Skip to content

Produced by the Sentencing Academy

Sentencing for Murder

A Review of Policy and Practice

Richard Martin
Assistant Professor, LSE Law School

Published: 9 Sep 2025 Last updated: Apr 2024


Executive summary

  • The only sentence a court can pass for murder is life imprisonment. The mandatory life sentence is comprised of three elements: the minimum term spent in prison (handed down by the sentencing court); the post-term (determined by the Parole Board and its assessment of the offender’s dangerousness); release on licence into the community, subject to conditions and the possibility to recall the offender back to prison.
  • The challenge facing the sentencing judge in murder cases derives in large part from the range of harm and culpability the offence of murder encompasses in England and Wales. Murder can be committed where the offender intends, or foresees as virtually certain, not only the victim’s death, but also really serious harm and then death in fact results.
  • Schedule 21 to the Sentencing Act 2020 governs the sentencing of murder. It was originally introduced in the Criminal Justice Act 2003. By creating a series of starting points and providing lists of aggravating and mitigating factors, Schedule 21 recognises that despite the mandatory nature of the life sentence, an identical level of seriousness cannot be attributed to each murder. In a series of cases, the Court of Appeal of England and Wales has interpreted and applied Schedule 21 to ensure sentencing judges achieve a just sentence in accordance with the guidance provided for by Parliament in Schedule 21.
  • According to the Sentencing Council, the introduction of the higher starting points in Schedule 21 has meant sentences for the vast majority of murder cases increased substantially. The average minimum term has risen from 13 years in 2000 to 21 years in 2021 – an increase of over 60%. This increase in minimum terms has also inflated sentence lengths for other serious offences, most notably manslaughter.
  • Schedule 21 has been the subject of a series of amendments over the last two decades driven by successive Governments, notably the expansion of cases within the higher-level starting points and the creation of a new starting point (25 years). Recent amendments to Schedule 21 have been announced by the Government and are contained in the Sentencing Bill, Criminal Justice Bill and the Sentencing Act 2020 (Amendment of Schedule 21) Regulations 2023. The first of these seeks to expand the types of cases to which the Whole Life Order will apply, while the latter two are in response to the Domestic Homicide Sentencing Review, which made recommendations to better reflect the causes, characteristics and harms of fatal domestic abuse. 
  • There have been numerous critiques of Schedule 21. A longstanding one has been the very justification for the mandatory, rather than discretionary, life sentence for murder in the first place. As for the sentencing framework in Schedule 21 specifically, criticism has focused on the disproportionality between Schedule 21’s starting points and the offender’s harm and culpability, as well as the ordinal disproportionality across starting points. More recent commentary has questioned the merits and operation of the Whole Life Order, as well as noting the limited empirical research on its effects on offenders.

1. Introduction

Since the abolition of the death penalty in England and Wales in 1965,1 only one sentence can be passed for murder: life imprisonment. Yet the sentencing judge must decide what portion of the life sentence the offender will spend in prison – the ‘minimum term’. The need to establish the minimum term on a case-by-case basis reflects the truism that despite the mandatory nature of the life sentence, not all murders are like; murder encompasses a wide spectrum of harm and culpability. Schedule 21 to the Sentencing Act 2020 provides the statutory framework the judge must have regard to when determining the appropriate minimum term. Drawing on the major research, sentencing decisions and official data published, this paper sketches Schedule 21’s application, amendments, and impact since its enactment in 2003. It begins with introducing the offence of murder, the mechanics of Schedule 21 and its guiding framework for the sentencing judge, before describing sentencing trends which Schedule 21 has given rise to. The discussion then turns to how the appellate courts have interpreted and applied Schedule 21 to achieve a just result, paying attention to the flexibility inherent in Schedule 21, as well as specific issues arising from young offenders, knife murders and the Whole Life Order. It then reflects on ideas and initiatives to reform Schedule 21 considering the ongoing critiques.

2. An Overview of the Statutory Scheme

A. The offence of murder

As the Report of the Royal Commission on Capital Punishment observed as far back as 1953, ‘there is perhaps no single class of offences that varies so widely in character and culpability as the class comprising those which may fall within the comprehensive common law definition of murder’ (at 433). The core definition of murder is an ‘unlawful killing with malice aforethought’ (derived from Coke’s Institutes, 3 Co Inst 47). What distinguishes murder from other offences which result in death – the various forms of manslaughter – is the mens rea, a legal term used to describe the mindset with which the offender killed. Most straightforwardly, the offence of murder is satisfied where the defendant intends to kill (it was their aim or purpose). This alone is wide enough to capture a range of murderers, as has long been recognised by the appellate courts, from ‘multiple sadistic murders at one end and mercy killings at the other’.2

Matters get even more complicated beyond cases of clear intent on the part of the offender to kill because of the breadth of the legal definition of murder. First, a defendant is liable to be convicted of murder where the defendant intended to cause really serious bodily harm and death results instead. This could include, for example, the offender who violently punches the victim in the head, causing a head injury which requires surgery, and the victim then dies in surgery. Second, a jury is entitled to convict a defendant for murder where they find death or really serious harm was a virtually certain consequence of the defendant’s conduct and the defendant knew this to be so.3 The result, in the words of Lord Hailsham, is that those being sentenced for murder range ‘from brutal, cynical and repeated offences… to the almost venial, if objectively immoral “mercy killing” of a beloved partner’.4

B. The mandatory life sentence

The only sentence that a court can pass for murder is life imprisonment.5 Where the offender is aged 18 but under 21 the mandatory life sentence is referred to as custody for life.6 For those under 18, it is referred to as Detention at His Majesty’s pleasure’.7 The life sentence is comprised of three components:

  1. The minimum term: the term that is served in full, in prison. This is the time the offender must spend in prison before they can formally apply to be released on licence. The core sentencing principle determining this term of the sentence is ‘seriousness’ (although deterrence has also been referenced as a principle here too – see R v Peters and Ors [2005] EWCA Crim 60, at para. 4).
  2. The post-term: a term of continued detention to be determined by the Parole Board based on considerations of public protection. The core sentencing principle at this stage is ‘dangerousness’.
  3. Release on licence: this lasts for the rest of the offender’s life, with conditions capable of being attached to the offender on release, breach of which can result in the possibility of recall to prison. A Probation Practitioner may give consideration to requesting the suspension of the supervisory conditions on the life licence after 10 years of continuous, trouble free resettlement and good behaviour in the community (Ministry of Justice, 2020, para. 5.3.7).

C. Schedule 21 to the Sentencing Act 2020

Schedule 21 establishes a series of general principles the sentencing judge ‘must have regard to’ in considering the seriousness of the offence and thus the appropriate minimum term on the facts of the particular case. Schedule 21 attracted little discussion during its legislative passage through the House of Commons, most probably because parliamentarians – mistakenly – anticipated it would have limited effect on overall sentencing practices (Roberts and Saunders, 2020). Schedule 21 is now attached to the Sentencing Act 2020, which consolidates almost all sentencing provisions. Schedule 21 has been subject to various amendments, the most significant of which have been the expansion of the types of cases falling within its higher starting points, the creation of a new starting point of 25 years and the introduction of a range of starting points for offenders aged 14-17.

The court has a statutory duty to sentence the offender taking into account the seriousness of the offence, or the combination of the offence and one or more offences associated with it.8 To get a sense of how this assessment of seriousness is determined, it is worth outlining the key stages the court will progress through in arriving at the minimum term the offender will serve.

Stage 1 – Selecting the starting point

Schedule 21 creates a series of ‘starting points’ that are ranked according to the seriousness of the offence, with most providing a list of cases that will normally fall within the category. In considering the seriousness of the offence, the court ‘must have regard’ to the general principles set out in Schedule 21. The starting points are summarised in Figure 1 below.

Figure 1:  Starting Point Sentences For Murder
Paragraph of Schedule 21Cases normally falling within this starting pointAgeStarting Point
Para 2(a) Murder of two or more persons, where each murder involves any of the
following—  
(i) a substantial degree of
premeditation or planning,  
(ii) the abduction of the victim,
or  
(iii) sexual or sadistic conduct,
(b) Murder of a child if involving the abduction of the child or sexual or sadistic motivation;
(c) Murder of a police officer or prison officer in the course of his or her duty;
(d) Murder done for the purpose of advancing a political, religious, racial or ideological cause;
(e) Murder by an offender previously convicted of murder
Over 21 (or aged 18-20 if ‘exceptional circumstances’)Whole Life Order                        
Para 3    (a) a murder involving the use of a firearm or explosive,
(b) a murder done for gain (such as a murder done in the course or furtherance of robbery or burglary, done for payment or done in the expectation of gain as a result of the death),
(c) a murder intended to obstruct or interfere with the course of justice,
(d) a murder involving sexual or sadistic conduct,
(e) the murder of two or more persons,
(f) a murder that is aggravated by racial or religious hostility or by hostility related to sexual orientation,
(g) a murder that is aggravated by hostility related to disability or transgender identity,
(h) a murder falling within paragraph 2 committed by an offender who was aged under 21 when the offence was committed.
Over 1830 Years
1727 Years
15 or 1620 Years
14 or under15 Years
  
Para 4The offender took a knife or other weapon to the scene intending to— (a) commit any offence, or
(b) have it available to use as a weapon, and used that knife or other weapon in committing the murder.
Over 1825 Years
1723 Years
15 or 1617 Years
14 or under
13 Years
Para 5The case does not fall within paragraph 2, 3 or 4.Over 1815 Years
1714 Years
15 or 1610 Years
14 or under8 Years

Stage 2 – Accounting for aggravating and mitigating factors

Having chosen the starting point, the court should apply any aggravating or mitigating factors to the extent that it has not already done so in its choice of starting point.9 The statutory aggravating and mitigating factors are listed below. The mental state of the offender is of particular significance in determining culpability in the context of murder, reflected in all but one of the mitigating factors referring to this.10 Schedule 21 does not identify all the aggravating and mitigating factors but merely provides relevant examples.11 As discussed below, there are statutory reforms afoot which will expand the aggravating and mitigating factors to take account of specific features of domestic homicides.

Aggravating factors (additional to those mentioned in paragraphs 2(2), 3(2) and 4(2)) that may be relevant to the offence of murder include—

  • (a) a significant degree of planning or premeditation,
  • (b) the fact that the victim was particularly vulnerable because of age or disability,
  • (c) mental or physical suffering inflicted on the victim before death,
  • (d) the abuse of a position of trust,
  • (e) the use of duress or threats against another person to facilitate the commission of the offence,
  • (f) the fact that victim was providing a public service or performing a public duty, and
  • (g) concealment, destruction or dismemberment of the body.

Mitigating factors that may be relevant to the offence of murder include—

  • (a) an intention to cause serious bodily harm rather than to kill,
  • (b) lack of premeditation,
  • (c) the fact that the offender suffered from any mental disorder or mental disability which (although not falling within section 2(1) of the Homicide Act 1957) lowered the offender’s degree of culpability,
  • (d) the fact that the offender was provoked (for example, by prolonged stress) but, in the case of a murder committed before 4 October 2010, in a way not amounting to a defence of provocation,
  • (e) the fact that the offender acted to any extent in self-defence or, in the case of a murder committed on or after 4 October 2010, in fear of violence,
  • (f) a belief by the offender that the murder was an act of mercy, and
  • (g) the age of the offender.

Stage 3 – Further considerations

Three general sentencing provisions are not restricted from application when the sentencing judge is arriving at the offender’s minimum term. Two of these provisions are aggravating. First, the court must treat as an aggravating factor each relevant previous conviction that it considers can reasonably be so treated.12 The sentencing judge will have particular regard to the nature of the relevance of the prior conviction to the murder offence and the time that has elapsed since the previous conviction. Second, if the murder was committed while the defendant was released on bail, the judge must treat that as an aggravating factor.13 The final additional provision is the offender’s guilty plea. The court must consider the stage in the proceedings at which the offender indicated the intention to plead guilty, and the circumstances in which the indication was given.14 Where the judge decides that it is appropriate to fix a minimum term, an appropriate credit for the guilty plea should be deducted from the minimum term which the judge would have imposed but for the plea.15 The maximum reduction for a guilty plea, set out In the Sentencing Council’s Reduction in Sentence for a Guilty Plea guideline, is the lesser of one-sixth of the minimum term or five years. There are no offence-specific guidelines for murder but the Sentencing Council’s General Guideline: Overarching Principles remains relevant for all offenders sentenced, including those for murder.

Stage 4 – Articulate reasons

The sentencing court has a general duty to state in open court, in ordinary language and in general terms, its reasons for deciding on the sentence.16 The court must state which of the starting points in Schedule 21 it has chosen and why, as well as stating any reasons for departing from the applicable starting point.17

D. Comparison with determinate sentences

The life sentence for murder can be contrasted with determinate sentences in several respects. Firstly, in general, Parliament establishes the maximum penalty for the most serious instances of the offence and the Sentencing Council, through its offence-specific guidelines, issues a series of starting points within category ranges, based on harm and culpability. For murder, however, it is Parliament, in Schedule 21, that establishes the starting points. Secondly, a sentencing judge ‘must follow’ the Sentencing Council’s guidelines, unless it is in the interests of justice not to do so, or the guidelines are not relevant to the case. In setting the minimum term for murder, however, the judge ‘must have regard to’ the general principles set out in Schedule 21.18 The law seems, therefore, to allow greater discretion to a judge determining the minimum term for murder than it does a judge passing sentence for a lesser offence using a Sentencing Council guideline. Thirdly, because the sentence of murder is fixed by law, the general principles of sentencing set out in section 142 of the Sentencing Act 2020 do not apply. A judge is, therefore, engaging in a ‘very different’ task when deciding the minimum term than when deciding the length of a determinate sentence, where the judge is only directly concerned with ‘seriousness’, the protection of the public being provided by the imposition of the life sentence.19 Finally, when it comes to the origins of the sentence, almost all determinate sentences are arrived at by the sentencing judge applying an offence-specific Sentencing Council guideline. Unlike Schedule 21 and many of its amendments which have been enacted in haste and with minimal consultation, the Council’s guidelines are the product of a lengthy, deliberative process. This involves prior research, policy and legal investigations, and Sentencing Council deliberation that result in a draft guideline being produced and consulted upon, in conjunction with a draft resource assessment and statistical bulletin. After considering the consultation responses, it is only at this stage that a final version of the guideline is developed (Sentencing Council, 2023).

Prior to the introduction of Schedule 21 in 2003, the guidance issued to courts contained a series of starting points which were lower than those created by Schedule 21. In 1997, the then Lord Chief Justice, Lord Bingham, recommended to judges that a minimum term of 14 years was to be served for the ‘average’, ‘normal’ or ‘unexceptional’ murder and a minimum term of 30 years in rare cases. In 2002 the Sentencing Advisory Panel – the predecessor to the Sentencing Council – gave guidance to the Court of Appeal on the length of minimum terms. It suggested that there should be a higher (15/16 years), middle (12 years) and lower (8/9 years) starting point. In contrast, Schedule 21 establishes a starting point of 15 years for a ‘normal’ murder. It is unsurprising, therefore, that the Sentencing Council identified Schedule 21 as one of the most significant legislative changes affecting the increase in the prison population, observing that because of Schedule 21, ‘sentences for the vast majority of murder cases increased substantially’ (Sentencing Council, 2018, p. 3).

The Court of Appeal has observed, with the passing of Schedule 21, ‘Parliament’s intention seems clear: crimes which result in death should be treated more seriously and dealt with more severely than before’.20 The average minimum term has risen from 13 years in 2000 to 21 years in 2021 – an increase of over 60% (figures cited in Prison Reform Trust, 2023, p. 12). Further, on average people serving mandatory life sentences for murder are spending more of their sentence in prison, spending 18 years in custody, up from 13 years in 2001 (Prison Reform Trust, 2023, p. 12).

The increase in minimum terms for murder has also inflated sentence lengths for other serious offences, most notably manslaughter. The average custodial sentence length for manslaughter almost doubled from 5.4 years in 2007 to 8.8 years in 2017 while the average prison sentence for sexual offences increased by 50 per cent (Roberts and Saunders, 2020; see also Jeremy, 2010; Sentencing Council, 2018). This is because these offences have been adjusted to ensure a degree of ordinal proportionality – the term used to refer to the relative seriousness of offences among themselves – with longer minimum terms.

As of June 2023, 7,071 prisoners were serving a minimum term for murder (Ministry of Justice, 2023a, Table A1.5i). As seen in Figure 2, the number of such prisoners has, overall, been growing for the past eight years. Particularly striking is the number of people serving a Whole Life Order, which has grown from 23 prisoners in 2000 to at least 67 in 2021, an increase of over 174 per cent (Ministry of Justice, 2023d).

Figure 2: Trends in Prisoners Serving a Minimum Term for Murder (2015-2023)

Source – MoJ (2023a), Prison population: June 2002 to June 2023, Table A1.5i

4. The Challenges of Achieving a ‘Just Result’ within Schedule 21

In a body of authorities concerning the application of Schedule 21, the Court of Appeal has repeatedly made clear that notwithstanding Schedule 21’s general principles and starting points, the sentencing decision remains one for the judge and that ‘justice cannot be done by rote’.21 The sentencing judge retains discretion under section 322(2) of the Sentencing Act 2020 to determine the minimum term ‘as the court considers appropriate’. What this means, according to the Court of Appeal, is that the judge ‘is free not to follow the guidance if in his opinion this will not result in an appropriate term for reasons he identifies’.22 Similarly, while judges must have ‘regard’ to the principles set out in Schedule 21, ‘[a]s long as he bears them in mind, he is not bound to follow them’.23 It is worth exploring in detail four situations in which the Court of Appeal has tried, with varying degrees of success, to navigate the prescriptive format of Schedule 21 and its statutory starting points so as to reduce the risk of arbitrary or unduly severe sentences.

i. Flexibility within a framework

The significant difference in the starting points in Schedule 21 can ‘present a sentencer with considerable difficulties in his quest to match the penalty to the infinitely variable circumstances of crime’.24 The Court of Appeal has further remarked that, ‘at first sight’ Schedule 21 appears to be ‘an overly prescriptive, unnecessarily complex, and, on occasions, wholly artificial, apparently all embracing, statutory framework’.25 It is within the context of these remarks that the Court of Appeal has tried over the last two decades to identify some flexibility in applying the starting points prescribed by Schedule 21. This is illustrated in two respects.

First, Schedule 21 expressly states, in paragraph 8, that consideration of aggravating or mitigating factors may result in a minimum term of any length (whatever the starting point), or in the making of a Whole Life Order. The starting points ‘must not be used mechanistically so as to produce, in effect, three [now four] different categories of murder. Full regard must be had to the features of the individual case so that the sentence truly reflects the seriousness of the particular offence.’26 Emphasising the movement between starting points, the Court of Appeal has stated the starting points ‘provide a very broad framework for the sentencing exercise. They are so far apart that it will often be impossible to divorce the choice of starting point from the application of aggravating and mitigating factors’.27

A recent high-profile example of how the application of weighty aggravating factors can function to determine the ultimate sentence more so than the starting point is R v Stewart and Ors.28 The appeal concerned, amongst others, the former Metropolitan Police Officer Wayne Couzens, convicted of murdering Sarah Everard. A defining feature was how Couzens abused his knowledge and status as a police officer to commit murder in a way described by the trial judge as ‘warped, selfish and brutal offending, which was both sexual and homicidal’. The bare facts – a single murder involving sexual conduct – brought the case within paragraph 3, a starting point of 30 years. Nevertheless, the Court of Appeal arrived at a Whole Life Order to reflect the gravity of Couzens’ misuse of his role as a police officer and other serious aggravating features. Reiterating that ‘Schedule 21 must be applied in a flexible and not rigid way to achieve a just result’,29 the Court of Appeal opted not to frame the nature of Couzens’ abuse of office as a new category of case belonging to paragraph 2 (as the trial judge had done) but rather to recognise the aggravating factors as being of a level of seriousness so exceptionally high that a Whole Life Order should be made.

In other cases, the Court of Appeal has emphasised that it is not enough to list aggravating and mitigating factors and then consider which list is the longer, akin to a mathematical calculation: various factors interlink, just as the potency of factors may vary.30 In R v Inglis,31 for example, the offender was the victim’s mother and, out of mercy, deliberately injected her son with a lethal dose of heroin after he had suffered catastrophic head injuries. There were multiple aggravating factors: the offender was in a position of trust, and carefully planned the killing of her particularly vulnerable son. However, the Court of Appeal remarked that the aggravating factors in the case should not, on the particular facts, be taken ‘to aggravate a murder committed by an individual who genuinely believes that her actions in bringing about the death constitute an act of mercy’.32 The minimum term was reduced from nine years to five years.

Second, the list of cases accompanying each starting point are not exhaustive. Paragraphs 2(2), 3(2) and 4(1) of Schedule 21 list the types of case where the seriousness is ‘normally’ to be regarded as ‘exceptionally high’, ‘particularly high’ or ‘sufficiently serious’. Drawing attention to the word ‘normally’, the Court of Appeal has explained that Schedule 21 does not exclude the possibility other cases might reach the indicated level of seriousness, albeit such cases are ‘probably rare’.33 The point applies in reverse. A case that normally comes within the ambit of paragraphs 2(2), 3(2) or 4(1) may not reach that level of seriousness because of its particular facts. A mechanical application of the starting point would risk creating, in the words of Lord Hughes, ‘absurd anomalies.34

An oft-cited instance of such an anomaly is Height & Anderson. The case concerned the murder of the defendant’s wife, killed by a paid contract killer hired by the defendant. The straightforward application of Schedule 21’s starting points by the trial judge resulted in the contract killer being subject to a starting point of 30 years (a killing done simply in pursuit of financial gain). This was double the 15 year starting point applied to the husband who incited and engaged the contract killer and for whose purposes the killing was carried out. As a matter of ‘elementary justice’ the Court of Appeal recorded its surprise at this outcome. It increased the husband’s starting point to 30 years to reflect the many aggravating features which made the case of ‘particularly high seriousness’.35

ii. Young offenders

In 2022, there were significant amendments to Schedule 21 made via section 127 of the Police, Crime, Sentencing and Courts Act 2022. The single 12 year starting point for those aged under 18 was replaced with a total of nine new starting points according to whether the offender was aged (a) 14 or under (b) 15-16 or (c) 17 at the time of offence (see Figure 1), as read in conjunction with what the starting points would have been had the offender been aged 18 or over. In the White Paper preceding these amendments, the Government’s view was that the ‘significant gap’ between the starting points for children and young adults was ‘unfair to the families of victims who can often feel as though the person responsible … has missed being eligible for an appropriate minimum term on account of a matter of weeks or months of difference in age’ (Ministry of Justice, 2020, para. 317). As observed by Wasik, these reforms were influenced by the high-profile case in which a 17-year-old victim was murdered in her own home by a former boyfriend of the same age; the boyfriend receiving a life sentence with a minimum term of 12.5 years (2021, p. 1058). Absent in the White Paper, though, was any reference to either the Sentencing Council guideline on ‘Sentencing Children and Young People’ or the appellate guidance on murder committed by young offenders (Wasik, 2021, p. 1058).

The offender’s age does not necessarily equate with their maturity. Maturity has long been recognised by the courts and Sentencing Council as relevant to the seriousness of the offence (Emanuel, Mawer and Janes, 2021) and the age of the offender is a specific mitigating factor under paragraph 11(g) of Schedule 21. In assessing the potential for unduly severe or arbitrary decisions under the newly amended starting points, it is illustrative to consider the Court of Appeal’s concerns in applying rigid starting points under the old starting point of 12 years for all offenders aged under 18. In Peters, Lord Judge observed the passage of an offender’s 18th or 21st birthday ‘does not necessarily tell us very much about the individual’s true level of maturity, insight and understanding’.36 A rigid application of the starting points risked accidents of time leading to years of a difference in minimum terms and, as such, the Schedule needed ‘to be approached with an acute sense of how inevitably imprecise the statutory criteria may sometimes be to issues of culpability.’37 How was this to be achieved? Consistency of approach and adherence to the relevant legislative provisions requires the sentencing judge to begin from the prescribed statutory starting point. However, Lord Judge in Peters stated that: ‘Schedule 21 does not envisage a moveable starting point, upwards or downwards, from the dates fixed by reference to the offender’s eighteenth or twenty-first birthdays…The principle is simple. Where the offender’s age, as it affects his culpability and the seriousness of the crime justifies it, a substantial, or even a very substantial discount, from the starting point may be appropriate.38

This principle was applied to one of the appellants in Peters who was 18 years and two months at the time of the murder. Accordingly, the trial judge took a 15 year starting point and discounted it to 12 years to reflect the appellant’s age. As the Court of Appeal observed, if the offence had taken place three months earlier, the legislative framework itself would have provided for a starting point of 12 rather than 15 years. Taking account of the appellant’s youth and emotional immaturity, the Court of Appeal reduced the minimum term to nine years.

iii. Knife murders

In 2010, paragraph 5A of Schedule 21 was introduced (now paragraph 4 of Schedule 21). This created a 25 year starting point where a knife or other weapon was taken to the scene, and the offender intended to use it commit an offence or have it as a weapon, and ultimately did so when committing the murder. By way of context, in the year ending March 2022, the most common method of killing was use of a sharp instrument, including knives (41% of victims) (House of Commons Library, 2023, p. 24). Notably, of those homicides with victims aged 13 to 19, 74% were caused by sharp instruments. In addition to these homicides, in 2021/22 there were 441 recorded offences of attempted murder involving knives or a sharp instrument (House of Commons Library, 2023, p. 26). Yet, as Roberts and Saunders (2020) have commented, by inserting a 25 year starting point for cases in which a weapon was taken to the scene, paragraph 5A introduced a deterrence-based severity premium which is at odds with the concepts of harm and culpability.

The primary difficulty arising from paragraph 5A – aside from the paragraph’s failure to provide an adjective to describe the level of seriousness of the offence – is the term ‘took the knife or other weapon to the scene’. This potential for injustice was described by the Court of Appeal in the case of R v Kelly and others.39 If a man makes up his mind to kill his partner, walks back to their home, picks up a knife in the kitchen and kills her with the knife, he will not have taken the knife to the scene. However, if the same man walks home, buys a knife on the way, and kills his partner in the kitchen in exactly the same circumstances, then, on the wording of paragraph 5A, he will have taken a knife to the scene. In the former, the defendant’s minimum term starting point will be 15 years; in the latter it will be 25 years – yet the culpability is comparable in both instances. The provision’s reference to ‘the scene’ makes things more problematic still. If the victim is in the kitchen, the defendant takes a knife from a drawer and kills him or her, that knife was not taken ‘to the scene’ per paragraph 5A. If the kitchen is at one end of the living room with no partition between the two, the victim is in the living room and the defendant takes a knife from the kitchen drawer and kills her, then this knife was still not ‘taken to the scene’.40

Where disparate minimum terms do arise in the context of paragraph 5A and the offender’s culpability remains similar, this would, in the words of the Court of Appeal, ‘not represent justice in anyone’s assessment’.41 In an effort to do justice, the Court of Appeal has applied paragraph 5A of Schedule 21 in the following manner. Where a knife is taken from a kitchen to another part of the same flat or house, including a balcony,42 it will not normally be regarded as having been taken to the scene, even if a door is forced open. However, if the knife is taken out of the house or flat into the street, or into another part of the premises, or on to a landing outside a flat, it will normally be regarded as having been taken to the scene. But even then, if a case is only just within paragraph 5A, because a knife was taken from a kitchen and used to inflict a fatal wound a short distance outside the door of the flat or house, this principle may well lead to a minimum term of less than 25 years.43 Bild (2011) is critical of both of paragraph 5A as a provision for creating an arbitrary distinction between whether a weapon was or was not taken to the scene, and the efforts of the Court of Appeal in Kelly to provide some clarity to it. As to the latter, Bild asks, ‘Is a front garden a different scene? How about paths and doorsteps? Could a defendant really be subjected to a two thirds increase in the starting point simply because an offence is one foot one side of a doorstep rather than the other? There is little in Kelly to dispel this concern’ (2011, p. 8).

iv. The Whole Life Order (WLO)

A WLO means that the statutory early release provisions do not apply. It does not preclude the possibility of release by the Secretary of State for Justice on compassionate grounds. Whole life prison terms only began to emerge in the latter half of the twentieth century, following the Government’s tariff-setting procedures introduced in 1983 (Appleton and Gilman, 2022). The Criminal Justice Act 2003 extended the scope of life imprisonment to include the WLO as the ultimate penalty in England and Wales. Recently, the Police, Crime, Sentencing and Courts Act 2022 expanded the WLO in two ways. First, under section 125, it expanded the range of circumstances for which a WLO will normally be an appropriate starting point for those who murder a child to include the premeditated murder of a child. Second, section 126, enabled judges to impose a WLO on offenders aged 18 to 20 years in ‘exceptional cases’ where such a sentence would be warranted.

Further amendments to WLOs are contained in the Sentencing Bill, introduced by the Government in November 2023. Clause 1 of the Bill will create a new duty for the court to impose a WLO for cases of murder which are currently normally subject to a WLO starting point, unless the court is of the opinion that there are exceptional circumstances which justify not making a WLO. As explained in the Bill’s Explanatory Notes, it will be for the court to consider the individual circumstances of a case in passing sentence to determine whether there are exceptional circumstances. The interpretation of exceptional circumstances is therefore left to judicial discretion (Ministry of Justice, 2023f, para. 16). In addition, Clause 1 will also move the category of a single ‘murder involving sexual or sadistic conduct’ into the new framework for WLOs. As a result, the court will be required to impose a WLO for these murders, rather than them being subject to a 30 year starting point. Following a number of high-profile murders, where women have been attacked and killed by male assailants and there has been a sexual or sadistic element to the murder, the Government considers this expansion of the WLO as necessary to ensure ‘those who commit the very worst crimes are given the most severe punishment available’ (Ministry of Justice, 2023f, para. 20).

The Court of Appeal has long made it clear that the threshold of seriousness to warrant a WLO is exceptionally high. Stated in Jones, it will be a case in which the facts ‘will leave the judge in no doubt that the offender must be keep in prison for the rest of his or her life’.44 It is ‘a sentence of last resort for cases of the most extreme gravity’,45 which is ‘reserved for the few exceptionally serious cases’ where ‘the judge is satisfied that the element of just punishment requires the imposition of a whole life order’.46 Restating the flexibility within Schedule 21, the Court of Appeal has explained that if one or more of the factors set out in paragraph 4(2) is present, it is likely to be case requiring a WLO. Nevertheless, the judge must consider all the material facts before concluding that a very lengthy finite term will not be a sufficiently severe penalty.47 The material facts include a guilty plea entered by the defendant, albeit a case which calls for a WLO is unlikely to be a borderline case in the first place.

The ability to review a WLO has been subject to sustained legal challenges before the European Court of Human Rights (ECtHR). In 2013, in Vinter and Ors v UK,48 the Grand Chamber of the ECtHR held that the WLO in England and Wales violated article 3 of the European Convention on Human Rights (the right to freedom from torture or inhuman or degrading treatment or punishment). The fact a prisoner might spend the whole of their life in prison was not a violation of article 3, but the article does require reducibility of the sentence. Specifically, it must allow the authorities to ‘consider whether any changes in the life prisoner are so significant, and such progress towards rehabilitation has been made in the course of the sentence, as to mean that continued detention can no longer be justified on legitimate penological grounds.’49 In Vinter, the ECtHR held the legal regime in England and Wales, including the absence of any dedicated review mechanism, meant the WLO was not reducible for the purposes of article 3. In 2017, in the later judgment of Hutchinson v UK,50 however, the Grand Chamber adopted a more generous, if not optimistic, reading of the domestic case law and policy documents to decide that there was, despite no material change in the legal regime, sufficient prospect of release and a possibility of review for the administration of the WLO to be compatible with article 3 after all.51

5. Criticism of Schedule 21 and Future Reform

Schedule 21 has attracted a great deal of scholarly critique over its framework and its effect on minimum terms for murder as well as sentencing for other serious offences (Roberts and Saunders, 2020). Even the Government itself has been critical of Schedule 21. In 2010, a Green Paper described the Schedule as being ‘based on ill-thought out and overly prescriptive policy. It seeks to analyse in extraordinary detail each and every type of murder. The result is guidance that is incoherent and unnecessarily complex, and is badly in need of reform so that justice can be done properly in each case’ (Ministry of Justice, 2010, p. 51). Despite such a frank acknowledgement of the Schedule’s flaws, no remedial reforms have emerged in the decade or so since. Having examined the Court of Appeal’s effort to address the risk of arbitrary or unduly severe results from a straight-forward application of Schedule 21, it is worth turning now to consider the sustained critiques of Schedule 21. This is followed by a brief discussion of potential reform options, drawing on some very recent suggestions, as well as those mooted since the very introduction of the mandatory life sentence.

i. Critiques of Schedule 21

The effect of Schedule 21’s wide gap in starting points means the presence of a single aggravating factor is meant to justify the doubling of the starting point, while the offender’s age can result in one of nine different starting points. Roberts and Saunders (2020) have raised concern that there is a danger this leads to unwarranted differences in minimum terms unless sentencers attach great weight to other elements of aggravation and mitigation. Relatedly, they have questioned on proportionality grounds the relationship between the starting points and aggravating and mitigating factors. They ask, should the infliction of mental or physical suffering before death (in paragraph 10) be regarded as broadly equivalent to sexual or sadistic conduct (in paragraph 5)? And should murders of other people performing a public duty (paragraph 10) also be treated as especially serious, perhaps equivalent to other cases that alter the starting point?

Mitchell (2016) argues that whilst all murders that contain the factors identified in the higher starting point categories of Schedule 21 are undeniably serious, it is difficult to see why cases have been categorised as they have. While the murder of a single police officer or prison officer is obviously very serious, Mitchell queries whether it is really of broadly equivalent seriousness to those of the murder of multiple people, especially where there is substantial premeditation or planning, or abduction of the victim. Murders motivated by the advancement of a political, religious, racial or ideological cause, are to be treated as exceptionally serious because they represent an affront to the views of many and put people’s lives and interests at risk. But Mitchell suggests a similar argument could well be made for murders by explosive, which indicates a lower starting point.

The inflation of the starting points for the minimum term that the introduction of Schedule 21 and its subsequent amendment has provided for has been linked to a climate of penal populism and crime control. The unanimous perception amongst the 26 judges and barristers in homicide cases interviewed by Fitz-Gibbon (2016) was that Schedule 21’s restrictive sentencing scheme, in so far as it requires judges to identify a starting point determined by Parliament, was fuelled by political motivations to appear tough on crime and retain Government control over sentencing.

The academic literature has considered the WLO in England and Wales from various vantage points. A principled critique of the WLO is made by Mitchell (2016) on the basis that it imposes a unique degree of mental punishment. For Mitchell, there are no obvious aggravating factors, or combination of factors, that warrant this degree of punishment, leading him to argue that any list of factors compiled for such a purpose would be purely arbitrary. Mitchell accompanies this with an instrumental basis for opposing the WLO. Because the offender has ‘nothing to lose’ and efforts to rehabilitate become redundant without the prospect of any future release, it undermines the likelihood that the offender will adopt a constructive or co-operative approach to the sentence. Worse still, Mitchell suggests the offender might pose serious problems of management and control for prison staff.

Pettigrew (2015), meanwhile, questions the ECtHR reasoning in Vinter – that the penological justification for the sentence can shift, and shift to such an extent continued imprisonment can no longer be justified. First, imposing a WLO is expressly linked to the heinous nature of the offence. For the Minister responsible for reviewing penological principles and the justification for continued imprisonment, the starting point to such a review will remain the heinous nature of a case that gave rise to the WLO in the first place. In other words, can retribution ever fully be exhausted? Second, the prospect of release will be diminished by other factors likely weighing on the relevant Minister’s mind – such as public outrage, media lobbying, and political cost, factors which, Pettigrew notes, have influenced the political decision to oppose release in previous cases.

There has been no comprehensive empirical study of prisoners serving WLOs in England and Wales. This seems especially significant given the increasing use of the sentence. Appleton and Gilman (2022) have, however, recently reviewed the Ministry of Justice data relating to prisoners serving a WLO in England and Wales. Their analysis revealed that a significant number of prisoners serving WLOs died at a relatively young age: approximately 20% died at an average age of 63 years old, having served, on average, 19 years in prison. This is 18.5 years below the national average life expectancy. Appleton and Gilman argue this raises ‘significant questions about the imposition and impact of WLOs, the detrimental effects of lifelong detention, the conditions of confinement and regime for this group of prisoners and their physical and mental health’ (2022, p. 6). Accordingly, they advocate for a review and assessment of what it is like to be sentenced to die in prison.

ii. Proposals for reform

In 2005, the Home Office tasked the Law Commission to conduct a comprehensive review of the law of murder. The Law Commission’s report, published in 2006, recommended creating a distinction between first-degree murder (intent to kill, or intent to cause serious injury where the killer was aware that his or her conduct involved a serious risk of causing death) and secondary- degree murder (intent to cause serious harm, or intent to cause injury or fear or risk of injury where the killer was aware that his or her conduct involved a serious risk of causing death). The Law Commission recommended that only first-degree murder would carry the mandatory life sentence. As such, it would be confined to the most serious kinds of killing, leaving available the discretionary life sentence for less serious, but still highly blameworthy, killings. Ultimately, this recommendation was not implemented.

A more modest but nonetheless progressive proposal is the introduction of a late sentence review mechanism for those serving the minimum term of the mandatory life sentence. Advocated for by Roberts and Saunders (2020), this would acknowledge that higher minimum terms are unlikely to achieve the preventive benefits relied upon to justify them, and an oversimplicity in asserting that the determination of the minimum term is concerned with considerations relating only to the circumstances at the time the offence took place. But what would trigger a late sentence review? Roberts and Saunders suggest it be narrowly tailored to ensure only the most meritorious cases are eligible – for example, where a prisoner demonstrates exceptional progress towards rehabilitation. A similar system operates for those aged under 18 at the time of their murder conviction. After serving the likes of, say, a 30 year minimum term, the offender is physically and psychologically a different person, especially if the offender was a young adult at the time of the offence. As for concerns of recidivism, figures reported by the Prison Reform Trust (2023) show that just 4% of those sentenced to a mandatory life sentence were reconvicted of any criminal offence within a year, compared to 42% of the overall prison population.

The reform of Schedule 21 to reflect the harm and culpability specific to domestic homicides is, at the time of writing, a particularly live issue. Questions have arisen as to whether the Schedule reflects contemporary understandings of the causes, characteristics and harms of fatal domestic abuse. By way of context, in the last decade there have been an average of 129 domestic homicides each year. Over this period, domestic homicides accounted for 60% of homicides with female victims, but 9% of homicides with male victims (House of Commons Library, 2023, p. 18). An open letter from the Victims’ Commissioner and the Domestic Abuse Commissioner in 2021 and a campaign by families of two women murdered by their male partners prompted the Domestic Homicide Sentencing Review, conducted by Clare Wade KC, which reported in 2023. The Review highlighted the gendered nature of homicides. The perpetrator of homicide was an intimate partner and/or ex-partner in 350 homicides, of which 87% are male perpetrators, and just 13% female perpetrators (para. 5.1.6). When women are convicted of murdering their male partners, they often tend to be the victims of previous domestic abuse by their partners (para. 5.3.1).

The Review recommended an increased focus on the extent of culpability by way of coercive control in domestic murders but advocated that this should not be achieved by introducing an additional (higher) starting point for domestic murders into Schedule 21. Instead, it recommended domestic murders should be given specialist consideration within the present sentencing framework under Schedule 21. Firstly, paragraph 9 (aggravating factors) should be amended to include the following factors: (i) if a murder takes place at the end of a relationship or when the victim has expressed the desire to leave; (ii) if there is ‘overkill’ (extensive and gratuitous violence perpetrated against the victim); and (iii) if there is history of coercive control of the victim of a murder by the perpetrator of that murder. Secondly, where there is a history of coercive control having been perpetrated by the victim of the murder against the offender, then this should be a statutory mitigating factor and paragraph 10 of Schedule 21 should be amended accordingly.

The Government’s response to the Review has been three-fold. First, in July 2023, the Ministry of Justice accepted recommendations (ii) and (iii) above, as well as the proposed mitigating factor (Ministry of Justice, 2023b) and has now sought to implement them by way of secondary legislation (Sentencing Act 2020 (Amendment of Schedule 21) Regulations 2023) tabled in November 2023. Second, recommendation (i) is to be given effect by a provision in the Criminal Justice Bill which will make it an aggravating factor that the offence was connected with the end of an intimate personal relationship with the victim. Third, in November 2023, the Government began consulting on a 25 year starting point for murders preceded by controlling or coercive behaviour by the perpetrator against the victim of the murder; such a change in starting point was expressly discouraged by the Review (Ministry of Justice, 2023e). In its scrutiny of the 2023 Regulations, the House of Lords Secondary Legislation Scrutiny Committee expressed its reservations about the Government’s piecemeal approach to implementing the recommendations, observing that whilst it understood the desire to make changes rapidly where possible, in general, it was better policymaking to make all related changes at the same time (Secondary Legislation Scrutiny Committee, para. 63).

6. Conclusion

This paper has examined the origins, content, application and critique of Schedule 21, the scheme that governs the mandatory life sentence for murder in England and Wales. Most striking is the inflationary impact Schedule 21 has had on the length of time offenders are serving the custodial part of their sentences. The average minimum term has risen from 13 years in 2000 to 21 years in 2021. Such inflation was the foreseeable consequence of the increased starting points legislated for in Schedule 21 and, over the subsequent two decades, the grafting onto the Schedule of Government-led amendments, the overwhelming majority of which have been more punitive. Particularly noteworthy has been the addition of the knife provisions and the recent multiplication of starting points for offenders aged under 18.

Operationally, Schedule 21’s starting points, the cases within each category and the listed aggravating and mitigating factors have proven difficult for sentencing judges to apply to the myriad facts and degrees of culpability that characterise murder, as illustrated in the decisions of the Court of Appeal. In an effort to reduce the likelihood of arbitrary or unduly severe outcomes arising in certain types of case, the Court of Appeal has tried, with varying degrees of success, to draw on residual flexibility within the Schedule.

Beyond the observations of the Court of Appeal about the Schedule’s potential for injustice, the scholarship has provided a sustained critique of the disproportionality within Schedule 21, the arbitrary outcomes it still gives rise to and the need, therefore, for significant reform. Such observations and criticisms do not appear to be entirely lost on policymakers, as reflected in the Government’s frank admission that Schedule 21 is ‘incoherent and unnecessarily complex, and is badly in need of reform’. For now, at least, this need remains unmet.

References

  1. Section 1(1) of the Murder (Abolition of Death Penalty) Act 1965. The Act suspended the death penalty for capital murder in England, Scotland and Wales for five years. In 1969, the abolition was made permanent. ↩︎
  2. Simon Brown LJ, R. (Anderson and Taylor) v. Secretary of State for the Home Department [2001] EWCA Civ 1698 at para. 56. ↩︎
  3. R v Woollin [1999] AC 82. ↩︎
  4. R v Howe [1987] A.C. 417, at p. 433. ↩︎
  5. Section 1(1) of the Murder (Abolition of Death Penalty) Act 1965. ↩︎
  6. Section 275 of the Sentencing Act 2020. ↩︎
  7. Section 259 of the Sentencing Act 2020. ↩︎
  8. Section 322(2) of the Sentencing Act 2020. ↩︎
  9. Paragraph 7 of Schedule 21 to the Sentencing Act 2020. ↩︎
  10. R v Jones and Ors [2005] EWCA Crim 3115. ↩︎
  11. Lord Woolf CJ in R v Last [2005] EWCA Crim 106, at para. 18. ↩︎
  12. Section 65(2) of the Sentencing Act 2020. ↩︎
  13. Section 64 of the Sentencing Act 2020. ↩︎
  14. Section 73(2) of the Sentencing Act 2020. ↩︎
  15. R v Jones and Ors [2005] EWCA Crim 3115. ↩︎
  16. Section 52(2) of the Sentencing Act 2020. ↩︎
  17. Section 322(4) of the Sentencing Act 2020. ↩︎
  18. Section 322(3) of the Sentencing Act 2020. ↩︎
  19. R v Sullivan [2004] EWCA Crim 1762, at para. 9. ↩︎
  20. R v Wood [2009] EWCA Crim 651 at para. 23. ↩︎
  21. R v Peters and Ors [2005] EWCA Crim 605 at para. 5. ↩︎
  22. R v Sullivan [2004] EWCA Crim 1762 at para. 16. ↩︎
  23. R v Sullivan [2004] EWCA Crim 1762 at para. 12. ↩︎
  24. R v Griffiths and Ors [2012] EWCA Crim 2822 at para. 15. ↩︎
  25. R v Kelly and others [2011] EWCA Crim 1462 at para. 9. ↩︎
  26. R v Jones [2005] EWCA Crim 3115 at para. 8. ↩︎
  27. R v Jones [2005] EWCA Crim 3115 at para. 7. See also R v Sullivan [2004] EWCA Crim 1762, at para. 16 and R v Griffiths and Ors [2012] EWCA Crim 2822 at para. 15. ↩︎
  28. [2022] EWCA Crim 1063. ↩︎
  29. R v Stewart and Ors [2022] EWCA Crim 1063 at para. 82. ↩︎
  30. See R v Peters and Ors [2005] EWCA Crim 605; R v Stewart and Ors [2022] EWCA Crim 1063. ↩︎
  31. [2010] EWCA Crim 2637. ↩︎
  32. R v Inglis [2010] EWCA Crim 2637 at para. 53. ↩︎
  33. R v Height & Anderson [2008] EWCA Crim 2500 at para. 28. ↩︎
  34. R v Griffiths and Ors [2012] EWCA Crim 2822 at para. 15. ↩︎
  35. R v Height & Anderson [2008] EWCA Crim 2500 at paras. 31-32. ↩︎
  36. R v Peters and Ors [2005] EWCA Crim 605 at para. 11. ↩︎
  37. R v Peters and Ors [2005] EWCA Crim 605 at para. 11. ↩︎
  38. R v Peters and Ors [2005] EWCA Crim 605 at para. 12. ↩︎
  39. [2011] EWCA Crim 1462. ↩︎
  40. R v Kelly and others [2011] EWCA Crim 1462 at para. 14. ↩︎
  41. R v Dillon [2015] EWCA Crim 3 at para. 32. ↩︎
  42. R v Senechko [2013] EWCA Crim 2308. ↩︎
  43. R v Dillon [2015] EWCA Crim 3 at para. 32. ↩︎
  44. R v Jones [2005] EWCA Crim 3115 at para. 10. ↩︎
  45. R v Wilson [2009] EWCA Crim 999 at para. 14. ↩︎
  46. R v Reynolds [2014] EWCA Crim 2205 at para. 5. ↩︎
  47. R v Reynolds [2014] EWCA Crim 2205 at para. 15. ↩︎
  48. Vinter and others v United Kingdom (Application nos. 66069/09, 130/10 and 3896/10). ↩︎
  49. Vinter and others v United Kingdom (Application nos. 66069/09, 130/10 and 3896/10) at para. 119. ↩︎
  50. Hutchinson v United Kingdom (application no. 57592/08). ↩︎
  51. See also R v McLaughlin; R v Newell [2014] EWCA Crim 188. ↩︎