I. Introduction
The purpose of this article is to examine how Andrew von Hirsch's proportionality theory understands personal mitigation in sentencing, to demonstrate how it applies to six actual offenders, and to test whether it is consistent with the public's sense of justice. Proportionality entails commensurability between offence seriousness and sentence severity. In this theory there is little room for the proportionate sentence to be reduced on the grounds of an offender's claim to personal mitigation. Nevertheless, von Hirsch does not have the field to himself. A very different view of what constitutes fair punishment is taken by Michael Tonry and the late Norval Morris. To them, an offender's less than full culpability, favourable rehabilitation prospects and general plight in the face of punishment often may warrant a reduction, sometimes substantial, in the proportionate sentence, and matters relevant to these three aspects of personal mitigation are interpreted broadly. Both von Hirsch and Morris and Tonry claim their views represent common-sense justice. With respect to this, the question arises as to which view better represents the general public's sense of justice. The present analysis answers this by means of an empirical study conducted in the community (Melbourne Criminology Sentencing Study). Why is this theoretical contention in respect of personal mitigation important? The fact is, it has implications for guideline reforms of traditional judicial sentencing policy, including those of the Sentencing Guidelines Council. And relevant to this is Robinson's concept of empirical justice, as it applies to proportionality and public opinion.Footnote 1 These matters are addressed in the light of the results of the empirical study of the public's sense of justice. Before proceeding to the study and its results, von Hirsch's and then Morris and Tonry's sense of justice are first set out, followed by an explanation of the significance of the study.
II. Von Hirsch on Personal Mitigation
According to von Hirsch, who in a recent work joined forces with Andrew Ashworth, the purpose of punishment in sentencing is to censure offenders in proportion to their blameworthiness. The more serious the offence, the more blameworthy is the offender, and thus the greater the deserved punishment. Seriousness involves the harm caused, risked or threatened by the offence, as well as the offender's culpability in committing the offence. In practice, seriousness and severity are construed as follows. First, the seriousness of actual offences is defined by way of (statutory) categories of offence (e.g. robbery); for this purpose some offence categories are divided into a couple of sub-categories (e.g. residential/non-residential burglary or intentional/reckless behaviour). Secondly, these various (sub-)categories are grouped and ranked according to a moderate number of (say ten) levels, constituting a scale of offence seriousness. This provides the framework for thinking about proportionate punishment. Thus, for each level of seriousness, there will be a corresponding range of sentence severity deemed to be proportionate. For the higher levels of seriousness, these will be terms of imprisonment. In this scheme, the range of sentence corresponding to each level of seriousness will be relatively narrow and there will be little overlap between adjacent ranges. These two consequences follow from the fact of proportionality being strongly controlling and seriousness being narrowly defined.
This is just deserts sentencing. For the purpose of this, normal human beings are assumed to be rational, moral agents, capable of responding to censure and the threat of censure. Von Hirsch and Ashworth say all this accords with common-sense notions of what is just and fair in the punishment of individual offenders.Footnote 2
What then is the scope for personal mitigation in their scheme? There are three aspects of this: the categories of mitigation and the factors falling within these categories; the potential reduction in sentence when these factors are present; and matters excluded as mitigating.
A. The Content of Personal Mitigation
There are four categories for this: reduced culpability; the absence of prior convictions; equity mitigation; and equality of impact.
1. Culpability
In regard to this, von Hirsch and Ashworth do not set out a list of factors. In fact, what they say is no more than indicative. The idea seems to be that mitigation relies on and represents an extension of the substantive criminal law. Thus, reckless and negligent will be mitigating (cf. intentional), but in individual cases distinctions by way of degrees within these categories may be warranted. As well, from the doctrine of excuses follows their acceptance of certain partial excuses as potentially mitigating. Instances cited in this regard include diminished capacity, partial necessity, duress, provocation, and unforeseeability of the harm. Whether an individual matter should be treated as mitigating will depend on the circumstances. This aside, youth is acknowledged as lessening culpability, its effect progressively lessening up to the age of majority.
2. Absence of prior convictions
First offenders are said to warrant less severe sanctions than otherwise would be proportionate. This is justified in terms of an understanding of and tolerance for human weakness; such individuals deserve a chance to respond to censure. This leniency may be extended to those with several prior convictions, but to a progressively lesser degree.
3. Equity mitigation
This involves compassion for the offender's suffering but justifiable in terms of the desert rationale. Within this category fall offenders who, by virtue of frailty due to advanced age or ill health, will find it hard to respond to the moral demands of censure. Also here come cases for which there has been a significant time delay, and in this period the offenders have changed for the better so that their crimes do not reflect so badly on them. Finally, there will be those who have acknowledged the wrongfulness of their behaviour by making voluntary reparation to the victim or – they appear somewhat uncertain about this – by surrendering to the authorities. Each type of case in its own way may require less censure and, thus, less punishment. General remorse, in respect of equity mitigation, is left open.
4. Equality of impact
Clearly, punishment, particularly imprisonment or community work, will fall more heavily on some; for example, the chronically ill, the disabled, and the elderly. When this difference is potentially significant, the severity of the punishment which otherwise would have been proportionate should be modified so there is approximate equality of impact. In this respect, the young offender merits special consideration under proportionality.
B. Potential Effect of Personal Mitigation
The drawing of conclusions about the potential effect of personal mitigation is not easy, since von Hirsch and Ashworth speak in only general terms about it. Nevertheless, there are clues in their analysis. Thus, in regard to the effect of mitigation, several general propositions must be borne in mind. First, the effect of any mitigation related to the offender normally must not disturb the relationship between offence seriousness and sanction severity. To do otherwise would be to undervalue the relative seriousness of the offence and not adequately censure the offender; indeed, it necessarily would be unfair to other offenders who commit the same offence. Secondly, as discussed above, the range of sanction severity corresponding to each level of offence seriousness will be comparatively narrow. Thus, as a general rule, the scope for personal mitigation will be very limited (i.e. the reduction in the proportionate sentence for the allowable mitigating offender factors, individually, even in combination, will be small relative to the full range of sanction severity). Nevertheless, the potential effect of mitigation is more complicated than this. This is best explained with respect to each of the four categories of mitigation.
1. Culpability
Differences in intent (e.g. intentional cf. reckless) may be regarded as substantial and thus reflecting different offence-seriousness levels. Indeed, as alluded to above, they are often distinguished in statute. However, distinctions in regard to the degree of intent, and so on, represent finer gradations. Thus, their effect on sentence will be less. For this reason, and in view of the practical limitations on the number of levels of offence seriousness, they would appear to lend themselves to being treated within offence-seriousness levels. A similar argument would appear to hold for dealing with the mitigating effects of partial excuses. Nevertheless, for the factor of youth, von Hirsch and Ashworth clearly envisage its effect being a reduction to a level of sentence severity proportionate to a lower level of offence seriousness; for young teenagers the reduction may be very substantial. In fact, they are expressly untroubled by any disproportionately arising on this ground. This is because youth for the individual offender is no more than a temporary state.
2. Absence of prior convictions
The potential discount on the proportionate sentence for good character would appear to warrant moderate but not substantial disproportionality. And the basis for von Hirsch and Ashworth being comfortable with this concession is similar. The state of being a first offender is temporary and does not involve a permanent distortion in the relationship between seriousness and severity.
3. Equity mitigation
In regard to the potential effect on severity of the factor of frailty in regard to censure, von Hirsch and Ashworth are quite clear. The reduction may be very substantial indeed; this factor may warrant a non-custodial penalty though the proportionate sentence be a term of imprisonment. Again, the resulting disproportionality is not problematic, the reason for this being the rarity of such cases. For the factors of time delay and reparation in regard to censure, they are less clear. But there would appear no grounds for a substantial reduction. It does seem that these may justify a disproportionate sentence, but perhaps limited by the sentence proportionate to the level of seriousness immediately below that of the current offence.
4. Equality of impact
The degree of offender sensibility to harsh punishment due to advanced age, illness or disability envisaged here is significant but much less than for equity mitigation. What is required, therefore, is no more than some modification to the onerousness of the proportionate sentence. The idea is that the harshness experienced by this offender should approximate the harshness experienced by an able-bodied offender subjected to the full proportionate sentence. It may be satisfied by no more than substituting physical work with clerical work. For the young offender, however, the discount in severity will be substantial.
C. Personal Matters Excluded as Potentially Mitigating
In order to qualify as potential mitigation in von Hirsch and Ashworth's scheme, a factor must be capable of being justified in terms of the seriousness of the offence – for personal mitigation, this means culpability – the severity of the sanction, or the censuring function of punishment. Moreover, of those factors whose relevance can be justified on these grounds, and are thus permissible, in general, they will be regarded as relevant only when they represent, in the context of the case, a significant feature of the circumstances. Fine-tuning makes no sense when the categories are comparatively narrow. Moreover, were the threshold for disproportionality to be reached readily, the scheme would collapse. For this reason, factors common to cases are viewed askance. As a consequence, in any given case, few factors would be expected to present themselves as personally mitigating. In regard to the mitigating potential of individual factors, the following observations appear to reflect the scheme's intended operation.
For a factor to justify a reduction in the offender's culpability, it must be indicative of less than full intent, or represent a partial excuse. Moreover, the standard for accepting a factor on these grounds is strict – the relationship to the offending must be direct and immediate. Remorse would be patently too remote. In fact, for this reason, social deprivation is rejected as a factor having the potential to reduce an offender's culpability. It was entertained, however, as a ground for compassion; such offenders were seen as experiencing less societal support for compliance. Nevertheless, von Hirsch and Ashworth rejected it on this ground also, because so many offenders would qualify for a less than proportionate sentence as to distort the scale of offence seriousness.
The factor of an offender having made a worthy social contribution over time or by way of a specific act (e.g. social service or an act of heroism) is also rejected. And the fate of the factor of an offender having overcome social disadvantage and made good is similar. In von Hirsch and Ashworth's scheme this follows as a matter of logic, because the behaviour is unrelated to the offence in question. Again, a broader interpretation of culpability would have allowed it to come within that category; this would involve viewing worthy social contributions and achievement in the face of social disadvantage as indicative of a person who is not as bad as the facts of the crime would suggest and, for this reason, deserving a lesser penalty.
Then there is mercy. Compassion, as von Hirsch and Ashworth call it, is also rejected as mitigating. Thus an offender's having experienced a recent misfortune and suffered emotionally because of it does not appear to excite their sense of compassion. In such cases, sanction severity is treated as remaining the same, because this matter is unrelated to the punishment to be served. Nor do they appear persuaded to offer a discount to offenders who suffer as a result of their offending. This includes the collateral consequences of conviction. Instances of these would be an injury sustained during the offence or the loss of employment. Foreseeability seems to be the reason here. But the former, if significant, might warrant a lessening of the sanction's severity on the grounds of equality of impact.
Finally, there is rehabilitation. Von Hirsch and Ashworth are not against offenders' participating in rehabilitation programmes as a part of their sentence, as long as the onerousness of the sanction remains roughly the same. What they do oppose is a disproportionately lenient sentence being imposed to facilitate treatment. This is because any discount for rehabilitation generally would be of no practical value unless it was substantial; yet the basis for this – amenability to treatment – is unrelated to proportionate punishment for the purpose of censure. The consequence of this would involve a serious undermining of proportionality. In any case, a substantial reduction for rehabilitation would be inequitable to offenders whose offending is similarly serious but for whom an efficacious treatment is not available. There is also for proportionality a problem with the very idea of a judgment being made about an offender's amenability to treatment. This is because of the conventional indicators of this (e.g. employment record and remorse) being unrelated to offence seriousness. Modest reductions in severity may be allowed for rehabilitation, but not to such an extent as to be of much practical significance.
More generally, von Hirsch and Ashworth appear to understand personal mitigation solely in terms of individual factors considered in isolation. There is no room for reflection on the type of person before the court. Thus, the culpability of offenders is not reduced by their being not as bad as indicated by the (sub-)category of offence considered alone. Such judgments rely on multiple factors. In any case, they often will be factors not related closely or at all to seriousness of the present offence, such as weakness in the face of temptation or past good works.
D. Ashworth on Personal Mitigation
Ashworth's independent observations on mitigation warrant separate treatment, because they offer several considerations beyond his joint work with von Hirsch, and on one or two matters he appears to arrive at a contrary conclusion.Footnote 3
For Ashworth, a factor should not qualify as mitigating if it fails an equity test (i.e. if it offers an avenue of mitigation not open to the financially or socially disadvantaged). On this basis, he rejects financial reparation – it discriminates against the poor – and loss of employment – the unemployed cannot take advantage of it.
With regard to equality of the impact of a sanction, Ashworth is somewhat parsimonious. He regards the fact of a prison sentence being served in solitary confinement as not ordinarily relevant. He also deals with the offender's family. Again, he does not see it as normally relevant, but he appears to entertain the possibility of exceptional circumstances, such as the fact of a woman having young dependent children.
One important difference between the two analyses concerns the relevance to personal mitigation of social disadvantage or a troubled upbringing. Ashworth comes down in favour of these factors, and is prepared to treat them as potentially mitigating an offender's culpability. The degree to which they should reduce culpability will depend on how direct their relationship is to the offending in question.
Finally, remorse as mitigating is not favoured by Ashworth; presumably, not least because it is removed from the offending. Thus, the linking in the joint work of remorse – acknowledgement of wrong – and mitigation in regard to reparation would seem problematic for Ashworth.
III. Morris and Tonry on Personal Mitigation
Morris and Tonry's conception of what represents fair punishment for individual offenders may be seen as a direct challenge to von Hirsch and Ashworth's theory. That is how it is treated here. Nevertheless, like von Hirsch and Ashworth's theory, their view of justice is grounded in common sense. Indeed, Morris and Tonry expressly state that their approach to punishment corresponds to what most ordinary people find to be fair and sensible in criminal cases. But unlike von Hirsch and Ashworth, they do not offer a highly developed theory; in fact, theirs is largely a statement of what they find to be fair and sensible as they look at life.Footnote 4
Proportionality with a human face is how Tonry describes their position. Thus, as for von Hirsch and Ashworth, the seriousness of the (sub-)category of offence sets an upper limit on the severity of the deserved punishment. But for Morris and Tonry, proportionality is not otherwise strongly controlling. There are two aspects of this in respect of personal mitigation. Firstly, personal mitigation relates not only to an offender's culpability, but also to their rehabilitation prospects and claim to mercy on the grounds of compassion. Secondly, personal mitigation may be seen as warranting a substantial reduction in the sentence proportionate to the seriousness of the offence considered alone. Proportionality does operate on the extent of any reduction, but somewhat weakly (the most serious crimes aside). Thus, there will be many cases for which the sentence is regarded as properly well below what in von Hirsch and Ashworth's terms would be proportionate to the (sub-)category of the offence; but uncommon will be cases in which personal mitigation is seen as warranting a severely disproportionate sentence.
Numerous factors are of potential relevance to personal mitigation, and their source is indeed broad: they may relate directly or indirectly to the offending, or even be not obviously related to the offending; and they may lie in the offender's past and present behaviour, character and circumstances. Thus, factors such as the offender's social deprivation, remorse, employment status, injuries suffered in the offence, or effects on the offender's family may be seen as relevant. It matters not that they are not directly or closely related to culpability or severity, are common to cases, are not major features of cases, or breach equity. Moreover, factors attain relevance not only in isolation, but also in interaction, and in personal, social and cultural contexts. Thus, a youth from a deprived background who has made good and supports a family, but who would lose his hard-won employment if imprisoned, may be seen as “worthy”Footnote 5 of a community sentence. Such a judgment is global, being in terms of the type of person before the court, and is made about a whole person. Indeed, it may confound factors relating to both culpability and rehabilitation, and indeed to mercy: the offender deserves less blame on account of the circumstances underlying his offending; he has the wherewithal to respond to rehabilitation addressing these factors; and losing his job may crush him. It shows an understanding of human weakness in the face of challenging circumstances, a preparedness to regard the offender as at heart a decent person, and a belief that an offender's future and feelings matter.
IV. The Significance of the Study
The positions of von Hirsch and Ashworth (Part II) and of Morris and Tonry (Part III) on the role of personal mitigation in sentencing are very different. Yet both claim their positions represent what is intuitively and commonsensically just. Without question, this contention is no mere theoretical squabble. The fact is, these contrasting senses of justice are mirrored in the contest between those for whom the essence of traditional judicial justice represents fair punishment and those who seek reform.
Traditional sentencing is to be found in many jurisdictions around the world. Thus, for example, it is orthodoxy in Australia, the United Kingdom and Canada, and was orthodoxy in the United States before the reform movements of the late 1970s onwards. The judicial instinct is to sentence offenders for committing offences (cf. impose sentences for offending). Thus, often in cases, sometimes though the offence is serious, judges will regard the circumstances of the offender as justifying significant mitigation of what would be the proper sentence on the grounds of the facts of the offence considered alone. Personal mitigation may involve the offender being seen to have less than full culpability for the offence, good rehabilitation prospects, and claims to mercy. Nevertheless, the seriousness of the case is always important, and sometimes will largely override the offender's claim to mitigation. Expressed in terms of principle, the severity of the punishment should be proportionate to the seriousness of the offence in question; but it also should be appropriate, having regard to the offender's personal mitigation. Thus, the sentence represents a balance; where it is struck will depend on all the circumstances of the case. For this purpose, a wide range of factors are considered individually and interactively, and in terms of the type of offender. For evidence of this, see legal analyses;Footnote 6 see also empirical studies.Footnote 7 This traditional approach broadly accords with Morris and Tonry's sense of fair punishment.
In contrast to this, von Hirsch's ideas are to be found in schemes representing attempts to reform traditional sentencing policy. One of the earliest, the state sentencing system developed by the Minnesota Sentencing Guidelines Commission, well illustrates this.Footnote 8 In this system, offence seriousness is a principal determinant of sentence, seriousness being defined in terms of offence categories and sub-categories and scaled by way of a limited number of seriousness levels. Associated with each level is a relatively narrow range of sentence deemed proportionate. Moreover, there is little toleration of personal mitigation. Indeed, a number of traditional social factors are expressly prohibited. These aside, personal mitigation is permitted, as long as the factor is substantial and compelling. In regard to this, there is a short non-exclusive list of factors closely related to seriousness. Nevertheless, personal mitigation will not normally warrant a sentence outside the guideline range. This approach to personal mitigation in the original guidelines has been somewhat relaxed in the light of appellate decisions, but the essence of the scheme remains very much intact.Footnote 9
The same propositions broadly apply (at least for the purposes of the present discussion) to the federal system developed by the United States Sentencing Commission.Footnote 10 In these guidelines offence seriousness is much more finely differentiated. Moreover, a moderate number of matters relating to personal mitigation are listed; but these are deemed to be not relevant or not ordinarily relevant, a provision strictly interpreted originally but less so now.Footnote 11
In England and Wales, too, the notion of proportionality is manifest in policy. There it appears as a framing principle in the Sentencing Guidelines Council's statement of the general principles to be followed by sentencing judges.Footnote 12 Moreover, the Council's guidance for particular categories of offence suggests its intention is to favour offence seriousness as strongly controlling. This is evident in the form these guidelines take. For this, typically, comparatively narrow ranges of sentence are set out for case circumstances defined by offence factors, and departures require an explanation. See, for example, the guidelines for robbery,Footnote 13 sexual offences,Footnote 14 causing death by driving,Footnote 15 and assault.Footnote 16 Moreover, in the Council's general guidance and, typically, in its specific guidance, there are far more listed aggravating matters than there are listed mitigating ones.Footnote 17
Clearly, what is under empirical enquiry in the present study is not just of academic interest but is important for policy. It is a question of whether the theory of proportionality underpinning reform is valid with respect to its claim to represent common-sense and intuitive justice. Although there are a limited number of studies of personal mitigation (reference will be made to two major ones subsequently) there appears to be no study directly addressing this matter.
V. The Empirical Study of the Public's Sense of Justice
The purpose of the empirical study is to test the validity of one aspect common to von Hirsch and Ashworth's and to Morris and Tonry's respective conceptions of criminal justice (Parts II. and III., respectively); namely, the claim by each that their view represents the general public's sense of justice.
A. Overview of the Study
For this purpose, the study relies on a large sample representing a cross section of the jurisdiction of Victoria, Australia. The general community seems to offer a valid standard against which to test these claims, for the following reason. If a particular view of just punishment does indeed represent common-sense and intuitive justice, rather than an idiosyncratic view, then it should encapsulate broadly the approach of the general community to the punishment of offenders.
In this study, four real cases involving six offenders were presented by the sentencing judges to members of the public in small groups; their task was individually to impose a sentence/sentences in one of the cases and then to discuss in their group what seemed to them to be the aggravating and mitigating factors and other relevant considerations. This took place at a cross section of workplaces around the State during working hours in groups of approximately fifteen. There was one case per group. The exercise was introduced to participants as trialling a process of judicial consultation. It was conducted between 2004 and 2006.
B. Rationale for the Method
To ensure a valid test of the two conceptions of justice in the light of the public's sense of fair punishment, the following four characteristics seem necessary. First, the cases involve serious offending by offenders with potentially strong claims to personal mitigation; the four offences cover a range of inherent seriousness as categories of offence; and the potential mitigation involve a wide range of personal circumstances. In this way, there is the opportunity of properly exploring the capacity of mitigation to be seen as warranting a less than proportionate sentence, the extent of any disproportionality, and the range of matters seen as relevant to this.
Secondly, the facts of each offence and the personal circumstances, behaviour and character – past and present – of each offender should be presented in detail; the offender should be portrayed as an actual person; and the participants encouraged to treat the offender as someone whose future lies in their hands. This is important, because it is hard to see personal mitigation emerging unless those doing the punishing feel a common humanity with the person whom they are punishing. This would appear to require that those participating in the study view the offender as real. For this, the description of the offender must be of a whole person, one who has a life, and not of a stick figure. This effect may be enhanced if the participants feel that their responses matter; hence, the exercise taking the form of a judge consulting about a real case. There are no apparent grounds for thinking that these conditions will of themselves elicit personal mitigation if it is not seen to be relevant; rather it is that these are the sine qua non for personal mitigation. Detail about the case is also required properly to assess the scope and nature of personal mitigation.
Thirdly, the participants' responses must be informed and made thoughtfully. For the former, the public need the opportunity to think about what matters are of potential relevance to justice (i.e. to develop their own senses of justice); they also need to have information about sentencing (e.g. sanctions available to the courts). To achieve thoughtfulness, the information about each case must be presented clearly, there be no pressure on the participants to make a quick response, and the exercise be seen as important.
Fourthly, the preceding two matters are about ensuring that the method is sensitive to the phenomenon of personal mitigation. However, these steps raise the possibility of two biases. One, of privileging the interests of the offender over the seriousness of the offence and the harm to the victim; the other, of favouring the judge's view of the case over what might be the contrary views of the participants. Thus, measures were taken to guard against these potential biases; the following examples illustrate this. In the account of sentencing policy, a similar emphasis was placed on matters pointing to severity and to leniency. And the judges' presentations of the facts of the cases were very detailed, including matters not obviously relevant, thus permitting participants to draw their own conclusions. Moreover, the judges did not argue for their approach to the case, but set out the matters of potential aggravation and mitigation which had exercised their minds, with no indication of where the balance was struck or how this related to the sentence. Finally, with respect to the discussion, participants were told that since the exercise represented consultation, what was wanted was their own well-thought-out views, not fawning agreement with what they thought might be the judge's perception of the case. Getting these countermeasures right was a matter of balance.
C. The MethodFootnote 18
1. Procedure
There were two sessions at each workplace. In the first session, the author gave a 70-minute talk. It included material on: the aims of sentencing; the circumstances under which each is considered relevant; the considerations relating to the various aims and their effect on sentence; and how these, often conflicting, are reconciled in the construction of the sentence; the sanctions available to the courts, what each may involve, and their (dollar) costs.
In the second session, the judge presented an edited version of his sentencing judgment. This covered a full account of the facts of the offence and circumstances of the offender (including summaries of any reports about the offender or statements from the victim), and the significance of this material for sentencing law and principles (e.g. the offender's culpability). Participants were also presented with the statutory maximum and crude data on current sentencing practice. This took around 45 to 55 minutes. The offender was given a name and spoken of in personal terms, and the circumstances of the offence and the offender presented as narrative and biography.
Participants then anonymously wrote down what sentence they thought was appropriate for the offender from a list of the available sanctions. They were reminded that this was a real case, but that they were not bound by sentencing law or practice. Finally, participants were invited to comment on and discuss the judge's sentence, including matters raised by him as potentially relevant or irrelevant.
2. The cases
What follows is a brief description of some of the circumstances of each of the cases. Each account is sufficient for sense to be made of the participants' comments.
• An armed robbery of $1,100 at a small gambling venue at night, involving minimal violence and an unloaded gun, and committed somewhat impulsively. The offender, 31, who had a very extensive record of dishonesty, but no violence, had used the money to support his wife and baby after losing money gambling. His upbringing had been chaotic. He wanted to reform, but his poor coping skills and temperament made life a challenge. As an informer, any term of imprisonment would have to be in protective custody.
• Multiple rapes at knifepoint, of a young woman, at night, by a neighbour who broke into her home. Before the offender left he apologised and asked her for a date. This young adult male had a drinking problem and was drunk at the time; he was of very low intellectual capacity, but was able to do menial work. Although he had priors for car theft, he was not regarded as antisocial. The victim suffered severe and continuing psychological trauma.
• Multiple stabbings of two young men by a young adult male carrying a knife, and the violent punching of a young woman by his 18-year-old girlfriend, who was also involved in one of the stabbings. Both had been drinking, the female heavily. One of the stabbings required quite major surgery. The first stabbing was initiated by a comment, which should have been inconsequential, by one of the victims; the second, as a result of the friends of the victim pursuing the offender. Both offenders were aggressive, but had troubled lives and were experiencing drinking and, the young woman especially, drug problems. Their work histories were poor. The male had served a term in a youth training centre, as a result of a previous stabbing; the female had priors but not for violence. The three victims received significant injuries – one victim, in particular, suffered very serious injuries (he recovered) – and experienced psychological trauma.
• The theft of about $1,000,000 worth of goods (cigarettes) from a company by two employees, over an extended period of time. The male, who was a senior manager, had organised the operation involving a number of offenders. He was now in debt, and his family shattered by these events. The woman, although a key player, acted under his direction and did not benefit greatly from the crime. She gave substantial assistance to the police investigation. Her husband left the family due to her offending, and she was the sole carer of two chronically sick and very dependent children. Both offenders were otherwise of good character.
3. The sample
There were 471 participants, comprising approximately an equal number for each case. The participants were employees at 32 workplaces, eight for each case: four were from across suburban Melbourne and four from around country Victoria; of each four, two represented primarily while-collar and two primarily manual work; and of each of these two, one was commercial (e.g. factory) and one non-commercial (e.g. library). There was substantial variation in the nature of the work within each of these occupational categories.
For each case there was a good representation with respect to gender, age (20–60), education (graduate/non-graduate), and occupational status (high, medium, low) within the participant's own workplace. However, overall there were biases against (i.e. fewer of) those over 50 (quite substantial), non graduates (slight) and the high status (quite substantial). Management at the chosen workplaces were asked to select participants randomly, within constraints formulated to achieve this representation. Some managers observed this, but there was a degree of targeted- and self-selection at some workplaces.
At three workplaces participants were first asked to make a general rating of judicial sentencing. The mean percentages for “much too soft”, “too soft”, and “slightly too soft” were, respectively, 11, 30 and 47. (For these participants, there was an overall strong bias to white-collar, country, commercial, female and those under 40.) This is evidence of the sample representing quite well the major demographic groups and the public's general disenchantment with judicial leniency.
VI. Comparing the Public's and the Academics' Sense of Justice
The question is whether, in the participants' comments about the six offenders, what was considered mitigating by them is more similar in content, scope, nature, and importance to the sense of justice of Morris and Tonry or to that of von Hirsch and Ashworth (Parts III. and II., respectively). This is done by first setting out and making observations about the factors found by the participants to be mitigating, as well as the sentences they imposed on the offenders. Then these findings are examined to see whether they represent by way of mitigation what would be expected by these respective theorists.
A. The Participants
In fact, the participants found numerous factors in the four cases to be mitigating, and all could be placed in one of the categories of culpability, rehabilitation or mercy. Factors illustrating each of these are now presented, together with an assessment of their significance.Footnote 19 Following this, the participants' sentences for the six offenders are analysed for evidence of disproportionality.
1. Culpability
About the factors cited by the participants as downgrading the seriousness of the cases by way of reduced personal culpability, several things bear comment.
First, the factors found relevant to the mitigation of culpability lay in all aspects of the cases, as defined by the following four categories, namely: the circumstances of the offender's commission of the offence; factors underlying the offending (relevant to the offender's circumstances); the offender's response to the offence; and, the offender's character. Thus, for the rape, the offender should be treated as less than fully culpable because, it was said, he: used the knife as a threat, not for harm; was of low intellectual capacity; expressed remorse to the victim immediately after the offence; and had no history of violence.Footnote 20
Secondly, most of the factors seen by the participants as mitigating culpability for the rapist appeared to represent individual aspects of the case. From this perspective, mitigation of culpability is by way of an aggregation of individual factors. However, one of the factors for the rape suggested that there may be more to the public's understanding of culpability. It is that of “poor silly man”. This points to mitigation of culpability being thought of in terms of a type of offender, involving a summary of a global take on factors. Some of the factors implicit in this label more or less directly underlie culpability (e.g. low intelligence); but, significantly, others in themselves are no more than indirectly related to the offending (e.g. the apparently sincere apology at the time of the offence) or, indeed, apparently unrelated (e.g. low self-esteem). Thus, the use of factors in mitigation is both interactive and contextual.
This analysis was seen to hold for the other offenders. For example, for the female in the assault, all four categories were found to be applicable. Thus, there was provocation by one of the victims (circumstances of the offending), she suffered from a personality disorder due to her troubled background (factors underlying the offending), there was remorse shown (response to the offence), and the offender had not been involved in previous violence (character). Moreover, a global take on culpability also was found for the armed robber – “loser” – and for the female in the theft – “foolish” (cf. “criminal”). Again, the determinants of these global assessments could be found in the individual factors considered by the participants to be relevant in the circumstances of each of these offenders.
Thirdly, mitigation involved more than patent objective facts; rather, the factors found to be relevant represent interpretations, as the following examples from the circumstances of the offences show. Although some of the interpretations were straightforward and little more than objective case facts (e.g. the male in the assault was drunk), not uncommonly factors relied on interpretations involving multiple objective case facts, even going beyond the facts. Some were quite complicated. These interpretations appear to involve the participants relying on their life experiences, even projecting themselves into the events. This is illustrated by the observations that the offending by the male in the theft was initially opportunistic, and that the female in the theft would have had difficulty in extricating herself from the tightly knit criminal group. Patently, the interpretation of case facts relevant to mitigation is interactive and contextual, relying on the offending but also on the offender's past and post-offence behaviour and past and present character and circumstances.
Fourthly, many of the factors put forward by the participants can be seen as evidence of their being prepared to make allowance for human weakness and challenging personal circumstances. This was more patent on the several occasions when a justification accompanied the citing of a factor. For the female in the assault, one of the factors said to underlie her offending illustrated this. Her drinking and drug-taking were not regarded as her fault; in view of her poor education and troubled upbringing, she would not have given consideration, when first taking drugs as a young teenager, to the dangers. This participant seemed to understand how a half-decent person could do something bad. In this way, offenders were treated as less morally reprehensible, and so less culpable, than was indicated by the facts of the offence considered alone.
Sometimes the threshold for invoking mitigation was low and the factors cited somewhat subtle. Thus, for example, even though the male in the theft was financially comfortable, the factor of his using the money to compensate his family for his absences on business was cited as mitigating (factors underlying the offending). And in regard to the male in the assault, who was carrying a knife at the time of the offence: this was seen as capable of justification on the grounds of his seeing this as necessary for self-protection (circumstances of the offence).
2. Rehabilitation
Rehabilitation as an aspect of offender mitigation was patent in the participants' responses for all six offenders. Participants looked for evidence of favourable rehabilitation prospects. The express assessments of the prospects of three of the offenders – rapist, and the male and female in the theft – being good, and of the rehabilitation of a fourth – the armed robber – being possible, demonstrated this. As evidence of favourable rehabilitation, participants again relied on one or more commonsensical factors in an offender's past and post-offence behaviour and past and present character and circumstances. These included no prior offending, youth, remorse, worthy conduct, current and past employment status, and having a supportive family.
Four of the offenders – those involved in the rape, the armed robbery and the assault – seemed vulnerable to serious offending as a result of personality problems. Participants discerned this and thought they required treatment. In regard to this, participants showed concern over the potential deleterious effects of imprisonment on an offender's rehabilitation prospects. Some of these related more or less directly to the offending. Thus, for each of the four offenders seen as requiring treatment, at least one of the factors cited (e.g. imprisonment for the rapist being criminogenic due to his low intelligence) underlay their offending, and also had been cited as reducing their culpability. And for two of these offenders – the armed robber and the male involved in the assault – the factors cited in this respect also directly related to the treatment proposed – life skills and anger management, respectively. Surely this is evidence of participants dealing with each case, not as a set of compartmentalised factors, but in terms of a real person, constructively and with humanity?
Other factors cited as deleterious, however, were more general, such as the observation, for the young woman involved in the assault, that custody would ruin her chances of employment.
3. Mercy
Mercy was for the participants a part of mitigation. It took the form of compassion for the offender and for the offender's family. For two of the offenders, the grounds could be regarded as especial. These were picked up by the participants. For the female in the theft, there was the extreme dependence of her chronically sick children on her as their sole carer. For the armed robber, there was his having to spend his time in prison under the harshness of protective custody. However, in addition to this type of factor, general grounds were cited. These related to the collateral effects of conviction/imprisonment and to the severity of the sanction. Concern over the armed robber's child growing up without a father's support, illustrates the former; an example of the latter is the mature age of the male in the theft. In this way, the participants appeared to identify with the offenders and their families as fellow human beings and to feel their penal pain, in some instances demonstrating a high level of sensibility (e.g. the shame felt by the family of the male in the theft).
4. Proportionality and personal mitigation
The participants' comments clearly showed that matters relevant to personal mitigation exercised their minds when they thought about what was a fair punishment. Important also are the sentences imposed by the participants on the six offenders; from these data it is possible to infer something about the relative importance of proportionality and personal mitigation to them.
The four offences as (sub-)categories of offence were all very serious. The rape would generally be regarded as by far the most serious offence, followed by the assault. Consider these two first. The rape was more serious as an offence (note, both were aggravated by multiple counts) but against this, not only did the rapist have no priors for violence while the male in the assault did, the participants appeared to find much other personal mitigation for the rapist, but less mitigation for the male in the assault. Despite these counteracting influences, the median non-parole period imposed by the participants on the rapist was five years, significantly above their (just over) three years for the male in the assault. This may be taken to indicate the pre-eminence of proportionality. Yet, notwithstanding these two offences being very serious, particularly the rape, these sentences are comparatively no more than moderately severe. Apparently, personal mitigation had a substantial influence; in fact, these sentences seem disproportionately lenient.
A similar comparison may be made for the armed robber and the male thief. Armed robbery as a category of offence is far more serious than is theft. Nevertheless, the present armed robbery as an armed robbery was no more than moderately serious, whereas the theft as a theft was very serious (note, the seriousness of both was very much aggravated: the former by the offender's extensive priors and the latter by multiple counts and the extended period of the offending). With respect to personal mitigation, much more was found for the armed robber. Thus, perhaps the similarity of the sentences for the armed robber and for the thief (both non-parole periods of two years) reflects the comparative importance to the participants of these two categories of offence together with the more substantial personal mitigation of the armed robber. Moreover, the sentence of two years' imprisonment for an offence of armed robbery – a serious offence category – may well indicate that personal mitigation may be seen as warranting a disproportionately lenient sentence but not so substantial as to negate the seriousness of the offence.
Clearly, the preceding reasoning is suggestive rather than conclusive. In view of this, a comparison of the theft and the armed robbery with the assault would not appear to add anything; indeed, it would be rather tenuous. However, the participants' sentences for the two female offenders do bear comment. The sentence of a moderate-to-short period of youth training (a custodial sentence) for the female in the assault again probably reflects a balance by the participants between what they saw as the strongly compelling but competing demands of offence seriousness and of personal mitigation. (Note, this young woman's offending being secondary was not as serious as that of the young man's, and her claims to personal mitigation were seen as being much greater.) In regard to the female in the theft, her offending also being secondary was not as serious as that of the male's, but was serious nonetheless. The fact of the participants imposing a non-custodial sentence on her demonstrates how personal mitigation, when compelling, may overwhelm offence seriousness.
B. Von Hirsch and Ashworth
The preceding analysis of the participants' responses is assumed to reveal the public's intuitive sense of justice (Part VI.A.). Whether von Hirsch and Ashworth's view of fair punishment (Part II.) corresponds to this is the present question. It is addressed by identifying which of the factors considered by the participants to be relevant to sentence accords with these theorists' strict take on proportionality. This is necessarily a somewhat imprecise exercise; for while von Hirsch and Ashworth do give express opinions on some matters, for other matters it is necessary to rely on their general principles.
In respect of culpability, von Hirsch and Ashworth would find few facts to be relevant and compelling. Almost certainly, the rapist's very low intelligence and the youth of the female in the assault would be seen as warranting a somewhat lesser sentence; but probably not beyond or much beyond the proportionate range, since he did not qualify as disabled, and she was (just) a young adult. The participants' regarding the rapist and the male in the assault as young at 24 years and 21 years, certainly for the former and probably for the latter, would be deemed to be too subtle and remote. So too would the personality disorders, having regard to the clinical evidence, of the armed robber and of the male in the assault. Similarly, the fact of the female in the assault being under the influence of her co-offender, and the motivation of the female in the theft being time off work for the sake of her sick children, would surely be seen as falling well short of the partial excuses of duress and necessity. And the victim's looking at and making a weakly derogatory reply to the female in the assault would not nearly qualify as provocation. Finally, the drunkenness of three of the offenders (rape and assault), in the absence of a clinical problem, probably would not carry much weight, not least because it would be regarded as too common. But the fact of the female in the assault being somewhat under the influence of drugs might well count for something in the light of her deep-seated drug problem.
Next, consider priors. Both offenders in the theft were first offenders. However, the fact of their offending involving multiple counts over a very extended period might be seen as negating much of the otherwise warranted mitigation. For the rapist and the female in the assault, there would be mitigation on this ground, but probably it would not be substantial. They had no prior violence, but did have records, albeit not extensive ones; and their offending, although representing one incident, involved multiple violent counts. And while the armed robber too had not been previously violent, his record of dishonesty and other offences was extreme.
Then comes equity mitigation. There are two potentially relevant factors here in the circumstances of the offenders. One is the time delay in the prosecution process for the offenders in the theft. In this time the male in the theft set up his own business; this being the basis of the observation by the participants that he was now leading a productive life. However, hard work had never been a problem for him and, in any case, the delay was partly due to his initial lack of cooperation. This is hardly a basis for his being seen as a “new man”, and thus, according to proportionality, requiring less censure. The other is the rapist's apology to the victim before he left her home. Whether this acknowledgement of wrong to the victim would count for much under proportionality is not clear.
The fourth potential source of personal mitigation is in the impact of the sanction. In regard to this, none of the offenders had strong claims on the grounds of advanced age, ill health, or disability. The mature age of the male in the theft was cited by participants in this respect, but he was able-bodied and only of middle age. In any case, this factor is not normally seen as warranting a disproportionately lenient sentence. Proportionality might make allowance for the possible harshness of the armed robber's sentence, involving as it did protective custody, but this might well be seen as within the band of normal variation in which imprisonment may be served, and thus as no more than of minor significance. Participants were concerned about the crushing effects of custody on the female in the assault due to her youth and the fact of her being particularly psychologically vulnerable. Proportionality, it appears, would recognise the former but perhaps not the latter.
The preceding discussion dealt with matters seen by the participants as relevant to personal mitigation, and having potential or plausible relevance to proportionate punishment. Few of these factors would have got across proportionality's high threshold, and when they did – the rapist's low intelligence and comparatively good record, and the youth, minor record and drug problem of the female in the assault – their combined effect for each of these offenders would not appear to warrant, according to strict proportionality, more than modest disproportionality. Nevertheless, this young woman's youth probably would be regarded as justifying her term of custody being served in an institution for young adults.
There also are the factors, mitigating according to the participants, but expressly forbidden by proportionality. For culpability, these would include the troubled backgrounds of the two offenders in the assault, and the shambolic background of the armed robber. These matters would be seen as too common and too remote from the offending. In regard to rehabilitation, most of the participants' assessments about the offenders' prospects relied on factors unrelated to seriousness – remorse, shame, good employment history and family stability. In any case, the participants' wanting for four of the offenders a programme of treatment, and for these four and one other a lesser or no term of imprisonment, due to the potentially deleterious effects on their rehabilitation prospects, would count for naught. This is because rehabilitation is unrelated to the censuring function. Finally, the participants raised matters related to mercy. Several concerned the collateral effects of conviction/imprisonment on the offender – such as the ruined life of the male in the theft – but proportionality would exclude these on the grounds of their being foreseeable. The majority of the participants' concerns involved the offenders' families. These too, presumably, would be rejected because they are unrelated to the offender's experience of the sanction's severity.
Finally, let it be observed, the participants found factors across the six offenders to be relevant to their culpability, which apparently are neither of potential relevance to nor expressly disallowed under proportionality theory. Examples include: the motive for the rape being sex not power (rapist); the remote nature of the offending (female, theft); the company's poor audit controls (male, theft); the lack of life skills (armed robber); the carrying of a knife for self-protection (male, assault); and the seeking of treatment prior to offending and (general) remorse (female, assault). These all would appear to be unrelated to or remote from what proportionality regards as the seriousness of the offending, or to represent a too-sensitive reading of the facts of the case.
Also relevant to this general point is the evidence of the participants' determining punishment partly in terms of the type of offender. Recall, this was their approach for at least three of the offenders (rapist, armed robber, and female in the theft). However, proportionality does not do type, rather it reasons in terms of individual factors. This is because judgments about type involve multiple factors and many of these factors will not be closely related or will even be unrelated to the offending, as illustrated by the participants' comments (e.g. the rapist's low self-esteem).Footnote 21
Ashworth's independent stance on proportionality and mitigation (Part II.D) generally makes little difference to the preceding conclusions. Thus, factors like employment status and certain family circumstances (e.g. a supportive family), used by the participants with regard to rehabilitation and mercy but banned by proportionality on account of their remoteness from the offending, are also disallowed by Ashworth on the grounds of social equity. Then there is the participants' concern about solitary confinement. It was regarded as carrying probably no more than modest weight under proportionality; Ashworth, however, expressly disallows it. Nevertheless, there are two potential exceptions. Ashworth is prepared to entertain exceptional social disadvantage, when it is relevant to the offending, and exceptional family circumstances in relation to imprisonment, as perhaps warranting significant mitigation. Thus, on these matters he might well agree with the participants in regard to the armed robber and the female in the theft.
C. Morris and Tonry
The question is, do Morris and Tonry's views on fair punishment (Part III) correspond broadly to the public's sense of justice (Part VI.A)? In fact, there is no need here for a detailed analysis, since Morris and Tonry speak only in general terms. And at this level, the similarity of the two approaches is strong, as will be evident in the following concluding summary.
VII. Conclusions and Implications
This present exercise has attempted to investigate whether the public's sense of justice is better reflected in the theory of von Hirsch and AshworthFootnote 22 or in the views of Morris and Tonry.Footnote 23 This section first deals with the answer, which is patent and thus may be dealt with briefly. The bulk of what follows covers related matters, including the implications for policy reform.
The public do not view personal mitigation through a proportionalist's eyes. There is more to their sense of justice than strict proportionality. To the public, numerous case factors are of potential relevance, and it is not uncommon for these to be seen as warranting a substantial reduction in the sentence otherwise proportionate to the (sub-)category of the offending. Contrary to the thinking of von Hirsch and Ashworth, fair punishment involves more than culpability and severity, and censure. The offender's rehabilitation prospects may be an important consideration. So too, in addition to the direct impact of the severity of the sanction on the offender, the collateral consequences of the sentence for not only the offender but also the offender's family may be important matters. In regard to these considerations, the participants were pragmatic and showed a high level of sensibility. Thus, a factor does not have to pass an equity test or be exceptional. The public's view of culpability is much broader too. The offending is relevant, but they also rely on factors which proportionality would see as indirectly related or even unrelated to the seriousness of the offending. Nor does it matter to the public if these are common features of cases. And, unlike proportionality, the public's sense of justice does not treat most offenders as readily capable of responding to censure or the threat of censure; the public are highly sensible to human weakness. Moreover, to the public, less blameworthiness is not only by way of a calculus based on patent individual factors. They are prepared to look hard at the offender in the broad circumstances of the case, and see a type who is not as bad as the baldly described offence would indicate, and on this basis see the case as warranting a substantially lesser sentence, one outside, indeed well outside the proportionate range. Put simply, with respect to personal mitigation, von Hirsch and Ashworth see a less serious offence, the public see a less serious offender, one whose decency, prospects and feelings matter. This notwithstanding, offence seriousness is important to the public; the difference is that it is far from all important. Reconciliation between the public's sense of justice and proportionality theory is patently beyond fine-tuning and compromise. Von Hirsch and Ashworth acknowledge there will implications to be pursued and justifications to be explored in their theory, certainly; but reconciliation would require the building of major extensions.
Nevertheless, the public, in differing from the estimable academics von Hirsch and Ashworth, need not be intellectually or morally ashamed. In fact they keep very good company. For, in general terms at least, as the public differ from von Hirsch and Ashworth, so too do Morris and Tonry. This is true in respect of the content, scope, importance and nature of personal mitigation. This last aspect warrants elaboration. Morris and Tonry, like the public but unlike von Hirsch and Ashworth, think of offenders for the purpose of sentencing as whole people. Judgments about their culpability, need for rehabilitation and sensibility to punishment are global and based on the offender's personality, behaviour, circumstances and life history. Offenders are thought of as fellow human beings whose humanity needs to be understood and to be seen as mattering. This, notwithstanding, the offending is taken seriously. For Morris and Tonry, proportionality has a human face; it does too for the public.
In regard to these conclusions, two points must be made clear. The first is that these conclusions are based on the balance of the public's sense of justice. Some of the comments of the participants were at least superficially consistent with the thinking of von Hirsch and Ashworth. Thus, some participants, in arguing against certain personal factors, used the language of proportionality (e.g. the observation that mitigating the culpability of the rapist risks undervaluing the seriousness of rape as an offence). Other comments, at least at first blush, had an equity overtone (e.g. the non-custodial sanction for the woman in the theft was said to be unfair to those without extremely dependent children). Nevertheless, there was much evidence, in the participants' contrary comments, of the sentiment of law and order. Thus, it was observed: there should not be lesser terms of imprisonment in order to facilitate an offender's rehabilitation; it is up to offenders not to become embittered, but to put their life in order. Patently, this is not a proportionality argument.Footnote 24 Consistent with the latter type of sentiment, for each of the six offenders a significant percentage of the participants imposed harsh sentences.Footnote 25 Nevertheless, the evidence for the balance of the public's sense of justice lying with Morris and Tonry and against von Hirsch and Ashworth is the manifest disproportionate leniency in the median sentences for the offenders and the overall moderation of these medians (Part VI.A.4.).
The second point concerns the fact of there being no measure of the percentage of participants who endorsed a particular factor. For this reason, the data do not permit conclusions about individual factors. Nonetheless, the data do speak to the general content, scope and nature of the participants' understanding of personal mitigation. Moreover, the just-mentioned sentencing data attest to the importance of personal mitigation. This is sufficient for present purposes.
Thus, von Hirsch and Ashworth's claim that their theory of fair punishment represents commonsensical and intuitive justice was found to be false; at least on the basis of the apparently reasonable assumption of the balance of opinion in a cross section of the community defining common sense and intuition. This criterion appeared to be put forward by them as at least one basis for the legitimacy of their theory. If this is right, then on this test of validity of their own making, their theory fails. The problem is not, of course, that proportionality is not a part of the public's sense of justice. Rather, the problem lies in von Hirsch and Ashworth's narrow focus on seriousness, and their assessing punishment in terms largely exclusive of the offender. By way of contrast, Morris and Tonry's approach to thinking about what is fair punishment was, consistent with their claim, found to represent commonsensical and intuitive justice. On this measure it looks good, and thus offers a better foundation for a system of punishment.
The present use of the public's sense of justice as a test of validity makes sense in the light of these respective thinkers' reliance on it. But does this criterion matter, this theoretical squabble aside? Can it reasonably lay claim to a fundamental test of a system of punishment's validity? According to Robinson it can.Footnote 26 The public's sense of justice, identified by research, is what he calls empirical justice. And to Robinson there are good grounds why the community's shared intuitions about what is just should underpin (cf. necessarily be identical with) society's formal system of punishment. For only in this way, he says, will the system have the moral credibility so essential to its crime-control function.
At this point, a consideration of the implications of the present study's findings for sentencing reform is apposite. From Robinson's perspective, it matters much whether von Hirsch and Ashworth's understanding of fair punishment, as it is applied by reformers, represents theoretical idiosyncrasy or something more general. In this light, the sentencing reforms in the United States and in England and Wales must be seen as troubling. The description of these sentencing reforms (Part IV.) showed that each in its own way broadly reflects the main elements of von Hirsch and Ashworth's thinking. Moreover, these reforms to a greater or lesser extent have replaced traditional judicial sentencing policy, which is consistent with Morris and Tonry's views. Moreover, traditional judicial sentencing policy in its essence well represents how the public think about what is fair punishment for an individual.Footnote 27 Is it not time, then, to pause and reflect upon what these reforms are doing to sentencing? Punishment is at risk of becoming something different – there will be less humanity and it will be impersonal – and, in these respects, punishment risks becoming something less.Footnote 28
In respect of this conclusion, two caveats need to be noted briefly, each introducing a complication. Robinson's justification for the criterion of the public's sense of justice being important requires a significant degree of consensus in the public's views. How much is an open question. That the present study, as noted above, demonstrated minority dissensus in the public's sense of justice is potentially problematic. Clearly, this matter cannot be taken any further here, but it should be kept in mind nonetheless. Secondly, the validity criterion of compatibility with public opinion requires that the public be aware of their sense of justice and have an accurate picture of sentencing practice. But the public do not have a proper understanding of these things. This is because this understanding is based on the media's reporting of crime and sentencing, in which the accounts of the personal details of offenders are generally given scant attention or, if the offender is referred to, it is often as a devious, thuggish predator.Footnote 29 Thus, in the public consciousness there is little appreciation of the characters and backgrounds of typical offenders or of the potential effect of those various circumstances on what they regard as fair punishment. What follows from this is that reported sentences for the more serious cases will often appear unduly lenient to them.Footnote 30 Moreover, the public will mistakenly regard current reforms of traditional sentencing policy, with their greater severity by way of less personal mitigation, as better reflecting their sense of justice.
What is lacking to this point in this discussion are observations about the validity and generality of the preceding conclusions. In regard to the former, the question arises as to whether the finding of similarity between the public's sense of justice and Morris and Tonry's view is an artefact of the method; perhaps the findings no more than represent the judges' views on mitigation and reflect a method favouring the offender. It appears not. In fact, transcripts of the judges' sentences show much was made of matters in aggravation, particularly the seriousness of the offending (including the impact on the victim) and the need for deterrence. Moreover, although potentially mitigating factors were mentioned for all offenders, only for the female in the theft was some stress put on mitigation (namely, the fact of her extremely dependent children). More generally, active steps were taken to address this potential problem, and they seem to have been effective (Part V.B.). As evidence of this, the participants' discussions were often lively, and their comments were anything but a mere reflection of the judges' views about the potential relevance to sentence of the various case factors. The disparity included mitigating factors and reasons for factors as mitigating not mentioned by the judges. Illustrating this, it was said, for the theft, that the company was at fault because of its lack of auditing controls. As well, there were the instances of participants making much of matters which the judges had little more than mentioned in passing or had not treated as very significant. For example, in the rape, the judge saw the offender's low intelligence as no more than of some importance, whereas the participants made much of it. And then there were the participants' arguments raised against factors put forward by judges as potentially favouring lenience or severity.
Generality too is an issue here. The first aspect of this is intra-jurisdictional. The present finding of personal mitigation as important is particularly significant because the cases covered a range of offences and all involved serious offending. Clearly, though, this study needs to be repeated with these and other types of offence and offender. The second aspect is extra-jurisdictional. Victoria is punitively moderate, at least in comparison with the harshness of England and Wales, and the swingeing United States.Footnote 31 Since the effect of personal mitigation is to moderate harshness, perhaps personal mitigation is not a part, or so much a part, of the public's sense of justice in these two countries. So what does research tell us about the generality of the present conclusions?
Unfortunately, there is not much research of direct relevance to this question. An important exception to this is the recent study of the public's attitudes towards aggravation and mitigation, conducted by Hough, Roberts and Jacobson in England and Wales.Footnote 32 The sample was large and cross-sectional, and the data were both qualitative (focus groups) and quantitative (face-to-face survey). The offending involved violence and dishonesty. In general terms, the conclusions are the same as those of the present study. The seriousness of the offence was important, and generally more important than personal mitigation, but the effects of personal mitigation on sentence sometimes appeared to be significant. A range of factors were potentially mitigating, and the justifications for these related to culpability, rehabilitation and mercy. And among these factors were considerations, like remorse, outside the scope of strict proportionality. In the public's judgments about these matters, the significance of a factor often was found in the circumstances of the offending and the offender, and including the offender's life history. Some of the public's qualitative judgments appeared individualised, and their basis contextual and thus interactive. There is one potentially significant difference between the conclusions of the two studies, however. It relates to the nature of personal mitigation. The present study found personal mitigation to constitute not only the aggregation of individual factors but also a global assessment based on individual factors (e.g. with respect to culpability, the offender was foolish not criminal). This was not apparent in Hough et al.'s conclusions.
Two differences between the present study and Hough et al.'s study in regard to the method are worth noting. In the latter study, participants imposed sentences for cases and made judgments about factors by considering their relevance alone and in the context of cases. The core case descriptions were quite brief and were elaborated by the addition of single or combinations of several factors taken from a list. Some of their participants said they would have found more detail about the cases helpful when making their judgments. And other participants nonetheless might have used this additional information had it been available. This suggests that the very detailed and extensive case descriptions in the present study – these descriptions taking the form of narrative and biography – were worthwhile. In fact, this may explain why the global character of personal mitigation was patent here, and why the number and range of factors identified in the present study were greater. Nevertheless, this difference in method did not appear generally problematic in view of the general similarity of the findings. However, the offending involved in Hough et al.'s study, although serious, was far less serious than the offending in the present study. The former study's offending varied around and just above the custody threshold; in the latter it was seen as warranting not-insubstantial terms of imprisonment. This difference in the level of detail in the case descriptions may be critical to whether personal mitigation is found when the offending is very serious. Against this, Hough et al.'s method permitted conclusions to be drawn about the significance of individual factors. There is a second difference of potential significance between the two methods. In the present study, an attempt was made to present the offender as real and the participants' responses as mattering: the cases were real and the study was presented as judicial consultation. Hough et al. did not attempt this. Again, while it did not appear to make a difference, it may be critical in whether personal mitigation is found to be important when the offences are more serious.
Another exception to the general dearth of major research on personal mitigation is the National Punishment Survey, conducted in the United States by Jacoby and Cullen.Footnote 33 In this quantitative study, participants were interviewed by telephone, presented with a number of brief case descriptions covering a wide range of offence seriousness, and asked to impose punishments. The authors' conclusion: the public wanted the punishment to fit the crime; matters relevant to the offender, youth side, accounted for little. Nevertheless, there must be grave doubt about the validity of this conclusion. For example, offender factors were few and presented with virtually no context. Moreover, the measure of importance – percentage variance – identifies only average effects; this will miss much (e.g. important factors appearing only rarely in case circumstances).Footnote 34 Of course, what would have been found about the importance of personal mitigation, had the method been appropriate to the phenomenon, must be treated as an open question.
Clearly, more research is required on the importance of personal mitigation to the public. In regard to this, the following observation should be borne in mind. Research perhaps more suited to the phenomenon of mitigation nonetheless may find mitigation, generally or for serious offending, to be of little importance to the public in one of these other jurisdictions. Nevertheless, this would not necessarily warrant a conclusion of similarity in the thinking about fair punishment between von Hirsch and Ashworth and the public. The basis of the finding would require careful scrutiny. Indeed, any observed similarity may be no more than apparent; it may in fact reflect law-and-order sentiments on the public's part, not proportionate thinking. This possibility will arise thus: although there are similarities between proportionality and law-and-order sentiments, there are important differences. The former does not entail high levels of sentence. Indeed, proportionate censuring favours moderate punishment. But tough sentences are a hallmark of law-and-order. Moreover, to the extent that both are unenthusiastic about personal mitigation, they do so for different reasons. The former does so on the grounds of theoretical purity; the latter for the purposes of protection and vengeance, in the light of a disregard for or antipathy towards the offender. These differences may easily be missed in the light of the similarities between the two.Footnote 35 But it is important to be aware of them when drawing conclusions about the nature of the public's sense of justice as revealed by empirical studies. This requires data on the public's preferred levels of sentence as well as their reasons for sentence.
VIII. In Conclusion
The theories of von Hirsch and Ashworth and of Morris and Tonry represent two major approaches to understanding what is fair punishment for individual offenders. Both agree on the importance of offence seriousness as a determinant of sentence. Yet, in regard to the importance of personal mitigation and what constitutes it, they are diametrically opposed: the former being parsimonious and narrow; the latter generous and broad. Despite this difference, both claim legitimacy for their respective approaches on the grounds of their positions reflecting the public's sense of justice. The present study tested these claims, and the outcome was patent: this aspect of theory validity lies with Morris and Tonry.
The present analysis has policy significance. Morris and Tonry's approach broadly represents traditional judicial sentencing policy. However, it is von Hirsch and Ashworth's theory which has struck a chord with reforming academics and policy makers. By this means, von Hirsch and Ashworth's notion of proportionality can genuinely lay claim to have influenced strongly the structure and content of new sentencing systems in the United States, and can be seen to be insinuating itself into current sentencing policy reforms in England and Wales. But there is a problem. This reform may appeal to the decision-makers' intellects, but it is different from the people's instincts. This is troubling if, as has been reasonably claimed by Robinson, a sentencing system should reflect the people's sense of justice in order to have their confidence. In regard to this, two questions arise. First, why has strict proportionality been so successful in gaining practical application? Proportionality has, of course, adherents, such as BagaricFootnote 36 and Easton.Footnote 37 And there are those who, while not avowed proportionalists, appear to find little in favour of certain offender factors traditionally treated as potentially mitigating (e.g. meritorious conductFootnote 38). Nevertheless, there is significant academic opposition to elements of strict proportionality. Thus, Hudson has argued for social and economic disadvantage – broadly interpreted – and related factors as mitigating culpability.Footnote 39 Also, in respect of culpability, Dingwall and Koffman have made a case for drunkenness,Footnote 40 and Hirst for human error in driving causing death,Footnote 41 and so on. Now to the second question: why has the disparity between what has been found here to be the public's sense of justice and what proportionality regards as fair punishment not appeared as a problem? What follows is relevant to both questions. Conventional wisdom has it that law-and-order sentiments represent the public's sense of justice. This is because in the media's presenting of sentencing decisions, the public do not see personal mitigation. Thus, they are not troubled by its not being taken into account, and in fact see sentencing as unduly lenient. Moreover, the sentiments of law and order and the tenets of strict proportionality converge somewhat with respect to their parsimonious treatment of personal mitigation. Thus, governments actively promote and happily accede to punishment systems based on strict offence seriousness. And academics and others holding contrary views will find it hard to get traction for their arguments. Proportionality, in finding support in this obfuscation, may have the security of the mob, but in this it should take no comfort.
