Punishing drivers who kill: putting road safety first?

This article seeks to argue that the new offence of causing death by careless driving is undesirable and ought not to have been created. From a legal perspective, the offence goes against some fundamental principles of criminal law and, from a practical view, it will fail to rectify previous problems with the pre-existing offences available as criminal charges in cases of road death. Furthermore, it will not satisfy the public’s appetite for a more punitive approach to drivers who kill. It will be argued that the law in this area should not be driven by this latter consideration but that it should be used primarily as a method of increasing road safety. A radical alternative offence structure of general driving offences is proposed with this as its objective.


INTRODUCTION
The Road Safety Act 2006 (RSA 2006) creates, 1 amongst other offences, two new homicide offences of causing death by careless or inconsiderate driving (CDCD) 2 and causing death by unlicensed, disqualified or uninsured driving. 3 Previously two statutory homicide offences relating to driving already existed: causing death by dangerous driving (CDDD) 4 and causing death by careless driving whilst under the influence of drink or drugs. 5 The government claims that the new offences will deter dangerous and careless driving: '[t]he whole purpose of such legislation is to ensure that we take greater care and that we avoid dangerous driving'. 6 This paper, focusing on the offence of CDCD, 7 seeks to argue that such a purpose is unlikely to succeed and to explore the reasons why the offence has been suggested by government, the responses to it and what the alternatives might be. * I am grateful to Mandy Burton, Chris Clarkson, the editors of Legal Studies and the anonymous referees for their helpful suggestions on earlier drafts. 1. The Act received Royal Assent on 8 November 2006. It may, according to the Home Office, be another year before the new offences come into force. 2. Road Safety Act 2006, s 20, inserting s 2B into the Road Traffic Act 1988. The offence is triable either way with a maximum penalty of 5 years' imprisonment when tried on indictment. 3. Road Safety Act 2006, s 21, inserting s 3ZB into the Road Traffic Act 1988. The offence is triable either way with a maximum penalty of 2 years' imprisonment when tried on indictment. 4. Road Traffic Act 1988, s 1. 5. Road Traffic Act 1988, s 3A. In addition, the offence of aggravated vehicle taking under s 12A of the Theft Act 1968 carries a higher maximum sentence (now 14 years) where death is caused. 6. Lord Davies of Oldham Hansard HL Deb, vol 675, col 1563, 22 Nov 2005. The second new 'causing death' offence could be argued to be equally, if not more, unjustified than CDCD. All it requires is that the defendant causes the death of another person by driving and at the time was either driving other than in accordance with a licence, driving This paper starts by looking at the impetus behind the Home Office's programme of reform and assessing whether the concerns the reforms seek to address are valid. It provides some results of an empirical study of criminal charges in cases of fatal road traffic collisions conducted by the author. The first argument to be addressed is that there was a gap between pre-existing offences, which meant that justice was not done, or seen to be done, in many cases. There is evidence that a 'grey area' exists between the two offences of careless driving and causing death by dangerous driving, which has caused some difficulties for the exercise of prosecutorial discretion. However, the main perception of the government is that this gap has prevented justice from being done because of the difference in sentencing powers between the two offences. CDCD is designed to satisfy public demands to narrow that sentencing gap for drivers of different levels of culpability who cause death. It is argued in response to the government that retribution for careless drivers who kill is neither desirable nor likely to be achieved by the new offence. The discussion then turns to the work of criminologists and psychologists whose research will be used to argue that the aim of increased road safety will not be achieved by the reforms. It will be argued that reform of the law may be desirable but a radical alternative is suggested with the aim of greater road safety in mind.

CLOSING THE JUSTICE 'GAP'
The government's 2005 Consultation Paper 8 sets out the issues identified by the review of bad driving offences that might be addressed by any law reform. One of the main issues identified was that in cases involving fatalities the prosecution's decision as to whether the driving was careless or dangerous assumes enormous importance because of the difference in sentencing between careless driving 9 and CDDD. 10 This 'gap' causes distress to bereaved families because if a prosecution is for careless driving only, or is for CDDD but fails, the fact of death receives little attention by the courts. 11 The Home Office also put forward the concern that charge bargaining or undercharging means that cases warranting punishment as CDDD are treated as careless driving. while disqualified or driving while uninsured against third party risks. There is no requirement to assess the standard of the defendant's driving or to establish whether it was careless or dangerous. However, it would seem that this offence is less controversial than that of CDCD. A number of respondents to the Consultation Paper who opposed the offence of CDCD supported this offence because they considered driving whilst disqualified or uninsured to be more culpable than carelessness might be: Home Office A Summary and Next Steps: The Review of Road Traffic Offences Involving Bad Driving (2005), available on the Home Office website at http://www.homeoffice.gov.uk/documents/cons-bad-driving-2005/. 8. Home Office Review of Road Traffic Offences Involving Bad Driving: A Consultation Paper (2005) ch 2. 9. 'Careless driving' is the common term used to refer to the offence of driving without due care and attention contrary to Road Traffic Act 1988, s 3. 10. Paragraph 2.1. The maximum penalty for CDDD is now 14 years' imprisonment (Criminal Justice Act 2003, s 285). The maximum sentence for careless driving is a fine at level five (raised from level four by RSA 2006, s 23) and 3-9 penalty points or disqualification. 11. Death is not recognised in the offence label. Death may be an aggravating factor in sentencing for careless driving ( Simmonds [1999] RTR 257) in imposing a fine and/or disqualification.
To prove CDDD the Crown must show that the defendant (D) drove dangerously and that in doing so he caused the death of the deceased (V). According to the Court of Appeal in Hennigan , 12 the driving need not be a substantial or major cause of death. All that is needed is that the dangerous driving is a cause, and something more than de minimis. Under s 2A of the Road Traffic Act 1988, as amended, a person is to be 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. 13 'Obvious' carries its ordinary meaning within the English language and the Court of Appeal has suggested that judges should not attempt to define it further. 14 'Dangerous' refers to a danger of injury to any person or of serious damage to property. 15 A problem inevitably arises from what is meant by a 'competent and careful' driver. The Association of Chief Police Officers (ACPO) and the Crown Prosecution Service (CPS) have attempted to give guidance to police officers and Crown prosecutors as to the meaning of such terms in their Driving Offences Charging Standard . 16 This lists the following as examples, amongst others, of driving supporting an allegation of dangerous driving: racing or competitive driving; speed which is highly inappropriate for the prevailing road or traffic conditions; disregard of traffic lights and other road signs, which, on an objective analysis, would appear to be deliberate; overtaking which could not have been carried out with safety. 17 In Conteh , 18 the Court of Appeal emphasised that dangerous driving involves a high threshold and that the judge should direct the jury that mere breach of the Highway Code does not itself constitute the offence of dangerous driving or CDDD. 19 In this case a conviction for CDDD was quashed, with the court commenting that if V had not died it suspected that the prosecution would not have considered a charge of dangerous driving.
Under case-law the test for careless driving was whether the defendant failed to meet the standard of 'exercising the degree of care and attention that a reasonable and prudent driver would in the circumstances'. 20 This has been placed on a statutory 12. (1971) (7) states that '[a] failure on the part of a person to observe a provision of the Highway Code shall not itself render that person liable to criminal proceedings of any kind but any such failure may in any proceedings . . . be relied upon by any party to the proceedings as tending to establish or negative any liability which is in question in those proceedings '. 20. Simpson v Peat [1952] 2 QB 24. footing by the RSA 2006, 21 which has inserted s 3ZA into the Road Traffic Act 1988 and defines driving without due care and attention as follows: '(2) A person is to be regarded as driving without due care and attention if (and only if) the way he drives falls below what would be expected of a competent and careful driver.
(3) In determining for the purposes of subsection (2) above what would be expected of a careful and competent driver in a particular case, regard shall be had not only to the circumstances of which he could be expected to be aware but also to any circumstances shown to have been within the knowledge of the accused.' The alternative form of the offence is inconsiderate driving, which now has the following statutory definition under s 3ZA of the Road Traffic Act 1998: '(4) A person is to be regarded as driving without reasonable consideration for other persons only if those persons are inconvenienced by his driving.' It would seem that inconsiderate, rather than careless, driving is very rarely charged, 22 and it may be that the inclusion of this method of committing CDCD is superfluous to requirements.
Both careless driving and dangerous driving, then, are crimes of negligence, and the difference between the two offences is one of degree: whether the driver has fallen below or far below the standard of the competent and careful driver. In the most recent electronic version of the Charging Standard, some examples of bad driving which previously provided evidence of careless driving, such as driving with an arm or leg in plaster or driving when fatigued, are now listed under examples of dangerous driving. The following examples may support an allegation of careless driving, according to the Charging Standard: overtaking on the inside; driving inappropriately close to another vehicle; driving through a red light; emerging from a side road into the path of another vehicle. 23 When faced with a case of bad driving resulting in death in the past, a Crown prosecutor was confronted with a difficult decision as to whether to charge CDDD, recognising the fact that death has been caused and allowing for a maximum penalty of 14 years' imprisonment, or careless driving, the same charge to be brought where a driver has fallen below the standard of the competent and careful driver and not caused death, with no potential for a sentence of imprisonment. Whilst the Home Office highlighted the possible problems of undercharging and charge bargaining in potential cases of CDDD, it noted that 'hard evidence bearing on these issues is not always available or conclusive'. 24

EMPIRICAL EVIDENCE ON BAD DRIVING OFFENCES
Some empirical evidence about the way in which prosecutorial discretion is exercised in such cases is available from a study conducted by the Transport Research 21. Section 30. 22. This form of the offence did not figure in any of the empirical cases discussed below and is very rarely discussed in the relevant literature. 23. Paragraph 5.6(1). 24. Consultation Paper, above n 8, para 2.2. Laboratory (TRL) 25 and from HMCPS Inspectorate's review of decision making in fatal road traffic collisions. 26 A third study has been completed by the author. 27 The TRL report was commissioned by the Department of Transport, Local Government and the Regions (as it was then) and involved a variety of research methods. Of these, it is probably a postal survey of judges, magistrates, police officers, Crown prosecutors and victims' families which is most revealing about the 'gap' between careless and dangerous driving. The TRL summarised these findings to illustrate that there are inconsistencies in the way in which different respondents viewed the blameworthiness of drivers displaying different levels of inattention, 28 and in the way that they assessed the need for dangerous driving to involve a deliberate act, 29 as well as the way in which speed is taken as an indication of dangerous driving. 30 Another finding was that there was 'a strong feeling amongst the police interviewed that the CPS downgrade Dangerous Driving cases to Careless Driving too often'. 31 The view of prosecutors, on the other hand, was that they were not overcautious but charged careless driving rather than dangerous driving where there was insufficient evidence to prove the latter. 32 The HMCPS Inspectorate (HMCPSI) reviewed prosecutorial decision making in cases of fatal road traffic collisions in ten CPS areas, starting in October 2001. 33 This involved interviewing CPS staff and analysing a sample of 164 files, 34 99 of which ended in criminal charges being brought. It was found that in a small number of cases the standard of decision making needed to be improved. 35 These were cases in which there was a choice between charging careless driving and CDDD, with the prosecutor having opted for the former charge in six cases where HMCPSI was of the view that the latter would have been more appropriate. HMCPSI's explanation for this was that in some cases the law had been misapplied or that the prosecutor had failed to appreciate that juries now have a greater propensity to convict for CDDD than was once believed to be the case. 36 HMCPSI considered that the prosecutors in these cases 'did not properly consider and weigh the various factors that fell to be considered on an objective, reasonable basis'. 37 The two examples of these cases provided by HMCPSI were one involving a case of speeding 38 and one of fatigue. 39 The results of the author's own research confirm that cases involving speeding and fatigue are indeed particularly problematic. This study had broadly two aims. The first was quantitative in nature, in discovering how often different charges are brought in cases of road death incidents (RDIs). 40 In cases where there are multiple options with regard to offences to be charged, for example if there is a choice between CDDD and manslaughter or CDDD and careless driving, which option is chosen? The second aim of the research involved a qualitative analysis in seeking to find explanations for the choices made by decision makers within the criminal justice system. What factors may influence the decision as to charge and what difficulties, if any, do decision makers encounter in reaching that decision? Ultimately, the objective was to discover how the law operated in practice and whether it did so in the way its designers intended.
The study was conducted through the analysis of secondary data, in the form of police and CPS reports relating to RDIs. Police and CPS files were accessed in the counties of Leicestershire, Lincolnshire and Northamptonshire. All cases of RDIs from these three counties occurring in the years 1999 and 2000 were selected for analysis, 41 with data collection being conducted in 2001-2003. A total of 316 files were analysed, of which 126 cases resulted in prosecutions for the relevant offences. 42 In addition to this file analysis, a small number of open-ended, semi-structured interviews were conducted with representatives of each police force and CPS area involved.
In analysing the decision making of the police and CPS, it became clear that, depending on the seriousness of the risk-taking involved in any one case, the decision maker (usually a CPS lawyer) 43 would have to make a choice between two courses of action. At the lower end of seriousness, the decision would be whether to take no further action against any surviving driver, or whether to charge careless driving where the fault of that driver was minor but sufficient to satisfy the test of falling below the required standard. Increasing up the scale of seriousness, if it was clear that this test for careless driving was met, the decision might become a choice between a charge of careless driving or CDDD.
In the most serious of cases a decision maker might be faced with the question of whether a charge of manslaughter ought to be preferred to one of CDDD, although 37. Ibid, para 5.75. 38. Ibid, para 5.65. 39. Ibid, para 5.66. 40. The term 'road death incident' is preferred to 'fatal road traffic collision' because it is a term used by some of the police forces involved in the study and better represents the approach taken by the police to their investigation of such cases. Officers are encouraged not to make assumptions about how a death has come about and to investigate the possibility of murder or manslaughter having been committed. 41. With the exception of cases where there was no surviving driver who could be accused of committing an offence. This included cases of single vehicle collisions and collisions involving two or more vehicles where all the drivers suffered fatal injuries. 42. The study was concerned with offences which contribute to the occurrence of a fatal collision and relate to the way in which a vehicle is driven. Other driving offences such as driving whilst disqualified, driving without a driving licence, driving without insurance and using a vehicle without an MOT certificate were excluded from the study. the Charging Standard advises that a manslaughter charge will very rarely be appropriate because of the existence of the statutory offence of CDDD. 44 Manslaughter in such a case would take the form of gross negligence manslaughter, applying the test in Adomako, 45 unless there was evidence of the vehicle being used as a weapon of assault, in which case manslaughter would take the form of constructive (unlawful act) manslaughter. Murder could be charged in any case in which there was evidence that the driver intended to kill or cause grievous bodily harm.
Thus, there exists a hierarchy of offences applicable in such cases, and where in the hierarchy a particular case lies depends on whether it has crossed the relevant line between offence categories: no further action or careless driving; careless driving or CDDD; CDDD or manslaughter. 46 The line between careless driving and CDDD is particularly important because it determines whether the maximum penalty will be one of a fine or 14 years' imprisonment.
Whether these lines have been crossed will depend on an interpretation of the evidence available, as well as an interpretation of what exactly the definition of an offence requires, assisted by the Charging Standard. The aim of the Charging Standard is to produce more consistent decisions in cases of motoring offences, particularly those causing serious injury or death. This may be of some help, but it is clear that prosecutors have difficulty in deciding on which side of the various lines the relevant driving fell. The focus in this paper is on the divide between careless and dangerous driving, which the Consultation Paper identified as being one of the difficulties with the current law.
The study found that immediately on either side of this line there exists a 'grey area' between careless driving and dangerous driving which is notably evident and problematic in cases involving particular features of excess speed and fatigue. If a person drives at a speed which is highly inappropriate for the prevailing conditions this may support a charge of dangerous driving, 47 and thus CDDD where a fatality results. There were 21 cases in which excess speed for the conditions was a factor leading to prosecution for careless driving, and eight where the charge was CDDD. In most cases, the collision investigator collecting evidence at the scene of a collision on behalf of the police will be able to establish, from marks left at the scene and the damage caused to the vehicles, the speed at which the vehicles were travelling on impact.
Two cases provide an illustration of the difficulties faced by prosecutors in making decisions about whether to charge CDDD or careless driving in cases involving excess speed. Both cases involved a car striking a pedestrian in a 30 mph speed limit whilst travelling at approximately 15 mph above the limit. In the first case, the defendant was driving a taxi in darkness without lights on his dashboard, being unable to see the speedometer. The collision investigator estimated that the car was travelling at 44 mph. In this case, careless driving was charged, although there were notes on the file describing the requirements for CDDD and careless driving, suggesting that the prosecutor was having difficulty deciding which offence to charge. 44. Paragraph 11.5. 45. [1995] 1 AC 171. The test requires that the defendant owed the victim a duty of care which the defendant breached, causing the victim's death. It is then for the jury to decide whether, having regard to all the circumstances, the breach should be characterised as gross negligence and therefore as a crime. 46. Where the additional factors of drink or drugs are involved, an alternative charge of causing death by careless driving when under the influence of drink or drugs might be preferred. 47. Charging Standard, para 7.7.
The second case took place in a different county, where the collision investigator estimated the speed upon impact to be 43 mph. The police, referring to the Charging Standard, recommended that CDDD be charged and the Crown Prosecutor agreed. However, careless driving was also charged because the prosecutor was concerned that the judge might find no case to answer on the CDDD charge due to a lack of eyewitness evidence.
These cases, amongst others, clearly show the grey area between the two offences where excess speed is concerned. With the exception of this last case, most of the cases resulting in prosecutions for CDDD which were based on excessive speed involved not only speeds which were in excess of the speed limit, but some other factor which contributed to the collision. This would suggest that despite the Charging Standard listing speed that is highly inappropriate for the conditions as evidence supporting a charge of dangerous driving, it is unlikely that a charge of CDDD will be preferred unless additional factors contribute to the dangerousness of the driving. 48 The other factor present in many of the cases which fell within the grey area between careless and dangerous driving was fatigue. There were 14 such cases in the sample, seven of which led to a prosecution for CDDD and seven for careless driving. The difficulty with this type of case is that it is clear that D was not paying sufficient attention to driving, but it is almost impossible to establish the degree of inattention. The Charging Standard in use at the time of the research listed 'fatigue/nodding off' as conduct which would support a charge of careless driving, 49 although the new version of the Charging Standard available online now lists driving when too tired to stay awake as evidence of dangerous driving. 50 More generally, however, the old version of the Charging Standard stated that acts caused by more than momentary inattention will normally lead to a charge of careless driving, 51 whilst failure to pay proper attention, amounting to something significantly more than a momentary lapse, will support an allegation of dangerous driving 52 (and thus CDDD). The distinction, then, between the degrees of inattention comes with the insertion of the word 'significantly', which is required for dangerous driving. This suggests that a competent and careful driver will occasionally suffer moments of inattention, but no more than that. The longer the duration of the inattention, the further below the standard of the competent and careful driver D will have fallen. With no witnesses to account for D's behaviour behind the wheel in the moments leading up to a collision, such a distinction is extremely difficult to judge or prove. 53 48. This is confirmed by a Consultation Paper issued by the CPS in relation to its charging policy in cases of bad driving following the enactment of the RSA 2006. Citing the case of

Director of Public Prosecutions v Milton [2006] EWHC 242 (Admin), [2006] RTR 21 the CPS
suggests that the current legal position is that 'while excessive speed alone cannot be a basis for convicting a driver of dangerous driving, where the speed was grossly excessive on a stretch of road with potential hazards then excessive speed could amount to dangerous driving': Crown Prosecution Service Prosecuting Bad Driving -A Consultation on CPS Prosecuting Policy and Practice (2006) para 148, available at http://www.cps.gov.uk/news/pbd.pdf. It was further suggested (para 154) that speeding is one area where public attitudes may have changed so that what was once seen as careless may now be considered dangerous. 49. Charging Standard, para 5.6. 50. This change was made following criticism of the Charging Standard by HMCPSI in its report, above n 26, para 5.66. 51. Charging Standard, para 5.7. 52. Ibid, para 7.7. 53. The 2004 version of the Charging Standard has omitted all reference to inattention in its guidance on careless and dangerous driving.
Eleven cases involved heavy goods vehicle (HGV) drivers failing to notice that the traffic ahead of them had become stationary or very slow moving, often due to queues at junctions. CDDD was charged in seven of these cases, whilst careless driving was charged in the remaining four. Some of these cases were clear examples of dangerous driving, whilst others caused real difficulties for the CPS in their decision making. In one case, the judge, in sentencing D for three counts of CDDD to which D had pleaded guilty, took into account the special nature of the driving responsibilities of HGV drivers as aggravating factors. Because of the size and weight of HGVs, they pose a greater risk to other road users than do cars. HGV drivers are subject to legislation limiting the number of hours they can drive without resting in order to try to reduce the risk of such drivers becoming fatigued and failing to exercise the degree of attention necessary when driving such dangerous vehicles. They are also subject to lower speed limits 54 in order to limit the danger they pose to other road users. There were examples in the sample of what can occur if both of these regulations are not complied with.
One such example is a case in which an HGV collided with a car that had broken down in the inside lane of a dual-carriageway. It was found that D had been travelling at 56 mph, the limit for the vehicle on that road being 50 mph. It was also found that D had failed to complete his tachograph charts and had taken insufficient weekly rest. The collision investigator was able to calculate that the car would have been visible to D for between 21 and 26 seconds prior to impact, and yet D only braked immediately prior to or upon impact. In interview, D said that the car was 10-15 feet away when he first noticed it, although he claimed that he considered himself to be alert at the time and was unaware that he had breached driving regulations. The original decision as to charge taken by a Senior Crown Prosecutor, having considered the evidence supporting charges of CDDD and careless driving, was to charge careless driving. However, the case was reviewed by the Chief Crown Prosecutor who altered the charge to one of CDDD. 55 There were two other cases in the same county with similar facts, one of which led to a charge of careless driving and the other to a charge of CDDD. 56 Cases in the other two counties correspondingly showed, through notes on the file, difficulties in choosing a charge in cases with similar facts.
The driver of an HGV who falls asleep poses a particularly grave danger due to the size of his vehicle. However, car drivers who fall asleep or are fatigued also create risks. There were three cases in which D was a car driver who had clearly suffered 54. See Road Traffic Regulations Act 1984, ss 81, 86, 88 and 89. Goods vehicles exceeding 7.5 tonnes maximum laden weight are limited to 40 mph on single carriageways, 50 mph on dual carriageways and 60 mph on motorways. 55. The prosecution for CDDD was stayed for abuse of process on the basis that the Chief Crown Prosecutor had altered the charge after being prompted to do so by V's relatives. For further discussion see Cunningham 'The Unique Nature of Prosecutions in Cases of Fatal Road Traffic Collisions', above n 27. The case reverted to the magistrates' court and D was convicted of careless driving. 56. The former case was decided before it became local policy for the Chief Crown Prosecutor to review all RDI cases. Although the latter case resulted in a unanimous verdict of CDDD, this conviction was quashed on appeal on the basis that the judge had failed to give the jury a good character direction. from fatigue, 57 and all three resulted in prosecutions for careless driving. What these cases seem to suggest is that HGV drivers who suffer from fatigue may be more likely to face charges of CDDD than car drivers, mainly for evidential reasons. From examining a tachograph chart the police can establish not only the speed at which a HGV driver was driving, but also the pattern of his driving leading up to the RDI and can determine at what point D reacted to the presence of a hazard prior to the collision. When dealing with car drivers, the police and CPS have no such evidence and must rely on D's account of his driving, any eye-witness evidence and the presence of any tyre marks on the road surface.
It appears that the courts have changed their view of the seriousness of driving while fatigued in recent years. In Boswell, 58 the Court of Appeal considered 'dozing at the wheel' to be a mitigating factor, but by the time of Attorney General's Reference No 26 of 1999 59 it had become recognised that such drivers are to blame for continuing to drive when the appropriate action would be to stop and take a rest. The Selby rail crash in February 2001 brought the issue of sleep-related collisions to the attention of the public, with much media attention surrounding the trial and sentencing of Gary Hart. 60 The sentencing guidelines on CDDD adopted by the Court of Appeal in Cooksley and Others 61 list 'driving when knowingly deprived of adequate sleep or rest' as an aggravating feature in sentencing.
Prosecutors were well aware that their decision was enormously important in establishing the maximum penalty open to the court on conviction, as well as determining whether the death would be reflected in the charge. The government's solution was to introduce an offence that would not only provide sentencers with the power to impose imprisonment on careless drivers who kill, but would also reflect in the charge the fact that death resulted. As such it could be argued that it will fill the gap and reduce the importance of this 'grey area' because both CDDD and CDCD are serious offences punishable by imprisonment. However, whilst there may be a few cases which fall within the bottom end of dangerous driving but are wrongly charged as careless driving, it is not clear that creating a new, more punitive offence is justified or will succeed in achieving any of the objectives set for it by the government. In this respect the motivations behind the reform merit fuller critical examination.

SATISFYING PUBLIC DEMAND
The main reason for the creation of CDCD appears to be that pressure has been placed on the government by road safety groups, often representing those unfortunate enough to have had first-hand experience of RDIs through the loss of a relative. They 57. In other cases where it was concluded that the deceased was at fault, there was a suspicion that the deceased had fallen asleep at the wheel. Where a car driver causes a collision due to his fatigue, he is less likely to survive than an HGV driver who does the same, which perhaps explains why there were fewer cases in which a car driver was prosecuted because of such driving. 58. (1984)  understandably feel the need to have their loss recognised by the criminal justice system and feel wronged when drivers involved in RDIs are merely convicted of careless driving. A new offence of CDCD might achieve retribution. However, retribution implies that those on whom a criminal penalty is imposed deserve to be punished to that degree. Retribution can be seen as a theory of punishment based on moral blameworthiness, according to which offenders should be punished proportionately to their offence. The difficulty is in determining what factors ought to influence an offender's moral blameworthiness. More precisely, there is disagreement over whether the result of an offender's actions affects his moral blameworthiness, or whether his moral blameworthiness depends solely on his state of mind. This question is central to the issue of whether CDCD is justified in punishing more severely a careless driver who has caused death than one who has not.
It is clear that killing someone by driving carelessly is more harmful than merely creating a risk of killing by driving carelessly. But this does not explain why a driver who kills deserves more punishment than one who does not. This is because whether or not a careless driver kills depends on chance. When Lord Hailsham, 62 and later Sir Brian MacKenna, 63 described CDDD as 'illogical' it was because of the central role that luck plays in the offence. Such an argument is even more forceful when applied to CDCD, since the blameworthiness required for the offence is of a lesser degree. As noted by Willett in relation to non-fatal traffic offences, 'it is often a matter of chance whether the damage done is great or not. It is … frequently the skill of other drivers or the alertness of the police that prevents more serious consequences'. 64 Improvements in medical care and technology also mean that collisions which in the past could have been fatal no longer lead to death if the victim is fortunate enough to receive the necessary care. 65 There is a substantial literature on the subject of how luck should influence criminal liability. However, most of this is discussed in the context of advertent risk taking. Exceptions to this include Ashworth, who argues that in punishing bad drivers: '[the] primary criterion should be the risk created, taking account of the magnitude of the harm risked and the probability of its occurrence. If the harm did occur this may provide evidence which assists in assessing magnitude and probability, but resulting harm does not alter the intrinsic seriousness of the risktaking '. 66 Similarly, Schulhofer argues that to allow fate to affect the moral quality of an act in retributive terms is perverse and that the culpability of an offender guilty of negligence must stem from his having taken a forbidden risk, and any result of his risk taking should not affect his penalty. 67 On the other hand, there are those who argue that any harm caused by a defendant who has broken the law should be attributed to him in allocating responsibility. Horder argues that the law is, in some circumstances, justified in departing from the correspondence principle. The correspondence principle suggests that in order to impose criminal liability, an offence should be drafted in such a way that the mens rea or blameworthiness element relates to the proscribed harm. With homicide offences, the correspondence principle would insist that death was intended or foreseen before liability could be incurred. Horder argues that defendants who direct their efforts towards harming someone should be liable for the harm caused, even where that harm is greater than the harm intended or foreseen, because they 'deserve' their bad luck. 68 Their luck is not 'pure'. It is unclear whether Horder would also apply this argument to CDCD, since this is not a crime which depends on D directing his efforts towards harming someone. His view might be that although the death in CDCD is 'pure' (bad) luck there is still the possibility that D can be held responsible for it, since the purity of the luck is tainted by the foreseeable outcome of death. He suggests that this provides justification for punishment of manslaughter by gross negligence. 69 However, this argument does not necessarily extend to CDCD. In cases of manslaughter by gross negligence the test for liability requires that the jury have regard to the risk of death involved in D's actions. 70 For careless driving alone to be proven there is no requirement that death be foreseeable. The definition of careless driving in the RSA 2006 requires only that D's driving fell below the standard of the competent and careful driver; there is no requirement to have consideration to the risk of even injury or damage to property as is required in relation to dangerous driving. 71 Whilst it may be undisputable that any case of CDCD must have involved the taking of a risk where death was foreseeable, since death did in fact materialise and it is arguably 'foreseeable' that death could result from any instance of bad driving, the 'moral distance' between death and careless driving is too great to be able to attribute the former to the latter. This is because the risk of death or injury created by the bad driving is not a requisite ingredient of the offence of careless driving on which the resultant crime of CDCD is constructed.

Lord Hailsham
Horder's main justification for departing from the correspondence principle is that where D engages in violent conduct, luck legitimately plays a role in extending D's liability to cover harm of the same form but more serious than that which was foreseen or foreseeable. In engaging in violence the defendant changes his normative position towards the victim and, in doing so, accepts the risks involved. In relation to violent offences (an assault which forms the basis of constructive manslaughter, for example), luck goes the last mile in a journey of violence upon which D has embarked. However, in relation to CDCD, luck has to make the whole journey on its own. There is no mens rea requirement in relation to either death or physical harm falling short of death, and so there is no similar starting point from which liability can be extended.
Some might argue, though, that drivers who drive carelessly should suffer the consequences of their actions because they too have altered their normative position by engaging in criminal activity. They, just like violent offenders, 'make their own luck'. However, two objections to this argument present themselves. First, is it possible to say that someone changes their normative position through driving carelessly? Most offences of careless driving involve a failure to live up to a particular standard, perhaps through inattention or from being distracted, and do not involve a conscious decision to commit an offence or take a risk. It does not make sense to say that someone who had a lapse in concentration has changed his or her normative position towards other road users in the same way that someone who chooses to inflict a blow in angry retaliation has changed his or her normative position towards the victim.
Secondly, when looking at fatal collisions resulting from careless driving, there are often several different contributions, many reliant on luck, which combine to determine the outcome. The further beneath the standard of a competent and careful driver D falls, the higher the likelihood of creating a risk, and vice versa. If D drives in a grossly negligent manner he or she will create a fairly high risk of death. Any death can arguably, therefore, be attributed to D. But where D drives just below the standard of the competent and careful driver, the resulting death may be dependent on a whole host of additional factors, over which D has no control.
An example from the author's study, discussed above, provides an example. D1 was waiting at a junction to turn onto a main road. She saw a car travelling from her right which indicated that it would be turning left into the side road in which D1 was stationary. D1 therefore judged that it was safe to turn onto the main road. However, D2 was driving on the main road behind the car that was turning into the side road, and at that point attempted to overtake the turning car, in contravention of the solid white line in the centre of the road which prohibited such a manoeuvre on the approach to the junction. D1 and D2 collided, killing D2's passenger (his wife).
In this case both D1 and D2 were to blame, and consideration was given to prosecuting both for careless driving. The police decided, however, not to prosecute D2 on the basis that he had lost his wife and it was not in the public interest to prosecute him. Given that D2's driving was judged to have been worse than that of D1, it was also felt that it would be unjust to prosecute D1 in the absence of a prosecution of D2. Let us consider what might have happened and what would influence the commission of an offence of CDCD in a similar case. Had D2 himself died in the collision would it have been fair to have prosecuted D1 for CDCD, given that D2 was himself responsible? As it was, V died as a result of a combination of D1 and D2's driving, and that outcome may have been influenced by the fact that she was elderly and frail (where a younger person may have survived the injuries), and the particular way in which the two cars impacted. Further factors could play a part, such as whether V was wearing her seatbelt, whether her car was fitted with airbags and whether an air ambulance was available to ensure that she reached hospital within the 'golden hour'.
Alternatively, had D1 pulled out in front of the driver of a car who was an advanced driver and so able to take evasive action, should her liability differ? A fatal collision might be avoided but D1's blameworthiness remains the same. Why should she, in this last scenario, benefit from her good luck and avoid a harsh penalty for the risk she has taken?
Honoré agrees that '[a]ny principle which can justify responsibility for bad luck must be fair. If it is to be fair, it must entail that when we bear the risk of bad luck we also benefit if our luck is good. Allocation according to luck must cut both ways'. 72 However, he argues that the law can be seen to be fair because it uses the lottery of 72. T Honoré 'Responsibility and Luck ' (1988) 104 LQR 530 at 539. life to allocate responsibility, and over the course of a lifetime we stand to win more than we lose. It is unclear why Honoré should submit that we are likely to win more often than we lose. To explain the allocation of responsibility on the basis of a lottery, as does Honoré, could be interpreted as implicitly encouraging risk taking. In any case, there is certainly no guarantee that if D1 in the above example were to be punished for causing the death of V her luck would eventually even out during the remainder of her lifetime.
The argument for different levels of liability depending on the occurrence or otherwise of harm seems to be based on nothing more than intuition. A logical approach is preferable to an intuitive one and, as argued by Ashworth, the law 'should censure people for wrongs, not misfortunes'. 73 It can be argued that the wrong in the offence of CDCD is driving carelessly and that the misfortune is to have caused death.
Thus, the objection to CDCD is one of principle. Neither careless driving nor dangerous driving requires subjective mens rea. They are both crimes of negligence with the latter requiring a higher degree of negligence than the former. Elsewhere in the law of homicide negligence is usually only punished when it is determined to be 'gross' as required by the test for manslaughter in Adomako. Criminal liability does not attach to those who cause death negligently in other ways. 74 Whilst the Home Office received many positive responses to its Consultation Paper, the legal profession, with the exception of the CPS, was more or less unanimous in its opposition to the proposals. 75 This opposition was further represented by various members of the House of Lords, in particular Lord Lyell of Markyate, former Attorney-General. He stated that 'to create an offence of causing death by careless driving is fundamentally wrong as a matter of justice', 76 since justice must be governed by the culpability of an offender and not the consequence of his actions. Lord Monson noted that whilst the ostensible reason for the proposal was to increase road safety, as discussed below, the largely unspoken reason was to give the public and tabloid press what they have asked for. 77 Arguably, this provision of the RSA 2006 is just the latest of many initiatives delivering the government's declared priority to rebalance the criminal justice system in favour of victims, 78 which is, in itself, an indication of the trend away from an elitist tradition and towards populism in criminal justice policy making. 79 Whilst there are historical, political and cultural explanations 80 for the Labour Government's involvement of 'ordinary people' within its policy-making process, Johnstone notes that the methods used in gauging public feeling are 'likely to produce punitive and illiberal responses typical of our immediate reactions to offensive or threatening behaviour. Labour's populist style is likely to produce a vengeful, emotive response'. 81 The offence of CDCD is a clear example of this emotive response from those on the receiving end of the results of careless driving.
Whilst there may be positive reasons to engage the public in law-making policy, if this is to be done it should be facilitated through informed, rationalised thinking. Arguably, one of the main reasons why lawyers were opposed to a new offence of CDCD is because they engage directly with a range of real cases and issues rather than receiving distorted images of the law from the media 82 and can see the arguments from all points of view. Given that lawyers constitute a small section of the electorate, however, it is not surprising that their views were ignored since, as noted by Ashworth, '[v]ote winning and populism appear to hold greater sway with the Government than respect for rights'. 83 Practical considerations may have been more likely to persuade the government that the new offence is misguided. Lord Monson's main point was that the new offence is unlikely to give the public and tabloid press what they have asked for because CDCD provides a sentence of a maximum of 5 years' imprisonment. 84 The government argued that this is needed to punish those who fall just short of the test for dangerous driving, but the danger is that in a large number of cases the courts will decline to impose custodial sentences in cases of CDCD because the standard of driving is at the lower end of culpability. This will, Lord Monson predicts, lead to outrage in the press and nothing will have been achieved. If, on the other hand, the courts do impose custodial sentences on a frequent basis this could be problematic in terms of increasing the prison population. 85 Other possible expenses incurred by the criminal justice system include a greater number of contested cases, since drivers will have a great deal more to lose in admitting that they drove carelessly.

INCREASING ROAD SAFETY
In its 2005 Consultation Paper the Home Office noted that a framework of law exists to support standards of good driving and promote road safety including, on the one hand, specific offences enabling basic standards to be enforced, such as speeding and using a mobile 'phone whilst driving and, on the other hand, the two general offences of careless and dangerous driving. 86 These two general offences can be described as endangerment offences, since they exist to punish risk taking and do not require the risk to have materialised before liability can be established. Fifty years ago a constructive result crime (CDDD) was created from the more serious of the two offences, and allows for those whose dangerous driving has caused death to be punished more severely. The new offence is in some respects consistent in doing the same in relation 82. Which Johnstone identifies as being an important cause of punitive attitudes amongst the public: ibid, at 168. 83. A Ashworth 'Criminal Justice Reform: Principles, Human Rights and Public Protection' [2004] Crim LR 516 at 517. 84. It should also be noted that the offence is triable either way, meaning that, in many cases, this maximum penalty of 5 years' imprisonment will not be available as they will be heard at the magistrates' court. 85. The government estimates that an additional 150 prison places a year would be needed to accommodate the numbers who might be caught by the new offences of CDCD and causing death by uninsured, unlicensed or disqualified driving: to the other general offence of careless driving. However, on closer inspection, the offence is highly illogical and irrational.
Williams stated that the reason for punishing negligence (of which careless driving is an example) is the utilitarian one, that we hope thereby to improve people's standards of behaviour (in this case their driving). 87 This was recognised at a Law Society conference in 1964 at which it was said that 'the vast mass of traffic offences were temporary lapses of ordinary men and women and the aim therefore was not so much retribution as making them more careful drivers'. 88 Punishment can be designed to improve behaviour in a number of ways. First, the threat of punishment may deter all drivers from driving badly (general deterrence). Secondly, having been convicted of an offence and suffered the consequences, a driver could be deterred from re-offending in the future (specific or individual deterrence). Some have argued that deterrence fails to operate in relation to driving offences, since if a driver is not deterred from taking a risk by the immediate threat of physical harm to himself or damage to his property, he will not be deterred by any other consequences which could ultimately result. 89 Williams argued that although this is true, driving offences can add to the pressure to make bad drivers change their habits. 90 As far as individual deterrence is concerned, most drivers convicted of CDCD will be deterred from re-offending, not so much by the penalty they receive, but by the very fact that they have caused the death of another human being. 91 Those who regularly drive carelessly under the illusion that 'accidents' only happen to other people and eventually become involved in a fatal collision will experience, as the result of that collision, what the outcome of risk taking may be. Compare this to drivers who drive carelessly but, fortunately, for whatever reason, manage to avoid having a collision. Some such drivers may be deterred from taking further risks because the 'close-call' has shaken them up. However, many drivers will have their behaviour reinforced every time they take risks without incurring harmful consequences. The law needs to focus on providing general deterrence against careless driving to prevent this last group of drivers from continuing to drive carelessly; in terms of individual deterrence it is unlikely that any penalty will have such an effective influence on an individual's driving as will the personal guilt they feel having caused death.
Whilst it is widely thought that general deterrence is an achievable goal of sentencing, this is dependent upon certain stipulated preconditions being met. 92 One of these conditions is that the potential offender must believe that there is a nonnegligible likelihood of being caught. 93 In relation to CDCD, it is highly likely that someone who drives carelessly and causes a fatal collision will be caught. However, (Stevens, 2nd edn, 1983)  in this situation getting caught is reliant on causing death. It is unknown what the success rate in apprehending careless drivers is. Although careless driving is an endangerment offence and does not require that any harm has been caused, in practice it may be that prosecutions are rare unless either the police have themselves witnessed an incident of bad driving or there has been a collision as a result of such driving. If the offender does not consider the possibility of causing death by driving carelessly he or she will probably believe that the risk of being caught is negligible.

G Williams Textbook of Criminal Law
Thus, in order to be deterred from committing the offence of CDCD, a driver must first assess the risk of causing death. If he or she discounts the risk of causing death, he or she will also discount the risk of being punished for doing so. For drivers to be deterred from driving carelessly they must believe that the threat of unpleasant consequences (whether they be physical harm or criminal punishment) is strong enough to persuade them not to take the risk. As noted in the Consultation Paper: '[t]he apparently strong correlation between prosecution for these [general] offences and involvement in a crash or collision implies that few drivers will be induced to drive more safely merely by weighing up the likelihood of being charged and convicted of the general offences, since they do not expect to crash in the first place.' 94 If this is true, then surely such an argument is even stronger in relation to CDCD, and the fact that an additional, higher penalty attaches to this new offence will be ineffective in deterring careless drivers. Thus, the second main objection to CDCD is that, although it carries a harsher penalty than the general offence, it will not improve safety standards on the roads because drivers will not identify their own driving with driving that is likely to cause death. This is confirmed by research conducted by criminologists and psychologists.
Corbett and Simon found that the main reason for the commission of traffic offences was that offenders weigh up considerations of safety in the immediate circumstances and decide for themselves whether or not to engage in an unlawful manoeuvre, rather than letting the law decide. 95 The motivation for committing offences differed slightly between low-level offenders and high-level offenders. Lowlevel offenders are more likely to say that their offending was inadvertent, whilst high-level offenders gave reasons of convenience or laziness to explain their offending. 96 Those who abide by the rules of the road do so because to break the rules is wrong and might lead to an accident, whilst those who choose to break the rules do so in the belief that they are in control and discount the risk of having an accident. Corbett, Simon and O'Connell have confirmed this in a follow-up study of high-speed drivers. 97 They found that as long as a speeding driver feels in control he feels that his accident risk is no higher than if he were driving within the limit. 98 Corbett and Simon's finding 99 that drivers tend to have their own personal code of driving standards quite separate from the rules imposed on them by statute and the 94. Paragraph 2.19. Highway Code casts doubt on the possible effectiveness of the current test for the offences of dangerous and careless driving, which require an offender's driving to be measured against that of the 'competent and careful' or 'ordinary and prudent' driver. If the ordinary driver drives according to his own code which is individual to him, can such a test succeed? This was certainly brought into question by the TRL study, which suggested that although the current test is an objective one, applying it in accordance with the Charging Standard involves subjective judgement on the part of those making the decision, since respondents did not agree on the level of inattention indicated by different activities carried out whilst driving. 100 Whether carrying out a certain activity amounts to falling below the standard of a competent and careful driver depends on the personal opinion and experiences of the police officer, prosecutor, magistrate or juror applying the test. Manstead et al found that those who commit traffic violations overestimated the number of other drivers who also committed such violations, whilst those that did not commit such violations underestimated how many others did commit such violations. 101 If the competent and careful driver is interpreted to mean the ordinary or average driver, this raises problems, since the average driver may not be particularly careful.

C Corbett and F Simon
Another potential hindrance to the successful application of the current test is that most drivers view themselves as being of above-average capability. 102 Reason et al report that those who believe themselves to be better drivers commit more violations than those who self-appraise more modestly. 103 They conclude that: 'Drivers who violate may see themselves as skilful enough to take risks (or, perhaps, to behave in ways which would be risky only for "less skilful" drivers). Alternatively, drivers who violate may come to think of themselves as good drivers because they get away with what they do.' 104 So if drivers have a tendency to view themselves as expert drivers, whose skilfulness is above that of the 'competent and careful' or average driver, they are likely to commit the offence of dangerous or careless driving without acknowledging that they do so, because they do not see the offence as applicable to themselves. This ties in with Corbett's findings that drivers develop their own code of practice outside the law. These offenders may be difficult to deter by means of general deterrence because they set themselves apart from the rest of the motoring public. Although they might recognise that most drivers should abide by the rules of the road in order to limit the number of collisions caused, they see themselves as justified in creating risks in their driving because their greater skill will ensure that they are able to avoid collisions. 105

105.
Corbett et al's study found that high-speed offenders who had in the past been involved in collisions blamed the other party involved. Their responses were often contradictory and they failed to recognise that stopping times are reduced at higher speeds, no matter how much attention drivers pay to their driving in compensation for driving at higher speeds. Corbett, Simon and O'Connell, above n 97.
A possible goal of law reform could be to change the mindset of such offenders in the same way as has been achieved in relation to drink-drivers. Writing in 1985, Riley argued that those who were not deterred from drink-driving needed educating on the effects of alcohol in order to improve their compliance with the law. 106 He found that a large proportion of those who chose to drink and drive did so because they were ignorant about how much alcohol would take them over the legal limit. This was aggravated by drinkers deciding for themselves whether they felt capable of driving or not, unaware that alcohol in fact promotes feelings of self-confidence and therefore ability to drive. Twenty years later, however, it is commonly recognised that drink-driving is not only a criminal offence, but that it is totally unacceptable behaviour. 107 The question is: what can we learn about attitudes to drink-driving that could be used to encourage attitude change in relation to other driving behaviour?
When drink-driving was first introduced as an offence it was a common occurrence that drivers would drive after drinking and it was not seen as 'wrong'. Bottoms uses drink-driving as an example of the law being capable of moulding morality. 108 Whilst most crimes involve acts which are considered to be impermissible by society, some prohibitions 'may on occasion not reflect positive morality at all, but rather may be imposed by those in power in the hope of securing obedience through deterrent calculation'. 109 He continues that sometimes 'the fact of the prohibition, and citizens' evolving response to it, can influence the development of a new strand of positive morality. Something very like this seems to have occurred in relation to drinking and driving '. 110 Bottoms sees normative compliance as pivotal in producing legally compliant behaviour, and the suggestion here is that normative compliance is probably the most effective way of reducing the prevalence of careless driving, alongside educational initiatives. Bottoms identifies three ways of achieving normative compliance, the first of which is through acceptance of or belief in a social norm. 111 The examples he gives are the norms against assaulting others or taking their property without their consent. In relation to traffic offences, it can be argued that the relevant norm is the norm 106. D Riley 'Drinking Drivers: The Limits to Deterrence (1985) 24 (4)

111.
The second is 'attachment leading to compliance'. This involves the offender forming a relationship to a particular person, group or institution which holds non-criminal values. This is perhaps something over which the authorities have no control, the difficulty being that young drivers perceive less pressure from their peers to avoid committing driving violations than do older drivers: D Parker et al 'Determinants of Intention to Commit Driving Violations' (1992) 24(2) Accident Analysis and Prevention 117. The third is 'legitimacy'. This relies on the potential offender recognising the power of the authority imposing the prohibition. This might be one of the indicators of the reason behind non-compliance with traffic offences. Although the offences are created through statute, as for other criminal offences, the way they are enforced may differ throughout the country: Dix and Layzell, above n 99. In particular, in relation to speed limits, it is the local authority that determines what the speed limit on a particular stretch of road should be. against endangering the lives of others. 112 Although such a norm is probably accepted by members of the public, the problem is that they do not necessarily associate what they see to be fairly minor traffic offences with the endangerment of others. As identified by Riley, 113 education of the public was necessary before many drivers accepted that drink-driving was indeed dangerous, and similar education is needed in relation to careless driving. The current offences of dangerous and careless driving do little to communicate exactly how a life has been endangered or what behaviour is prohibited, which suggests that these offences require reform.

AN ALTERNATIVE OFFENCE STRUCTURE
The above arguments are that the new offence of CDCD is undesirable, since it will lead to injustice and will not achieve the aims set for it. However, given the work of criminologists and psychologists showing that the test of the competent and careful driver is unlikely to succeed in practice because of the way drivers are motivated to abide by the law or to break it, and given that there exists a 'grey area' between careless driving and dangerous driving, it might be worth considering whether the general offences could be improved. The aim should be to amend the offences to encourage safety on the roads and discourage risk taking. Whilst on the surface there appears to be no gap between the two offences, given that they require the application of the tests of falling far below the standard of the competent and careful driver, or alternatively falling below that standard, it is inevitable that such a test will be difficult to apply if there is disagreement as to what is required of a competent and careful driver in any given situation.
The TRL report argued that in practice the current law operates by distinguishing between those who deliberately take risks and those whose risk taking is not deliberate. 114 This is not an actual requirement of the law, which should be applied in a completely objective manner. 115 However, it may be true that, to a certain extent, we do judge those who take deliberate risks differently from those who do not. 116 Additionally, it can be argued that deliberate risk taking should be dealt with separately from inadvertent risk taking because of how such a differentiation might increase the law's ability to deter risk taking. This is supported by psychological research that has distinguished between violations, which can be defined as the deliberate infringement of some regulated or socially accepted code of behaviour, and errors, which include unintended infringements. 117 According to Parker et al:

112.
It is important that others' lives are endangered. Drivers may feel entitled as a question of autonomy to take risks if it is only their own life they see as being endangered. 113. Above n 106. 114. See above n 25. 115. Whilst this is acknowledged by the CPS in its recent Consultation Paper on charging policy, it identifies the question of whether it is appropriate to consider the driver's intentions in deciding whether or not a manoeuvre was dangerous as one that could be an appropriate one to ask: CPS, above n 48, paras142-146. 116. Hawkins found that deliberation was a factor in favour of prosecuting in cases of breaches of health and safety legislation, and that even though such offences are usually of strict liability, evidence of deliberate breaches are taken more seriously: K Hawkins Law as Last Resort (Oxford University Press, 2002)  '[e]rrors and violations differ both in their psychological mechanisms and in the kinds of remedial actions necessary to combat them. Errors arise as the result of information-processing problems; violations have a large motivational component. Errors may be understood in relation to the cognitive function of the individual. Violations, however, are a social phenomenon and can only be understood in a broader organisational or societal context. Errors can be minimised by retraining, redesign of the human-machine interface, memory aids, better information and the like. Violations should probably be dealt with by attempting to change attitudes, beliefs and norms, and by improving overall safety culture.' 118 Both violations and errors are potentially dangerous (in the sense of creating a risk of harm to others), but what distinguishes them is the psychological processes which lead to their commission. Examples of violations include speeding, lane indiscipline and close following ('tailgating'). Examples of errors include failing to notice pedestrians crossing and not checking the rear-view mirror before executing a manoeuvre. Errors can be distinguished from lapses, which involve absent-mindedness and are not dangerous, such as leaving a roundabout at the wrong exit. 119 This grading could be used as the basis of a new structure of bad-driving offences. Errors could replace the current offence of 'careless driving', with extended use of the National Drivers Improvement Scheme (NDIS) to deal with offenders. 120 Violations would roughly cover what is currently 'dangerous driving' and would lead to higher penalties than both errors and the current offence of dangerous driving. This would be justified by the subjective fault requirement needed to prove the offence and by the need to change attitudes through individual, general and educative deterrence. It should also reduce the size of the 'grey area' between the two levels of offences, although it is difficult to predict whether it would eliminate it completely. Those cases involving speed which currently fall within the grey area should be resolved as cases of violations, ensuring that such offenders do not escape with a charge for the lesser offence. With careful drafting of the rules on which violations could be based, it might also be possible to include driving when fatigued within the definition of a violation.
In analysing a sample of 84 defendants in cases of CDDD, dangerous driving and careless driving, the TRL used Reason et al's concept of behaviour to categorise the data in comparing the treatment of different types of bad driving. 121 However, the 118. Parker et al 'Driving Errors, Driving Violations and Accident Involvement', above n 117, at 1036. 119. Ibid, at 1037. 120. The NDIS has been used as an alternative to prosecution in cases of careless driving not involving serious injury where the offender accepts the offer. This involves both classroom discussions on the causes of collisions and practical driving exercises, the aim of which is to highlight and eliminate any dangerous or inappropriate elements of driving style: C Burgess and P Webley 'Evaluating the Effectiveness of the United Kingdom's National Driver Improvement Scheme', unpublished article available at http://www.ex.ac.uk/~cnwburge/pages/ ndis03.html. ACPO has produced guidance notes on the use of NDIS, available at http:// www.acpo.police.uk/asp/policies/Data/nat_driver_improve_scheme_v4_april05_09x05x05.doc. Sections 34 and 35 of the RSA 2006 now allow for a reduction in punishment in the form of penalty points and periods of disqualification following an attendance on an approved course. 121. Pearce et al, above n 25, p 22. categories were not used as intended by the originators, 122 in that the TRL decided to include within the 'lapses' category lapses in concentration which resulted in danger to others. For example, included within this sub-category was a case in which a driver used a mobile phone and a case of a lorry driver who had exceeded his work hours and fallen asleep. 123 Both of these would be better categorised as violations. In addition to the three categories of lapse, error and violation, a fourth category of aggressiveness was added by the TRL as a more serious category than violations. According to the TRL's analysis all cases involving aggressiveness were charged as dangerous driving or CDDD and all but one of the 24 cases classed as violations were charged in this way. Twenty of the 24 cases classed as errors were charged as careless driving. 124 This suggests that a new offence structure along these lines could closely follow decisions under the current law, but it is argued that the simpler distinction between the offences would lead to decisions according to blameworthiness being made more easily.
This suggestion is not without problems. One predictable criticism of the proposal is that it will re-create the problems experienced in relation to the offence of reckless driving. One reason why the offence of reckless driving was replaced with one of dangerous driving 125 was that reckless driving was difficult to prove because of the need to show that the defendant had either failed to consider the risk or had deliberately run a foreseen risk. 126 Since violations, by their definition, require an element of intentionality, this again creates problems for the prosecution in proving the defendant's state of mind. However, it can be argued that proving intent in this instance is no more difficult than proving mens rea in relation to other criminal offences. In giving his judgment in the case of G and Another, Lord Bingham of Cornhill pointed out that in contested cases where mens rea is denied, the jury is able to infer mens rea 'from all the circumstances and probabilities and evidence of what the defendant did and said at the time'. 127 Defendants in cases involving fatal collisions, when interviewed, are often quite open about what they were thinking in the moments preceding the collision. 128 In only a minority of cases would a subjective test cause problems where the defendant refused to answer questions in interview, and in such cases it may be possible to infer what their state of mind was from the surrounding circumstances. 129 The Law Commission has recognised that, in some cases, the task of the jury in finding intention or recklessness to cause harm is easier 122. Ibid. 123. Ibid, p 47. 124. Ibid, p 46. 125. The Road Traffic Act 1988 was amended by the Road Traffic Act 1991, which implemented proposals made by the North Report: Department for Transport and Home Office Road Traffic Law Review (HMSO, 1988). 126. The test for recklessness which applied was that from the case of Lawrence [1982] AC 150. This test requires that (1) the defendant does an act which creates an obvious risk of causing physical injury to some other person or of doing substantial damage to property and (2) when he does the act he either has not given any thought to the possibility of there being any such risk or has recognised that there is some risk involved and has nonetheless gone on to do it. 127. [2003[ ] UKHL 50, [2004 1 AC 1034 at [39].

Cunningham Criminal Charges Brought in Cases of Road Death Incidents in the East
Midlands: Implications for Law Reform, above n 27. 129. It is of course also possible for a jury to draw adverse inferences from such silence under the Criminal Justice and Public Order Act 1994. than applying the current test for dangerousness of falling far below a certain standard. 130 Furthermore, what distinguishes this suggested offence from previous motoring offences is that the intention which must be proved relates not to the creation of a risk, but to the violation of a traffic rule. Spencer put forward a similar suggestion to the North Committee at the time of the Road Traffic Law Review in 1988. 131 He was unhappy with the Review's suggestion for a new offence of 'very bad driving' (which later became 'dangerous driving') because he predicted that the courts would have difficulty in defining it and it was too vague. 132 He noted that: ' [e]xperience amply shows us that it is unworkable to have a vague offence of bad driving, with a mens rea requirement that forces the prosecution to prove that the driver realised his behaviour actually created a risk for other road users. But if the offence were defined at a low level of abstraction, so that it included driving in any one of a number of ways which are generally recognised as being unacceptably dangerous, a mens rea requirement would surely not be unworkable at all.' 133 It would be difficult to prove that the defendant foresaw a risk of death or injury to others as a result of his or her bad driving, but this is not what would be required under the proposal I advance here. All that would be required is that the defendant deliberately did the act which amounted to the contravention of a rule. 134 It is true that such a proposal is likely to be unattractive to the families of those killed on the roads, since they will continue to feel that their bereavement is ignored by the criminal justice system. However, such a problem may be insurmountable given principles of criminal law. What this offence construction could conceivably help achieve, however, is the aim of greater road safety. Those who commit violations would hopefully be deterred from doing so, whilst those who commit errors, which cannot be deterred by their very nature, 135 would become safer drivers as a result of further education and training.

CONCLUSION
The argument of this paper is that the government was ill-advised to introduce the offence of CDCD. The causing of death on the roads is an unfortunate fact of modern 134. Spencer's suggestion was that an offence be 'drafted in general terms, but incorporating a partial definition expressly putting a number of recurrent cases within the general clause for the avoidance of future doubt: racing another vehicle, overtaking on a blind corner, driving when seriously under the influence of drink or drugs, driving a vehicle with seriously defective steering or brakes, etc': ibid. The suggestion here is that in order to avoid a statutory provision for an exhaustive list of rules which could form the basis of the new offence, it would be simpler for it to refer separately to the rules in the Highway Code, which could be updated. 135. This is a point made by Merry and McCall Smith, who use the work of the psychologist Reason to suggest that medical practitioners will unavoidably commit errors during their career and that the law should focus on preventing violations instead: life and, whilst we should do all that we can in terms of engineering, education and law enforcement to reduce the dangers, we should look carefully at the ability of substantive law to deter risk taking through the creation of punitive result crimes. The loss of life is the worst harm that can be experienced by humans, both in terms of the direct victim and those loved-ones that he or she leaves behind. It is only natural that the bereaved seek to allocate blame and explain their loss, but the danger of punishing for death those who have driven carelessly is that the only good it will do is to 'give fleeting satisfaction -and it would only be fleeting -to some relatives and provide another sacrificial lamb on the altar of a tabloid newspaper'. 136 Even the provision of fleeting satisfaction is doubtful, given that it is likely that only in the rarest and most serious of cases (perhaps those in which the correct charge would be CDDD rather than CDCD) will the courts feel justified in imposing a custodial sentence of anything near the maximum of 5 years. 137 It may be that, in a few cases, those who warrant punishment for CDDD avoid conviction because of the difficulty in applying the test of falling far below the standard of the competent and careful driver, and the fact that disagreement over what this means results in a 'grey area' between the offences of dangerous and careless driving. The government is convinced that the objective tests required for the general offences are workable, and so has failed to consider if these offences could better be defined. It may be that, as crimes of negligence, the current law is 'as good as it gets', but given the findings of criminologists and psychologists regarding the motivation of traffic offenders, there is at least some doubt over whether they can achieve the aim of increasing road safety. This area of law deserves further and more careful consideration, with the possibility of introducing a new offence structure based on the distinction between what psychologists term 'violations' and 'errors'.