The Sentencing Academy Response to Consultation: Assault Offences Guideline
14 September 2020
The Sentencing Academy welcomes this opportunity to provide input into the Council's consideration of the revised guideline for assault offences. 1 Our comments pertain only to revised guidelines for some of the offences.
The Academy welcomes the revision of the Sentencing Council’s Assault Guideline and the Sentencing Guidelines Council’s Attempted Murder Guideline. The Academy regards the continual monitoring and revision of guidelines to be a core part of the Council’s role and the Academy would encourage the Council to undertake further monitoring and analysis of its guidelines, and publish the findings of that work.
1. Overarching issues and the context of the guidelines
The Academy disagrees with the Council’s statement that there is ‘general confusion on how to interpret and apply the Step 1 factors of ‘injury which is serious in the context of the offence’ and ‘injury which is less serious in the context of the offence’, across all the assault offences’ (p. 8). This is a matter that the Court of Appeal (Criminal Division) has dealt with on numerous occasions, most notably in the decision of R. v Smith (Grant Christopher) [2015] EWCA Crim 1482; [2016] 1 Cr. App. R. (S.) 8 in which the Court gave what we regard as clear guidance on this issue.
It does not follow from the fact that the point is taken in court, arising from a dispute between prosecution and defence as to the severity of the injury, that there is confusion. On the contrary, we suggest it is to be expected that this is a point on which the court is frequently required to adjudicate by virtue of it being so central to the offence and thus so influential on the resultant sentence.
If one analyses the case law, it can be seen that the Court of Appeal (Criminal Division) (and one suspects the Crown Court too) apply the guidance in Smith with little difficulty. We would therefore not encourage a change to this approach as courts and practitioners are used to this approach to the assessment of offence seriousness and the law is relatively settled on this point.
Regarding the point in relation to differing interpretations of ‘sustained or repeated assault on the same victim’ and ‘significant degree of premeditation’, our view is that the flexibility of the guideline allowing for different interpretations is a benefit of the guidelines. Of course this should not allow for inconsistent (and thus unfair) applications of the guideline, but it does not follow that because interpretations differ, one (or both) are wrong. We agree that more could be done to provide clarity on the ‘sustained or repeated assault’ factor, though of course it is important to note the Council’s general approach to such factors is to avoid a mechanistic or overly prescriptive definition that would unduly constrain a sentencing judge.
​
2. Common Assault
Structure: For all offences except common assault, assaults on emergency workers, and assault while resisting arrest the revised guideline abandons the previous guidelines structure in favour of the two-dimensional matrix structure of later guidelines (such as Robbery). We support this move towards harmonising the structure of Council guidelines, and believe the later structure represents an improvement on the original assault guideline. However, there seems no reason for an exception for the two offences noted. In particular we note that the two offences adopt the same model as the other offences in the guideline (save for attempted murder), namely that of result crimes, and therefore all offences share similar features. Perhaps culpability has been reduced because the harm is limited (maximum of six months’ imprisonment). If seriousness is restricted, why is this dimension not reduced to two levels? Council offers no explanation or justification for this departure from the structure used for all other assaults, and we cannot conceive of one.
Recommendation: Restore 9-cell matrix for these offences.
Finally, the consultation document notes ‘the first step of the guidelines is to consider the culpability level’ (p. 13). Why? The other guidelines do not begin with the determination of the offender's level of culpability. High Culpability Factors ‘Prolonged Assault’: the aggravation springs from either the prolonged duration of the assault, and/or its repetitive character. An offender could kick the victim five times within a few seconds. This would not constitute a prolonged assault. This factor should include repetition: ‘prolonged or repeated assault’.
‘Strangulation’ has been added as a factor. We oppose the insertion of specific conduct as a factor, as it can lead to anomalies. If the offender places his hands momentarily around the victim’s throat this assault may, or may not, be more serious than a powerful punch in the face. Let the court decide on the level of culpability, or harm. An alternative approach would be to employ a culpability factor to describe the use of an item or a tactic to inflict a greater degree of harm, e.g. a shod foot, a headbutt, an elbow, or strangulation. Each of these would fall within this general description without the need to be prescriptive (and thus suggest that other behaviour not included is deliberately omitted by the Council and is thus regarded as less serious).
​
Low Culpability Factors
‘Lack of premeditation’ has been removed. We agree. If the absence of premeditation mitigates, and premeditation aggravates, wherein lies the base offence?.
‘Excessive self-defence’ has been moved to Step 2. We disagree. Many assaults arise as a result of an excessive response to mild or moderate provocation or even assault, and this may be a compelling claim for diminished culpability. When the assault arises out of an excessive, criminal response to provocation, the assault carries an element of ‘but for’ of the provocation. This is an important reduced culpability factor which should be located at Step 1. We note that analyses of the Council's own Crown Court Sentencing Survey data show this to be a very significant factor, much more predictive of seriousness than ‘subordinate role’ although that factor remains at Step 1.2 So theoretically and empirically there appears little reason to consign excessive self-defence to Step 2, where its influence will be greatly constrained.
​
Recommendation: Restore ‘excessive self-defence’ to Step 1.
​
Aggravating Factors
First, ‘spitting/ coughing’ is a particularly unpleasant form of assault; it is not an aggravating way of committing the offence. All other aggravating factors are enhancements to an act of assault. Spitting may well cause more harm and distress than, say, a slap in the face. Or it may not. For example, if the offender spits on a clothed limb of the victim. The offender’s level of culpability should be left to the court to determine. As a general rule, the form of the offence should not be construed as an aggravation. The effects of spitting can be placed within the guideline, such as exposure to the transmission of disease, which can encompass spitting and coughing without the need to specify the nature of the behaviour.
Second, the factor ‘presence of children’ should be more general: it should include the presence of others, particularly children or relatives of the victim. If the victim is assaulted in front of his or her child, why is this always and a priori more aggravating than being assaulted in the presence of his or her elderly mother? Many assaults are intentionally committed in the presence of friends of the victim, and this is a clear source of aggravation. Additionally, it is in our view correct to recognise the increased harm caused by the distress of others who witness an attack of a loved one.
​
​
Recommendation: Remove spitting/ coughing as a factor or consign it to Step 2; expand presence of children factor to include other people.
​
Mitigating Factors
Council has quietly removed two factors, ‘single blow’ and ‘isolated incident’ from the guideline. We agree with removing the first factor: the base offence should be a single blow, with the aggravated factor of repeated or prolonged blows. The fact that the assault constituted a single blow should not mitigate. We are ambivalent about removing the second factor. The fact that the incident was ‘isolated’ supports an out of character, one-off offence and is a legitimate mitigating factor.
Summary: Taken together, we believe the new guideline will result in a severity uplift, although Council's intention was ‘not to change sentences’ 3 for this offence. This may be cause for concern since this offence accounts for almost 40,000 convictions, easily the most common of the assault range of offences. We note that Council feels unable to predict the impact of this guideline, and we encourage Council to continue to ‘explore ways of obtaining the necessary data’.
Finally, several of the recommendations for this guideline read across to all the others. For example, we recommend presence of children be expanded and ‘isolated incident’ restored even for the more serious forms of assault, such as section 18 offences.
3. Assaults on Emergency Workers
Here our concerns relate to the starting point sentences and sentence ranges, specifically their relationship to common assault. As the Council recognises, this offence is an aggravated form of assault, the higher culpability or harm arising from the occupation of the victim. Parliament has legislated for a higher maximum penalty for this form of assault (12 months rather than six months) and this may soon be doubled to two years.
In our view, the starting points and sentence ranges are disproportionate. Returning to the common assault guideline, an assault falling in 1A attracts a starting point of a high-level community order, far below the maximum of six months imprisonment. If the victim is an emergency worker, the starting point jumps to eight months, well above the common assault guideline and above the midpoint of the guideline’s sentence range. We do not believe the occupational status of the victim justifies such a jump in severity. As a general observation, category starting points are usually set just below the midpoint of the category range. In this case that would result in an starting point of 4-5 months. This convention should apply here.
In addition, the guideline range itself is problematic: it spans the statutory range, an anomaly in Council's guidelines. We encourage Council to adopt a general policy on the relationship between the statutory range and the guideline range. There are two views. The guideline can span the entire statutory sentence range, or it may stop short, on the rationale that a guideline provides guidance only for cases that fall within a limited range (perhaps the central two-thirds of cases in the distribution). Cases attracting the maximum penalty or a sentence very near the maximum are clearly exceptional in some way and can be dealt with outside the guideline.
Sentencing statistics are instructive here. In the period after Parliament had raised the maximum sentence, fewer than one in five offenders (17%) received an immediate prison term and 10% a suspended sentence order for an assault against an emergency worker. These rates are very similar to the imprisonment rates for assault (14% immediate custody; 12% suspended sentence order).4 Yet custody was within the category range in five of the six categories in the emergency worker assault guideline. In short, the starting points and sentence ranges for emergency worker assaults may result in disproportionate sentences in relation to common assault as well as current practice for both offences.
Recommendation: Reduce the starting points and sentencing ranges.
Summary: We are apprehensive this guideline may increase sentence severity for this offence which is another one for which Council has been unable to estimate the impact on sentencing outcomes. Unlike common assault, Council intends to raise sentence levels for this offence.5 Our concern is with respect to the degree of this uplift, which may be disproportionate.
4. Attempted murder
The draft guideline for attempted murder seeks to do two apparently incompatible things. The consultation document states that this guideline is being ‘revised and updated at the same time as the Assault guidelines to ensure relativity of sentences given that it represents the most serious non-fatal assault offence’ (p. 47) and yet it also notes that ‘there have been concerns that some sentences in the existing guideline for attempted murder are too low, and are in some cases very much lower than a same facts murder offence’ (p. 52). Since the very significant increases in starting point lengths introduced by Schedule 21, sentences for murder have been de-coupled to a large degree from those imposed for the most serious non-fatal violent offences. To attempt to ensure proportionality with both murder sentences and sentences for serious nonfatal violent offences is to seek to square a circle. By increasing the starting point for some attempted murder offences, this draft guideline may indeed close the gap with sentences for murder; however, in doing so it opens up a larger gap with the offences below. Parliament has always made it clear that it is murder that is ‘unique’ and that is the justification for its different sentencing regime (a mandatory life sentence and statutory starting points). It may, therefore, be preferable to ensure that attempted murder is sentenced in a manner that is proportionate to all related non-murder offences rather than to murder itself – particularly as attempted murder offences lack the key element of murder (i.e. a death).
​
To illustrate this issue, take the starting point for a Culpability A, Harm Category 1 section 18 offence (12 years) and compare this with the same category starting point for attempted murder (35 years). Whilst an intention to kill is clearly more serious than an intention to cause serious injury, it is difficult to see that it merits a starting point almost three times the length in cases where the harm caused is identical. There seems to be an excessive focus on intention over that on harm (in practice this is likely to be a particular problem in the most serious harm cases where, for example, both offenders have left the victim in a permanent vegetative state, as is possible at the top end of these offences; under this guideline the top of the range for one is 16 years, the starting point for the other is 35 years). One could argue that the starting point for the section 18 offence is too low in order to achieve proportionality with attempted murder (which is the next ‘rung’ up on the ladder) but it is difficult to see how this level of starting point gap ensures ‘relativity of sentences’ as is expressly desired by the draft guideline.
The disparity between the starting points for section 18 offences and the comparable category for attempted murder may place excessive pressure on the charging decision (for example, a Culpability A, Harm Category 3 section 18 offence – where the victim suffers merely ‘really serious harm’ – has a starting point of five years, one-fifth of the starting point for a Culpability A, Harm Category 3 attempted murder case where there could be no harm whatsoever yet has a starting point of 25 years). On the other hand, the uplift in the attempted murder guideline does appear to close the gap with murder – although we are not aware of considerable concern at the ‘under-sentencing’ of attempted murder.6
Also on the theme of a possible under-consideration of the impact of harm, the starting point for a Category 3 Harm offence is noticeably closer to Category 1 Harm offences for the higher culpability offences than was the case with the previous guideline, in which the starting point was half the length of the highest harm offences (e.g. 30 years and 15 years in the existing guideline compared to 35 years and 25 years in the draft guideline). This is obviously not an issue that arises in murder cases (where the harm is fixed) and that makes Schedule 21 considerations far more straightforward. Greater consideration might be given to differentiating between a case where ‘Injury results in physical or psychological harm resulting in lifelong dependency on third party care or medical treatment’ and a case in which perhaps the victim suffers no injury whatsoever. The starting point gap between 35 years and 25 years seems inadequate to fully account for the differing levels of gravity between these offences. Furthermore, the Culpability A, Harm Category 3 starting point (25 years) is greater than the top of the range for unlawful act manslaughter and for manslaughter by reason of loss of control (and also slightly higher than the highest starting point for manslaughter by reason of diminished responsibility) and this gives rise to concerns that the intention in a ‘no harm’ attempted murder case outweighs the harm (i.e. a death) in some of the highest culpability manslaughter cases.
At the lower end of the draft guideline, an offence in which the offender genuinely considers it to be an act of mercy appears to be an uncomfortable bedfellow with the other factors in the same culpability category – particularly an offence of attempted murder linked to ‘excessive self-defence’. There is a danger that the guideline could lead to a longer sentence for an attempted murder ‘act of mercy’ offence than would be imposed for a murder offence. We are aware of a minimum term as low as three years being imposed for murder in what might be considered to be ‘mercy killing’ circumstances and this is lower than the bottom of the range for a Culpability D, Harm Category 1 offence. This raises a question about the suitability of the range for this unique very narrow band of cases.
Premeditation Factor The guideline inserts premeditation as a factor associated with offenders of medium culpability, the second lowest. We recommend a different approach to this key factor. Premeditation is one of the most powerful aggravating factors at sentencing. Schedule 21 recognises three levels: if the court finds a ‘substantial degree’ of ‘premeditation or planning’ for each of two or more murders this triggers a whole life order starting point. If the court finds a ‘significant degree of planning or premeditation’, this constitutes an aggravating factor applicable to all cases. Finally, the absence of premeditation is mitigating factor. For sentencing murder, premeditation aggravates while the absence of planning mitigates. This makes little sense. As a general proposition, the absence of an aggravating factor should not constitute a source of mitigation. 7 The ‘base’ condition of an action is surely a spontaneous decision. Culpability then increases to reflect the duration and degree of planning. The greater the planning, the higher the level of blameworthiness. A spontaneous killer is less culpable than the one who premeditates the offence, but this should not move the former into a zone of mitigation. It simply means he or she is not subject to additional punishment in recognition of the aggravating effect of planning. Offenders may thus be arrayed on a dimension of premeditation, with spontaneous cases at one pole and those killing after considerable planning at the other end of the dimension. Premeditation is therefore actually a dimension, and increases in severity should correlate with movement along the scale towards the end of maximal preparation.
We therefore make the following recommendation regarding premeditation in the attempted murder guideline. If there were three categories, we would assign ‘significant planning’ to the highest category; ‘evidence of planning’ to the second tier; and ‘spontaneous assault’ to the lowest level of culpability.
​
​
Offence Committed in Presence of Others
We previously advocated inclusion of this factor in the guidelines for lesser offences of assault (see above). Recent years have witnessed a number of cases of murder (and attempted murder) in which the offence took place in the presence of other people, often the victim's friends or family.8 Witnessing a murder will be a highly traumatic experience for members of the public who happen to be at the scene; for the victim’s relatives, the experience will be lastingly traumatic. On harm alone, this circumstance elevates the offence significantly, but an offender choosing to commit or attempt murder in front of others is demonstrating a wilful indifference to others which bespeaks a higher level of moral blameworthiness. The courts have recognised the aggravating effect of this circumstance even if it is not incorporated into Schedule 21.9 This factor should therefore be included in the attempted murder guideline.
​
‘Merciful Intention’ Attempted Murder
A small number of cases of attempted murder involve a defendant who attempts to kill in the genuine belief that he or she is sparing the victim suffering for which there is no relief. These cases are radically different in culpability than the vast majority of attempted murder. The number of such cases is small. Incorporating them into the guideline creates more problems than it solves. For example, the ‘merciful intention’ mitigation then may be offset by aggravating factors such as premeditation. In our view these cases are sufficiently distinct to be omitted from the guideline, leaving judges free to craft a proportionate sentence by applying the offence seriousness and any other relevant guidelines. We recommend these cases be removed from the guideline. In the alternate, a new lowest category of culpability could be added.
Role of Good Character
There is also an anomaly relevant to sentencing a crime as serious as attempted murder, where the offender intends to kill. ‘Good character and/or exemplary conduct’ appears as a mitigating factor. The guidelines for several sexual offences10 carry the limitation that: ‘In the context of this offence, previous good character/exemplary conduct should not normally be given any significant weight and will not normally justify a reduction in what would otherwise be the appropriate sentence’ and all sexual offences guidelines include this statement: ‘The more serious the offence, the less the weight which should normally be attributed to this factor’. 11
​
The anomalous result is that an offender convicted of intending and attempting to kill (with a maximum of life imprisonment) benefits from good character which is denied to an offender convicted of a sexual offence with a maximum penalty of seven years.
Recommendation: Re-consider the proposed new starting points and ranges with a view to retaining (or even enhancing) proportionality with section 18 offences and manslaughter. We recommend retaining the three tier culpability model from the existing guideline with the determining factors either being the extent of premeditation or simply a replication of the existing approach of the Sentencing Guidelines Council guideline.
Summary: Whilst we welcome the revision of the attempted murder guideline to bring it into the same format of other Council guidelines, there does not seem to us to be a compelling reason to increase the starting points and ranges in order to retain proportionality with Schedule 21. It is important to note that the existing guideline itself was crafted with Schedule 21 in mind. The greater danger, perhaps, is the further loss of proportionality with those offences immediately below attempted murder. If the starting point for the highest category was reduced back to 30 years, offences in which the offender has brought a knife or other weapon to the scene could be incorporated into this higher category and this opens up the possibility of returning to only three culpability categories rather than the proposed four. As this draft guideline appears to privilege culpability over harm, the previous guideline which simply separates non-higher starting point offences into ‘planned’ and ‘spontaneous’ would be a consistent approach to which to return. The factors currently in the lowest culpability category could become mitigating factors to be considered at a later step. Offences related to acts of mercy fall so far outside all other offences covered by an attempted murder guideline that they may well appropriately fall outside the proposed ranges; however, they are sufficiently rare to not endanger the overall integrity of this guideline.
​
Professor Julian Roberts
Dr Jonathan Bild
Dr Lyndon Harris
​
1 We are grateful to Professor Barry Mitchell and Professor Tom O’Malley for their comments on our draft response.
​
2 Pina-Sanchez et al. (2018) ‘Mind the step: A more insightful and robust analysis of the sentencing process in England and Wales under the new sentencing guidelines’, Criminology & Criminal Justice 1–34, Appendix IV, p. 29.
​
​
4 Sentencing Council (2020) https://www.sentencingcouncil.org.uk/publications/item/assault-offencesstatistical-bulletin/. 5 https://www.sentencingcouncil.org.uk/wp-content/uploads/Assault-Offences-draft-resource-assessment.pdf, p.10.
​
6 The sentence imposed on Emile Cilliers for the attempted murder of his wife – life imprisonment with a minimum term of 18 years – suggests that sentencing judges have ample scope under the existing guideline to impose a sentence that is proportionate to a murder offence.
​
7 Previous convictions would appear to be the exception to this rule: an absence of previous convictions mitigates while increasing numbers of previous convictions aggravates. The exception is explained by the fact that the mitigation arising from the first offender status is conceptually different from the aggravation triggered by accumulating additional convictions; the two categories (first offender; repeat offender) does not in fact derive from a single dimension, hence the exception. Imagine applying the Schedule's logic to some other aggravating factors, either derived from Schedule 21 or more generally. Paragraph 10(c) makes the infliction of mental or physical suffering on the victim before death an aggravating factor. The offender who kills the victim without inflicting such suffering cannot claim mitigation as a result.
​
8 R. v Whittle [2019] EWCA Crim 1282; R. v Saunders, Sentencing Remarks 28 February 2017. Judiciary of England and Wales.
9 For example, R. v Whittle [2019] EWCA Crim 1282; and R. v Saunders, sentencing remarks 28 February 2017. Judiciary of England and Wales and more recently, R. v Pencille, https://www.judiciary.uk/wpcontent/uploads/2019/07/pencille-darren-SENTENCING-REMARKS-for-circulation-12-7-2019.pdf para 8.
10 For example, section 4 of the Sexual Offences Act 2003 - Causing a person to engage in sexual activity without consent, where this does not involve penetration. The highest starting point for this offence is 4 years, compared to 35 years for attempted murder.
​