairly represent the nature of D's wrongdoing is commonly accepted as one of various principles governing criminal liability, 34 and has been hugely influential across many different areas of criminal law. 35
In describing offences, one of the main concerns for fair labelling is the description attached to D's wrongdoing.
36
Fair labelling requires that the nature of D's wrongdoing is accurately described.
37
S.3ZB is a form of homicide since it criminalises the killing of another person. The wrong in homicide offences is the culpable killing of another,
38
which is what s.3ZB attempts to target. The Government was of the view that ‘[t]he fault element of this offence would derive from the driver being on the road when he should not have been’.
39
This was confirmed in Hughes when the Crown argued that the purpose of s.3ZB was: to impose criminal liability for a death if it involved the presence of the defendant at the wheel of a car on the road where he had no business to be. The fault is sufficient, it is said, in driving et all when he had no right to be on the road.
40
A plausible explanation for why culpability is assigned to D for the additional actus reus element in constructive crimes 42 generally is that the possibility of the additional actus reus occurring is not random and is related to the initial wrong that D commits through the base offence. In other words, the additional actus reus is an ‘intrinsic’ risk to the wrongfulness of the base offence. 43 Consider the ‘GBH Murder’ case of Vickers, 44 where D, with the intention of causing only serious harm, attacks V which results in V's death. Here, the risk of death is intrinsic to the wrong that D does in intentionally causing serious harm. The likelihood of death occurring from such harm forms part of the reason why inflicting serious harm intentionally is wrongful and should be avoided. 45 In failing to engage adequately with the reasons why he ought not to cause serious harm, D fails to engage adequately with the reasons why he should not cause death. 46 So, not only does D display defective use of his capacities in respect of serious harm, but he also exhibits such deficiencies in respect of death. As a result, there seems no difficulty about holding D blameworthy for V's death in such circumstances. However, in the context of s.3ZB, the risk of V's death is ‘extrinsic’ to the wrongfulness of the base offence: it is unrelated to the wrong that D does in driving unlicensed or uninsured. Homicide is a very different type of offence from driving unlicensed or uninsured, and the prospect of death does not supply reasons why driving unlicensed or uninsured is a wrong – there is little, if any, risk of death resulting from driving unlicensed or uninsured. 47 It is therefore inappropriate to ascribe blame to D for the death of V based on only D's driving unlicensed or uninsured.
Moreover, since s.3ZB represents a form of full strict liability, where both the underlying offence of being unlicensed or uninsured and the aggravating element of causing death can be constituted by events for which D is not culpable, D can therefore be liable for a homicide offence for which he bears no mens rea at all. Thus, s.3ZB does not appropriately capture the wrong of culpable killing. The offence therefore unfairly imposes liability on D for a homicide offence. It makes D criminally responsible as a killer and thus labels D as a culpable killer when D does not culpably kill.
Although s.12A carries a higher sentence if the vehicle is involved in a fatal accident, the death of V is not an element of the offence.
48
Therefore, it is not strictly speaking a homicide offence. As defined by s.12A, the personal injury which constitutes one of the aggravating circumstances must have occurred after the taking of the vehicle and before its recovery, but there is no requirement that D should have been driving it. All that is required is that D was party to the taking of the vehicle and was in, or in the immediate vicinity of, the vehicle when the accident occurred. Indeed, D may have been a passenger as per Wilson.
49
These considerations suggest, then, that the taking and driving of the vehicle is merely incidental to s.12A and it is the injury which is the gravamen the offence. Thus, the wrong that this offence attempts to target is the culpable causing of injury to another. For s.12A, the government was of the view that culpability for causing injury stems from D's role in taking the vehicle in the first place. In debating the Aggravated Vehicle-Taking Bill, the Minister of State for the Home Office stated that: [t]he Government believe that anyone who deliberately involves himself in an episode of illegal car-taking incurs moral responsibility for whatever happens – whether or not he was in the driving seat at the time. If things go wrong, he should take the legal consequences, and he has no one to blame but himself.
50
Thus, both s.3ZB and s.12A misrepresent the wrongfulness of D's conduct (hence the potential of harsh liability findings occurring in both offences if INUS is applied). That is objectionable of itself since it falsely censures D. But, as will be discussed next, it is especially objectionable because of the negative effects that follow conviction. More worryingly, because INUS causation supports the position of the government on these offences – by establishing causation between D's unlawful conduct and V's injury or death – it therefore imposes liability for offences that have fair-labelling concerns, thereby causing injustice in these cases.
Is Strict Liability in Stigmatic Offences Ever Justified?
The problem in dispensing with a blameworthy requirement under INUS causation suggests that we should rethink strict liability in stigmatic offences such as s.3ZB and s.12A. However, arguments exist for the general imposition of strict liability which align with the Government's rationale for these offences. Therefore, strict liability in such offences might be justified. This section discusses those views. It argues that strict liability is generally beneficial in minor non-stigmatic crimes for practical reasons. However, in more serious stigmatic crimes, like s.3ZB and s.12A, practical reasons are outweighed by the need for fair labelling. Therefore, due to fair-labelling concerns, strict liability is not appropriate in these types of offences and so a blameworthy requirement needs to be imputed into them.
Arguments in Favour of Strict Liability
There are ostensibly two main arguments in support of strict liability: (1) strict liability offers procedural convenience for law enforcement administration and (2) strict liability may also increase the deterrent effect of an offence. So, there are practical arguments for allowing strict liability offences. 51
Procedural Convenience
First, the use of strict liability arguably increases the efficiency of criminal justice administration. Proof of subjective mens rea elements such as ‘intention’ or ‘recklessness’ presents a difficulty for the prosecution. It has been argued that this difficulty of proof arises particularly in prosecutions of corporate bodies, ‘since there may be no one person whose conduct can be attributed to the corporation and with whom the corporate mind can be identified’. 52 However, arguably proving mens rea is also burdensome for the prosecution in the context of private individuals. This is because D has ‘peculiar knowledge’ of his state of mind at the time of the offence; only D has privileged access to his intention, knowledge or belief. It is therefore difficult for the prosecution to ascertain subjective mens rea. As Simester states, ‘[t]o prove and know the content of a person's mind is typically harder, because more a matter of inference, than to prove and know the occurrence of physical events’. 53 By contrast, strict liability is much more easily applied to both corporations and the conduct of private individuals since it can be administered without reference to D's mental state. As a result, administration of criminal justice would be greatly slowed if prosecutors are required to prove mens rea with respect to all elements of the actus reus before the court.
The significance of strict liability in this context is that it would increase the efficiency of administering justice if the prosecution did not have to spend time proving mens rea with respect to all elements of the actus reus. This is particularly true with offences that regulate the specialist commercial behaviour of corporations (i.e. regulatory offences), because the prosecution is likely to be inadequately placed to learn the nature of D's own specialist activity in order to prove fault. 54 In turn, the costs involved in adjudicating strict liability offences are also likely to be decreased, since the number of legal elements required to be proved at trial, and so the number of potential issues, is considerably reduced. 55
There are well over 9000 criminal offences in the criminal law of England and Wales. 56 If all the current offences in England and Wales required proof of mens rea in respect of every element of the actus reus, administration of criminal justice would be not merely unmanageable but also too expensive. Criminal adjudications need to be efficient and cost-effective. Simester therefore argues that the state essentially has two options: first, to eliminate several criminal offences; secondly, to simplify some offences by removing some of their elements, in particular their mens rea elements. 57 However, if all the existing offences are in accordance with the harm principle, then the second option is preferred because it allows these harms to be feasibly proscribed. If it is impossible to prevent all proscribed harms while maintaining an efficient criminal justice system, the conflicting interests of all stakeholders (i.e. defendants and other community members, including victims) may be facilitated in part by excluding the mens rea elements of certain criminal offences. 58
Reducing the mens rea element from the definition of an offence can also promote convenience in terms of enforcement. This is seen typically in constructive strict liability offences as they attribute liability to D for the unforeseen consequences of his conduct. Indeed, the need for enforcement convenience formed part of the Government's rationale for enacting s.12A. The Aggravated Vehicle-Taking Act 1992, which inserted s.12A,
59
was brought in to tackle the problem of joyriding – the practice of criminal gangs who take vehicles illegally and then drive them dangerously or cause damage or injury.
60
A particular concern of the Government was that: [a]ll too often it is just not clear which member of a gang caused the particular damage, or who actually had his hand on the wheel when the car crashed and someone was killed or injured as a result. They might all have had a turn; it might have been dark; the car might have been travelling too fast for clear identification; or perhaps the vehicle was found abandoned and burnt out in a wood and the taker simply denies having had any responsibility for it.
61
Nevertheless, absence of empirical evidence, it is not clear that the criminal justice process will operate more efficiently if mens rea is not required to be established at trial. 64 But even if the removal of the requirement to prove mens rea were to increase the efficiency of criminal justice administration, there is a profound moral concern with this argument: it presents an issue of procedural convenience as a reason to do away with one of the key pillars of criminal law – culpability – and thus change substantive law. 65
Deterrence
Another argument in support of strict liability is that it may increase the deterrent effect of an offence, thereby protecting the public. 66 That is to say, the imposition of strict liability will likely cause individuals to take more care in their conduct thus more effectively preventing criminal risk-taking. 67
Such deterrence-based reasoning is implicit in one of the most common arguments for strict liability: that public protection sometimes requires a high standard of care on the part of those who undertake certain risk-creating activities. The rationale for this line of argument is that a high level of care needs to be taken by individuals who perform risk-generating activities, and without strict liability the careless will have little incentive to eliminate or reduce those risks. The threat of strict criminal liability therefore provides motive for individuals involved in risk-creating activities to adopt a higher standard of care which might not otherwise be taken, in order to safeguard against harm. 68
The deterrence-based argument for strict liability seems to be reflected in the law. This is true particularly in regulatory offences where there is a need to protect the public from risks associated with, say, industrial activities where the harmful impact of the actus reus is potentially widespread and severe. For instance, Lord Salmon in Alphacell Ltd v Woodward said that strict liability ‘encourages riparian factory owners not only to take reasonable steps to prevent pollution but to do everything possible to ensure that they do not cause it’. 69
Indeed, the need for deterrence formed part of the Government's rationale for enacting both s.3ZB and s.12A. The Road Safety Act (RSA) 2006, which inserted s.3ZB, 70 was brought in to deal with drivers who put road users at risk through their irresponsible and unlawful behaviour. The Government was concerned that too many people are seriously injured or killed on the roads and sought to reduce the number of those injured or killed. 71 The Government's aspiration is that s.3ZB works as ‘a useful general deterrent effect on bad driving’ thereby reducing the risk of those killed on the roads. 72 As discussed above, s.12A was brought in to tackle the problem of joyriding.
However, there is no empirical evidence to suggest that strict liability is effective in deterring criminal risk-taking. 73 But even if there is a deterrent effect on some conduct, that having strict liability offences will cause individuals to be more careful in some cases, it is arguable that imposing strict liability is no more effective than a standard of culpable negligence. The negligence standard requires individuals to do all that they reasonably can be expected to do to be attentive and careful. As Robinson states, ‘the culpable negligence standard already takes into account the seriousness of the risks, and it already demands greater vigilance to avoid greater risks’. 74 Therefore, the need to increase the deterrent effect of an offence could also be met by a negligence-based standard. 75
Proponents of strict liability in criminal law tend to assume that it might encourage individuals to be even more careful than the circumstances reasonably would require. 76 But even if that were true, such a high standard of precaution (i.e. going beyond what is reasonable in the circumstances) is unnecessary and undesirable because it criminalises reasonable conduct. For instance, it would be reasonable to expect that road users obtain insurance and go about their daily business unless advised that their policy is no longer valid. But if there were a clerical error which invalidated the policy and the road user was not made aware, then that road user would be liable for s.3ZB if they were then involved in a fatal accident. This means that s.3ZB demands going beyond that and requires road users to validate their policy every time they wish to use their vehicles.
This leads to another concern: that strict liability may also desist individuals from even modest risk-creating activity altogether. But it is surely not the intention of criminal law to discourage people from engaging in risk-creating activity. The economy would not be very productive if that were to be the outcome. So, criminal law's deterrent aim must be to deter people from commencing activity (or from performing it in a manner) which creates unreasonable risks of harm occurring. 77 It is therefore appropriate for criminal law to demand that individuals do only what is reasonable in the circumstances, 78 but that demand can be met by negligence-based offences.
Practicality Versus Fair Labelling
Strict liability offences, then, might be procedurally convenient. The prosecution does not have to prove mens rea with respect to all elements of the actus reus and it also promotes convenience in terms of enforcement whereby D's specific involvement in a crime need not be certain. Strict liability offences may also have a deterrent effect. For instance, for s.3ZB, it is plausible to imagine that if D is aware he is uninsured or unlicensed, then he may not run the risk of driving on the road out of fear of conviction for a serious crime if involved in a fatal accident. Similarly, for s.12A, it is possible to envisage that D may not run the risk of taking a vehicle and driving on the road out of fear of conviction for a serious crime if involved in an accident. Moreover, because s.12A makes everyone in the car, not merely the driver, liable for the aggravated offence, it may deter some joyriders from simply going along for the ride. Therefore, there may be some practical benefits to be gained from strict liability crimes. It might be the case, then, that strict liability is beneficial, especially in minor ‘non-stigmatic’ crimes (those that import little or no element of stigma or condemnation and carry trivial sentences). However, such practical benefits must be weighed against the fair-labelling concerns in more serious ‘stigmatic’ crimes (offences that import and retain an element of stigma or condemnation and impose sentences of imprisonment) such as s.3ZB and s.12A. 79
As identified above, s.3ZB and s.12A breach the fair-labelling principle because they misrepresent the wrongfulness of D's conduct. S.3ZB unfairly imposes liability on D for a stigmatic homicide offence. It therefore makes D criminally responsible as a killer and thus labels D as a culpable killer when D does not culpably kill. Similarly, s.12A falsely imposes liability on D for a serious crime that carries connotations of endangerment. It makes D criminally responsible for causing injury to another and thus labels D as a dangerous person when D does not culpably cause injury to another. These offences therefore result in false censure.
The reason why D ought not to be falsely censured is supplied by the adverse implications it has on D's life following conviction. Within the criminal justice system, the offence D is convicted of will have implications for any future conviction. For instance, D's previous convictions may be admitted during trial as evidence of his bad character, thereby aiding the judge or jury to reach a verdict on D's guilt or innocence; 80 and where D is found (or pleads) guilty, the corresponding sentence can also be influenced by D's previous convictions. 81 It would clearly be unfair to D if his criminal record misrepresented his wrongdoing as it could lead to erroneous judgments about D's bad character in future trials. Where convicted, it could also improperly increase the sentence imposed on D. As such, there is a need for convictions to accurately capture D's wrongdoing. As Chalmers and Leverick state, ‘[i]f decisions are to be made about the offender's fate that rely on previous convictions, it is only fair that the information communicated is accurate and sufficiently informative’. 82
Outside the criminal justice system, criminal convictions also have various harmful social consequences. Communities develop negative judgments of wrongdoers through the official state censure that is manifested in criminal convictions. Such judgments are the result of complex social phenomena caused by the stigma that is attached to different crimes, and society uses that stigma to disparage and discredit the individual in question. 83 As Simester notes, a criminal conviction ‘marks D out in such a way that it becomes appropriate, within the community, for the regard in which he is held to be affected’. 84 As such, certain social and professional exclusions typically follow – notably, by ‘employers or landlords to deny jobs or housing, respectively, to offenders’. 85 Gibson notes that, ‘[w]hilst it is not clear exactly how a specific offence label affects access to jobs or housing, it is probable that the more stigmatic the crime, the more likely that access will be compromised’. 86 But the more stigmatic the crime, the more likely D will also be ostracised by his friends and family, or his wider peers.
S.3ZB is undoubtedly a stigmatic offence. As the Supreme Court notes: [t]his is a statute creating a penal provision, and one of very considerable severity. The offence created is a form of homicide. To label a person a criminal killer of another is of the greatest gravity. The defendant is at risk of imprisonment for a substantial term. Even if, at least in a case of inadvertent lack of insurance or venial lack of licence, a sentence of imprisonment were not to follow, the defendant would be left with a lifelong conviction for homicide which would require disclosure in the multiple situations in which one's history must be volunteered, such as the obtaining of employment, or of insurance of any kind. Nor should the personal burden or the public obloquy be underestimated; to carry the stigma of criminal conviction for killing someone else, perhaps a close relative, perhaps as in the kind of situation referred to in para 19 an innocent child, is no small thing.
87
[t] he first point to be made about section 12A of the Theft Act is that it is in no sense a regulatory or “quasi-criminal” enactment. Aggravated vehicle-taking is a serious crime. Driving offences causing serious injury or damage are a source of growing public concern. The aggravating factors which differentiate the section 12A offence from the basic offence expose the defendant to a maximum sentence of 14 years imprisonment, the same as for causing death by dangerous driving. Although the death of the victim is not strictly speaking an element of the offence, the increased maximum sentence for cases where someone has been killed reflects the real stigma associated with it.
88
Thus, the false censure of D – supplied by both s.3ZB and s.12A – not only unfairly condemns D for the misrepresented wrongfulness of D's conduct, but it also has far-reaching implications for D following conviction: it could inappropriately affect any future conviction and sentence, and unfairly exposes D to harmful social consequences. It is arguable, then, that these fair labelling issues outweigh any practical reasons, such as procedural convenience and deterrence, for the use of strict liability in stigmatic offences such as s.3ZB and s.12A. Consequently, due to fair-labelling concerns, strict liability is not appropriate in these offences.
Stigmatic crimes retain an element of moral condemnation which conveys a public implication of culpable wrongdoing. 89 However, for stigmatic offences, such as s. 3ZB and s.12A, the state does not rely on the proposition that D is culpable for their wrongdoing. So, the state cannot claim to be punishing D in accordance with D's overall desert – culpable wrongdoing. Instead, it is simply punishing D in virtue of the fact that only wrongdoing occurred. Therefore, there are fair labelling concerns for imposing strict liability in stigmatic crimes more broadly.
Remedying the Concerns: Exiling Strict Liability in Stigmatic Offences
As discussed above, the fair-labelling concerns that INUS causation would produce in stigmatic offences such as s.3ZB and s.12A stems from INUS dispensing with a blameworthy requirement in the causal analysis. It follows that a blameworthy element needs to be imported into these types of offences. For s.3ZB and s.12A, specifically, a blameworthy requirement for the aggravated actus reus elements. This would ensure that these offences appropriately capture the wrongdoing that they aim to target (i.e. culpable killing or culpably causing injury). This section discusses how these offences should be transformed if INUS were to be implemented in English criminal law. It claims that the preferred option is to import a blameworthy requirement for the aggravated actus reus elements of s.3ZB and s.12A generally provided by D's careless driving.
The Requirement of Careless Driving
The preferred option is that s.3ZB and s.12A each require blameworthiness as to their aggravated actus reus elements, generally provided by the poor standard of D's driving. But what should that standard of driving be?
Guidance is given from the main statutory driving offences involving serious injury or death. Excluding both s.3ZB and s.12A, such offences include causing serious injury or death by dangerous driving; 90 causing serious injury or death by careless, or inconsiderate, driving; 91 causing serious injury or death while driving disqualified; 92 and causing death by careless driving when under the influence of drink or drugs. 93
Like s.3ZB and s.12A, these offences are constructive crimes since they all require a base offence of driving to an unlawful standard (sometimes in specific circumstances) and in doing so caused death. What is common amongst these offences is that they all require some form of blameworthiness which derives from D's poor standard of driving. For the offences of causing injury or death by dangerous driving, ‘dangerous’ driving is the basis on which blame is ascribed. An individual is regarded as driving dangerously if ‘(a) the way he drives falls far below what would be expected of a competent and careful driver and (b) it would be obvious to a competent and careful driver that driving in that way would be dangerous’. 94 The emphasis on the meaning of dangerous driving is thus assessed objectively and denotes very bad driving. 95
In relation to the offences of causing injury or death by careless, or inconsiderate, driving and causing death by careless driving when under the influence of drink or drugs, ‘careless’ or ‘inconsiderate’ driving is the basis on which blame is ascribed. An individual is regarded as driving carelessly (without due care and attention) if ‘the way he drives falls below what would be expected of a competent and careful driver’.
96
Whereas a person is regarded as driving inconsiderately (without reasonable consideration for others) only if other persons are inconvenienced by his driving.
97
Again, the emphasis on the meaning of careless or inconsiderate driving is assessed objectively.
98
Careless driving is thus a crime of negligence, but the degree of negligence is below that required for dangerous driving. As the Court of Appeal stated: careless driving on its own almost always involves culpability at the lowest possible scale. In one sense, every driver is careless when he makes a mistake. Every driver, even the best, and most experienced, and normally careful, does so from time to time. That does not mean that he has fallen “far below” the appropriate standard.
99
Finally, the offences of causing injury or death while driving disqualified, requires ‘something open to proper criticism in the way in which he or she was driving which contributed more than minimally to the death’. 101 What is needed in terms of the quality of driving to prove this offence remains to be seen. But the Supreme Court noted that the driving does not need to amount to careless or inconsiderate driving, but there must be some conduct that amounts to more than the mere presence of being on the road. 102
As discussed above, the standard that D's driving must be ‘something open to proper criticism’ was required in both Hughes and Taylor regarding s.3ZB and s.12A, respectively. However, given the vagueness of such a test, and the fact that the quality of driving could be less severe than negligence, it is inappropriate as the standard of driving on which blame can be ascribed. Moreover, given the uncertainty surrounding the quality of driving required, it is unclear whether a risk of injury or death is intrinsic to the quality of driving. As discussed above, the intrinsic connection between the initial wrong that D commits and the additional actus reus in constructive crimes is important to D's blameworthiness for that actus reus. 103 It is for this latter reason that the standard of ‘inconsiderate driving’ can also be rejected. There is little, if any, risk of injury or death resulting from being inconvenienced: the risk of injury or death is unrelated and thus extrinsic to this standard of driving. This could explain why there are rare, if any, cases prosecuted as causing injury or death by inconsiderate driving. 104 It is therefore inappropriate to ascribe blame to a driver for the injury or death of another based on the driver's inconsiderate driving. 105
But the extrinsic risk of injury or death is not an issue for both dangerous and careless driving. For dangerous driving there is a high risk of injury or death resulting from very bad driving: the risk of injury or death is related and thus intrinsic to this standard of driving. Indeed, the likelihood of injury or death occurring from dangerous driving forms the very reason why such driving is wrongful and illegal in the first place.
106
Simester et al. agree.
107
They provide an example of a motorist, D, who drives through a village at a speed 20 miles per hour in excess of the speed limit, loses control of his car, mounts the pavement and kills a pedestrian, V. For the authors, D would rightly be liable for causing death by dangerous driving notwithstanding the fact that V's death was a matter of (bad) luck because that luck ‘is not random or unrelated’.
108
They confirm that: [i]t is D's dangerous driving … that brings the uncontrolled factors into play, and makes the luck relevant. Where the luck is intrinsic, there seems no difficulty about holding D responsible and culpable for the outcome.
109
However, as discussed above, the rationale for both s.3ZB and s.12A was to reduce injury or death by deterring individuals being on the road when they had no right to be there. But, compared with careless driving, such deterrence may be reduced if dangerous driving was the required standard of driving for these offences. Because of the higher degree of culpability for dangerous driving, individuals are far more likely to run the risk of driving vehicles on the road while unlicensed or uninsured, or after unlawfully taking them, out of a lesser fear of conviction (because of knowledge that liability will not be imposed unless it is in the unlikelier situation that D is driving dangerously). In essence, because of the higher degree of culpability for dangerous driving, such a standard of driving for s.3ZB and s.12A may result in a reduced deterrent effect of those offences, compared with a standard of careless driving. Therefore, the preferred blameworthy requirement for the aggravated actus reus elements of s.3ZB and s.12A be provided by D's ‘careless’ driving. As a result, the wording in these offences should be amended to reflect this.
Indeed, if s.3ZB and s.12A had the requirement of careless driving, then liability in cases such as Williams, Hughes, McGuffog, Uthayakumar, Marsh and Taylor would not follow from an application of INUS. Although causation would be established in these cases, the defendants were not driving carelessly and thus are not to blame for the injury or deaths of their respective victims.
Conclusion
INUS causation is problematic for stigmatic strict liability result crimes looked at through the lens of offences involving injury or death by driving, specifically s.3ZB Road Traffic Act 1988 and s.12A(2)(b) Theft Act 1968. Unlike the standard tests of causation in English criminal law, the liability outcomes in cases involving these offences, such as Williams, Hughes, McGuffog, Uthayakumar, Marsh and Taylor, would change so that instead of being acquitted, the defendants would be criminally liable for their respective offences. This is because INUS removes the blameworthy requirement which has been injected into the standard tests of causation. However, such liability outcomes yield fair-labelling concerns. This is due to both s.3ZB and s.12A misrepresenting the wrongfulness of D's conduct: s.3ZB represents culpable killing when D does not culpably kill; and s.12A represents culpably injuring another when D does not culpably injure another. That is objectionable of itself since it falsely censures D. But it is especially objectionable because of the negative effects that could follow conviction.
Strict liability is generally beneficial in minor non-stigmatic crimes for practical reasons: strict liability offences might be procedurally convenient, and they may also have a deterrent effect. But in more serious stigmatic crimes, like s.3ZB and s.12A, such practical justifications are outweighed by the need for fair labelling. Therefore, due to fair-labelling concerns, strict liability is not appropriate in these types of offences.
Because of problems in dispensing with blameworthiness under INUS causation in offences such as s.3ZB and s.12A, a solution could be to circumscribe strict liability by importing a blameworthy element into them. For s.3ZB and s.12A, specifically, importing a blameworthy requirement for the aggravated actus reus elements of the offences generally provided by D's careless driving. This would ensure that these offences appropriately capture the wrongdoing that they aim to target (i.e. culpable killing or culpably causing injury).
Footnotes
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
