Abstract
In past publications, we have demonstrated harm’s centrality to crime, by presenting evidence on their relationship that dates to ancient legal systems and spans centuries via Beccaria, Smith, and Mill. In this article, we argue that binding criminology and criminal policy more firmly and explicitly to harm would benefit criminology and criminal policy, not just for the sake of criminal justice, but also for that of justice writ large. We show how doing so can, with support from a newly developed harm assessment framework, promote justice and we point to evidence of progress and feasibility. Our understanding of justice draws from Sen’s work, which conceives of justice largely as a matter of social realizations, capabilities, and freedom. We recommend expanding criminology’s mission to embrace the study of harm and targeting harm reduction in criminal policy to solidify their connection to justice.
Keywords
Introduction
Twenty years ago, Hillyard, Tombs, and their colleagues (Hillyard and Tombs, 2004; Hillyard et al., 2004) drew from critical criminology to propose replacing the notion of crime with that of social harm and directing social policy toward harm reduction. In subsequent publications, they suggested a new discipline, called zemiology, dedicated to the study of “social harm,” which, for some, was meant to stand apart from criminology (see, for example, Hillyard and Tombs, 2007; Hillyard et al., 2004; Yar, 2012 and Canning and Tombs, 2021). Other scholars of social harm have suggested a less radical path. Kotzé (2018: 97), for example, has criticized “the imaginary dualism between the study of crime and harm” and pleaded for the development of an integrated approach with a “more holistic understanding of both criminalized and un-criminalized harms.”
In our work, we have taken the less radical path, first, by demonstrating harm’s centuries-long relevance to crime and criminal policy, second, by developing a systematic and empirical framework to assess the harms of crime, and, third, by suggesting its broader applicability (e.g. Greenfield and Paoli, 2022; Paoli and Greenfield, 2018 and Greenfield and Paoli, 2013). As we have shown, evidence of harm’s centrality appears in ancient legal systems (e.g. Berman, 1983; Eser, 1966) and tracks intermittently—with varying acknowledgment—to the present via Beccaria (1995 [1872, 1764]), Smith, (1976 [1790]), and Mill (1978 [1859]) through writings on crime, legal theory, and security, and in criminal policy and law. Indeed, contemporary societies have labeled many offenses criminal precisely because of their harmfulness or because their criminalization was meant to prevent possible harm (e.g. Duff and Marshall 2014).
In this article, we posit that binding criminology and criminal policy more firmly and explicitly to harm—both conceptually and operationally—would benefit not just criminal justice, but also justice writ large. 1 Our understanding of justice draws on Sen’s work (2009), which conceives of justice largely as a matter of social realizations, capabilities, and freedom, but with due regard for the forces operating around them. We show how orienting criminology and criminal policy toward harm can, with support from the harm assessment framework, promote justice through contributions to decisions on criminality, strategic and tactical priorities, sentencing, and remediation and to evaluations of policies and interventions. We also point to evidence of progress and feasibility in applications of our framework and in recent initiatives on harm in policy and policing agencies in the United Kingdom (e.g. UK Office of National Statistics (ONS), 2022) and elsewhere. Finally, to bind criminology and criminal policy to harm, we propose expanding criminology’s mission to embrace harm, in part, by positioning harm assessment as a core disciplinary task, and re-centering criminal policy on harm. The expansion and re-centering of each, respectively, would likely be mutually reinforcing (Table 1). If criminology and criminal policy can take on harm together, they might each be better able to contribute to justice.
Potential gains to justice from taking on harm in criminology and criminal policy.
We proceed as follows. To start, we review our approach to harm, harm reduction, and harm assessment. Our intention is to provide enough methodological grounding to support our recommendations. (See Greenfield and Paoli (2022) for the details of our harm assessment framework and comprehensive applications to drug trafficking and production. For applications in other varied contexts, see, for example, Greenfield et al. (2016), Paoli et al. (2018), and Peters and Paoli (2020)). Then, working with Sen’s (2009) conceptualization of justice, we discuss our understanding of the relationship between harm and justice. Next, we explore how orienting criminology and criminal policy toward harm can contribute to justice, by pointing to opportunities to address harm in criminology and criminal policy, results of harm assessments, and recent harm-based initiatives in policy and policing. We conclude with our final recommendations for taking on harm to advance justice.
Our approach to harm, harm reduction, and harm assessment
In this section, we discuss our approach to harm, harm reduction, and harm assessment. We begin with our definitions of harm, itself, and of harm reduction. To explain how we are engaging with harm assessment, we introduce our framework.
Defining harm and harm reduction
Our definition of harm consists of a conceptual anchor and a concrete specification. Referring to Feinberg (1984: 31–36), a noted legal philosopher and the source of our anchor, we treat harm as “a setback to a rightful claimant’s legitimate interests” (Greenfield and Paoli, 2022: 1). Thus, an activity is harmful when it impinges, intrudes on, or damages a rightful claimant’s—or stakeholder’s—legitimate interests. Recognizing the inherent challenges of establishing rightfulness and legitimacy—the normativity of both is inescapable—we set explicit bounds on what counts and what does not when we conduct harm assessments. For example, we tend to exclude losses of ill-gotten gains from consideration as harms.
To make harm analytically tractable we define it more precisely as a multifaceted phenomenon—or a bundle of attributes—consisting of a negative outcome of a particular type, severity, and incidence that is associated with a particular activity, such as drug trafficking, and is borne by particular individuals, institutions, or the social or physical environment, which we refer to as classes of bearers. Institutions include private-sector entities, such as businesses and NGOs, and public-sector entities, such as government agencies. The physical environment includes wildlife and ecosystems and the social environment, by contrast, involves relationships among individuals and institutions. A single activity, even a simple crime, can yield multiple harms of different types, severity, or incidence that can be borne by different classes of bearers.
The specificity of our definition supports systematic, empirical assessment and can allay concerns that the concept of harm is “neither fixed nor analytically robust” (Lacey and Zedner, 2023: 169). Yet, if we were to consider non-criminal activities (Presser, 2013), it is capacious enough to encompass Tifft and Sullivan’s (2001: 198) early definition of social harm—that is, “actions or arrangements that physically and spiritually injure and/or thwart the needs, development, potentiality, health, and dignity of others”—and, as we discuss later, it may also bear extending to other more-recent concerns for structural conditions.
Extending the concept of interests, we delineate the type of harm in terms of damages to one or more of the interest dimensions of each class of bearers. The dimensions, consisting of functional integrity, material circumstances, reputation, and privacy and autonomy, represent different aspects of their interests and represent capabilities that create paths to attaining:
a certain quality of life or standard of living for individuals (Sen, 1987);
a mission for institutions; or
sustainability for the physical or social environment.
Inflicting harm in any dimension means creating an obstacle to attaining a certain quality of life, a mission, or sustainability. As we discuss below, the living standard, the mission, and sustainability serve as benchmarks for gauging the severity of harms for each class of bearers, respectively. With this approach, we have expanded on von Hirsch and Jareborg (1991: 11–14), who identified similar “interest dimensions” and used the living standard (Sen, 1987) to gauge harms to individuals, by casting a wider and less anthropocentric net, and we have established a firm link to Sen’s (1987, 2009) treatment of capabilities.
Two main elements of our definition of harm—type and bearer—comprise the taxonomy (Table 2) that is the backbone of our harm assessment framework. The taxonomy sets out the universe of stakeholders that can bear harms and the types of harm they can bear.
Taxonomy of types and bearers of harms.
Source: Authors’ adaptation of Greenfield and Paoli (2022: 91) and earlier versions, extending von Hirsch and Jareborg (1991) and Sen (1987).
X = applicable; N/A = not applicable.
Functional integrity refers to physical, psychological, and intellectual integrity.
Functional integrity refers to operational integrity.
Functional integrity refers to physical, operational, and aesthetic integrity.
The interest dimensions can mean slightly different things, depending on the class of bearers. For example, individuals can experience functional harm physically (e.g. from a wound), psychologically (e.g. from trauma), and intellectually (e.g. from a loss of access to education), whereas private and public institutions can experience it operationally (e.g. from a disruption in production or services) and the physical or social environment can experience it physically, operationally, and less tangibly. For the social environment, concerns can include public nuisance, social fragmentation, perceptions of insecurity, and loss of trust, potentially reaching well-beyond local communities to the cohesion of a society.
As for the meaning of harm reduction, we acknowledge that it has been debated intensely (Greenfield and Paoli, 2012) and adopt an expansive understanding that covers the prevention, mitigation, and remediation of harm. Harm reduction, in our work, can mean stopping a harm from occurring, lessening its severity or incidence, or redressing it.
Operationalizing harm
Our approach to harm assessment is encapsulated in our framework. The framework, which draws from wide-ranging literatures and disciplines, consists of a multistep process and a set of complementary tools (Figure 1). Our approach is, as we have said, systematic and empirical, meaning based in verifiable observation, but not necessarily quantitative. We cull information from court records, case files, interviews, and various other sources, some of which may contain hard numbers on bearers, damage, and so on, and we use the evidence to rate, rank, and prioritize harms, but we do not convert our findings into point estimates or indices. Our approach also represents a significant departure from other recent efforts, such as Sherman et al.’s (2016) Crime Harm Index, which uses sentencing guidelines to measure harm quantitatively, yet less empirically (Greenfield and Paoli, 2022: 61–62).
The harm assessment framework.
First, we
Next, we
To rate the severity of a harm, we consider the extent to which the damages associated with an activity intrude upon any or all of the interest dimensions of the bearer, in relation to the corresponding benchmark, that is, the living standard, the mission, or sustainability. At one end of the scale, the severity can be marginal and, at the other end, it can be catastrophic, with the possibilities of moderate, serious, or grave harm in between. For examples at each end, a slap, with no lasting physical, psychological, or other consequences, would do little to intrude on an individual’s interests, hence living standard, but a homicide would effectively eliminate all interests. Still, a slap could result in greater harm, if, for example, the victim fell and was badly injured or traumatized, in which case it would amount to an assault, which we would treat as such. Table 3 sets out the rating scheme for each class of bearers.
Rating severity against benchmarks.
Source: Authors’ adaptation of Greenfield and Paoli (2022: 105) and earlier versions, extending von Hirsch and Jareborg (1991) and Sen (1987).
As shown in Table 3, a harm can impinge on an individual’s mere subsistence and survival or encroach at progressively higher levels of life’s quality. Similarly, for an institution or the environment, the damage can be so severe as to render it unviable.
To rate the incidence, we use a scale that ranges from rarely, verging on never, to continuously, verging on always, bridged by seldom, occasional, and persistent. To illustrate, a rare harm might register in less than 5%–10% of all the case files for a certain criminal activity, a continuous harm might register in more than 90%–95%, and seldom, occasional, and persistent harms might occupy the mid-range in even increments.
If evidence is lacking on severity or incidence, it might be necessary to consolidate the levels, for example, to marginal, serious, and catastrophic or to rarely, occasionally, and continuously.
We can
Matrix for ranking and prioritizing harms.
Source: Authors’ adaptation of Greenfield and Camm (2005), drawing from U.S. military guidance.
VH = very high; H = high; M = medium; L = low priority; VL = very low priority.
Although we cannot say whether the highest-ranking harms to individuals matter more or less than those to institutions or the environment, we can, nevertheless,
With an eye to policy, we also
In our investigations, we consider:
the remoteness of a harm (Ryberg, 2004: 64–66; Simester and von Hirsch, 2011: 70–88), which is the temporal, spatial, or social distance between the harm and the activity, and
the extent to which a harm is intrinsic to the activity or, instead, emerges from policy, interventions, or other aspects of the context in which it occurs.
We investigate the latter largely through counterfactual reasoning (see, for example, Pearl and MacKenzie, 2018).
In our investigations, we typically drill down at least as far as policymakers can reasonably act in a given arena and sometimes farther. For example, in the case of drug trafficking (Paoli et al., 2013), we identified the relationship between illegality, law enforcement, and violence, whereas, in the case of human trafficking (Greenfield et al., 2016), we considered the effects of closed borders. In those and other applications we have found good reason to look further than the criminal activity, itself, as the cause of a harm, because doing otherwise might point wrongly. Accordingly, we refer to the harms associated with an activity until we have learned whether the activity is the source of the harm.
All told, we see similarities and differences in our approach and those of zemiologists and other scholars of social harm. Our taxonomy shares ground with those developed by zemiologists (e.g. Hillyard and Tombs, 2004; Pemberton, 2015 and Canning and Tombs, 2021) and implicitly with Doyal and Gough’s (1991) theory of human needs, on which Pemberton (2015) and others rely. That said, we occasionally categorize harms differently. For example, we do not call out harms of (mis)recognition, which constitute a key dimension of social harm for Yar (2012) and Pemberton (2015), but we do speak of harms to autonomy, which, according to Canning and Tombs (2021: 67), “are very close to ‘harms of [mis]recognition.’” We also grant standing to institutions and the environment, which tend to attract less zemiological attention than individuals. The major, most fundamental difference, though, is that we offer a comprehensive methodology for systematically and empirically assessing harms.
The relationship between harm and justice
Our understanding of justice draws from Amartya Sen’s (2009) pathbreaking work, “The Idea of Justice.” Sen, an economist, philosopher, and Nobel Laureate, offers a perspective on justice that departs notably from Rawls’ (1971) quest for perfectly just institutions, by emphasizing social realizations and, specifically, whether people have the capabilities and freedom to attain them. Sen (2009: 15–18) refers to Rawls’ theory of justice as transcendental and his own approach as comparative, entailing rankings of alternative scenarios.
To clarify, Sen contrasts two concepts of justice in early Indian jurisprudence. The first concept, niti, relates to organizational propriety and behavioral correctness, and the second, nyaya, relates to “the lives that people are actually able to lead” and “the world that actually emerges” (Sen, 2009: xv, 20). With the phrase “able to lead,” Sen homes-in on whether people have the capabilities and freedom to attain a certain quality of life or living standard, and, with the second phrase, he points to broader societal outcomes, albeit largely anthropocentric. Sen does not, however, dismiss the relevance of the processes and other forces that can operate along the path to attainment (e.g. Sen, 2009: 215) and speaks of “comprehensive outcomes” (Sen, 2009: 22: 215–217) that include the final results of a policy, the way the results are achieved, and whether they are achieved justly (Tyler, 2006). Still, Sen (2009: 210–213) advocates for sensitivity to consequences—“the results of one’s choices and actions must matter in deciding what one should do”—and argues strenuously for evidence-based “practical reasoning,” with which to diagnose injustice, identify ways to reduce injustice, and, thus, advance justice in open, democratic debate (Sen, 2009, for example, vii, ix, 93–94, 44–46).
One might say that our approach to harm—and its assessment—aligns intrinsically with Sen’s perspective, given our long-standing focus on consequences (Paoli and Greenfield, 2015) and our aim to provide an instrument of “practical reasoning.” We emphasize nyaya, as Sen does, and take a similar, but yet-broader, less anthropocentric view of justice, in which individuals, institutions, and the environment have the capabilities to attain a certain living standard, a mission, or sustainability, respectively. True, institutions exist in service to others—potentially, as “forces along the path to attainment”—and, on that basis, they may not merit independent standing, but we grant them standing by extension. We frame capabilities as interests, including privacy and autonomy or “freedom” (Greenfield and Paoli, 2022: 96), and harms as arising from damages to those interests, which create obstacles to attaining a certain living standard, a mission, or sustainability. Accordingly, harm—treated as a setback to a rightful claimant’s legitimate interests—represents an instance of injustice.
Responding to Sen’s appeal for consequence-sensitivity and evidence-based practical reasoning, we offer our framework as an instrument for:
Diagnosing harm, an outcome that represents an instance of injustice.
Contributing to harm reduction, hence, reducing injustice.
Advancing justice.
Moreover, by ranking harms according to their harmfulness and supporting comparative analysis, we mirror Sen’s interest in comparative justice. That is, we offer a means of comparing the justness of outcomes under different scenarios of policy and interventions.
Furthering the case for evaluation, “effectiveness” holds special meaning in criminal policy because it employs means, namely punishments, that are inherently—and intentionally—harmful. While defending the state’s duty to punish, Tadros (2011: 1), a prominent contemporary jurist, admits that “punishment is probably the most awful thing that modern democratic states do to their own citizens” (see also, for example, Hart, 2008[1968] and Clear, 1994). Criminal punishment, especially imprisonment, has a profound impact on citizens’ lives that, to our knowledge, has no comparison in other policy arenas. Although unintended, the impact often reaches beyond the convicted person to their family and community (e.g. Travis et al., 2014), yielding additional harm to them. Imprisonment is also costly in purely financial terms. In the extreme, US corrections costs approached $90 billion a year in 2017, with an incarcerated population exceeding 2 million (Buehler, 2021: 4, Maruschak and Minton, 2020: 2). In Europe, the incarcerated population is much smaller, but the costs of incarceration are still substantial.
Thus, we maintain (e.g. Greenfield and Paoli, 2022) that policymakers have a distinct obligation to clearly articulate the objectives of criminal policy; determine whether the means are achieving those objectives; and prevent the means from becoming objectives of their own (Goldstein, 1979: 236), without regard to the harms they might cause. If punishment, specifically, can only “be justified in virtue of the good effects that it can have” (Tadros, 2011: 2), the evaluation must be rigorous. Systematically and empirically evaluating the benefits of criminal policies and interventions in terms of harm reduction would constitute a first step. However, even if the means of policy appear to be working, it is incumbent upon the policy community to weigh the costs, including any unintended consequences, against the benefits.
Before turning to how criminology and criminal policy can play a larger part in advancing justice, we pause to consider, again, the meaning of advancing justice. Until now, we have spoken of harm reduction as a singular objective, as if reducing harm for one bearer—or on balance—would necessarily advance justice for all bearers. But a policy that reduces harm on balance can reduce harms for some bearers at the expense of others and alter how different bearers experience crime, criminal policy, and harms. As Feinberg (1984: 35) contends, any legal system determined to “minimize harm” must incorporate judgments of the comparative importance of interests of different kinds so that it can pronounce “unjustified” the invasion of one person’s interest of high priority done to protect another person’s interest of low priority.
Accordingly, we should not consider effectiveness in isolation, but should also consider the distributional effects of crime and the policies that are—or could be—adopted to address it.
Embracing harm in criminology
Against that backdrop, we argue that criminology should embrace the study of harm and strive to contribute to reducing harm equitably, as many critical criminologists and scholars of social harms also advocate (see, for example, Canning and Tombs, 2021), and we call for positioning harm assessment as a core disciplinary task. Unlike some zemiologists, we are not seeking to provide an alternative to criminology, but to strengthen criminology’s connection to justice, by expanding its mission, and to support—in methods and analysis (Table 1)—criminal policy. Criminology already has much to say about why crime occurs and what it looks like and, with this embrace, criminology would be better able to speak to the consequences of crime for individuals, institutions, and the environment. The embrace may also lend support to victimology (e.g. Green and Pemberton, 2017 and Letschert and van Dijk, 2011) and, potentially, to green criminology (e.g. Greenfield and Paoli, 2025; White, 2013). Criminology would then be well positioned to explore whether some policies or interventions would contribute more to reducing harms than others, how the effects span different interests and bearers, and, perhaps, whether harm or its reduction justifies criminalizing specific activities. 2
Criminology can also play a part in addressing harms dealt with outside criminal law, including those that private and public institutions generate, without claiming they are all crimes (see Sutherland, 1945). Clearly, other disciplines dealing with the environment, safety, health, and so on, would lead the work in those fields, but criminology can also consider activities thought to have substantial adverse consequences and bring empiricism to bear on decisions about criminalization. In turn, criminologists might find novel ways of preventing and restoring harms outside criminal law (see, for example, Braithwaite, 2000, 2022).
Finally, criminology can deepen its connection to what Williams (2005: 3, quoted in Bottoms, n.d.) terms a society’s primary political question, that is, “the securing of order, protection, safety, trust, and the conditions of co-operation.” Through harm assessment and greater attention to harm reduction—prevention, mitigation, and remediation—criminology can help ensure that its efforts are used to promote a more just order (see Walgrave, 2021).
Stanley Cohen (2016 [1979]: 86) called for “a moratorium on nearly all types of standard criminological research and publication,” especially work on the causes of crime, which he would have replaced, in part, with work on “differential perceptions of the seriousness of various crimes and the level of appropriate punishment,” “the degree to which certain forms of deviance can be tolerated without invoking the criminal sanction,” and “the extent and nature of harm, victimization and damage by crime . . .” We think that Cohen’s proposed moratorium is extreme, but that his proposed research program merits close consideration. Such research can, as we discuss below, benefit criminal policy to the ultimate benefit of justice.
Re-centering criminal policy on harm
In this section, we show how re-centering criminal policy on harm, with support from rigorous harm assessment, can benefit policy in pursuit of justice. Figure 2 indicates where, among the phases of criminal policy, the framework can play a part domestically, though we also discuss potential international applications. The advancement of justice could occur more or less immediately, for example, by identifying a source of harm, suggesting options for alleviating harm, reinforcing an ongoing effort to alleviate harm, or addressing disparities in the consequences of crime, criminal policy, and interventions across society. We also point to lessons from applications of our framework and to recent initiatives on harm in the United Kingdom and elsewhere as evidence of a constructive role for a harm-based lens.
Potential contributions of harm assessment to criminal policy.
Establishing criminality
Taking a consequentialist perspective, our harm assessment framework can contribute to decisions about whether specific activities, criminal or otherwise, warrant criminalization. The “harmful-conduct principle” and the “harm-prevention principle” (Duff and Marshall, 2014) suggest evaluating, respectively:
Whether an activity, criminal or not, is harmful enough to be deemed criminal. We do not suggest adopting a particular harm threshold for criminalization, but rather using the framework to better calibrate policy responses to harm.
The strength of the relationship between the harm and the activity to determine whether criminalizing the activity would prevent substantial harm. If the harm is remote, one might ask how criminalization would affect the outcome and for whom.
In this way, an activity might be deemed “criminal” when doing so reduces harm, while considering the distributional effects of the activity and response. A harm assessment might also provide strong arguments for avoiding an excessive extension of the harm principle and for questioning the rise in preventive offenses (see, for example, Duff, 2007; Harcourt, 1999 and Ashworth and Zedner, 2014, for related concerns).
Setting strategic and tactical priorities
Comparing harms across different types of already-criminal activities can provide initial evidence for setting strategic priorities in crime control, policing, and crime prevention. Policymakers might, for example, rank activities according to their harmfulness and, preliminarily, look to address the worst first. Still, they would need to consider the feasibility, costs, and distributional effects. 3 The exercise might also allow policymakers to contrast (1) the policy priorities that would emerge from a systematic, empirical assessment of harm, (2) current priorities, and (3) public perceptions of priorities, and to initiate a public debate when the priorities differ (Adriaenssen et al., 2020). A comparative assessment could also help the police—along with others—identify a community’s most serious problems and weigh competing interests in responses to them (Goldstein, 1990). Identifying especially harmful perpetrators or harmed bearers—the “power few” (Sherman, 2007: 229)—might also suggest avenues of crime control and prevention and help the policy bodies set tactical priorities, possibly with less risk of stereotyping or profiling. The High Point Drug Market Intervention (Corsaro et al., 2012) took such an approach to reducing violence.
Reviewing sentencing policies and practices
Assessing the severity of the harms of a crime, irrespective of the harms’ incidence, can yield evidence to review the sentences established by criminal codes and sentencing guidelines and to review related sentencing practices. These policies and practices are nearly all founded, at least in theory, on the principle of proportionality of the punishment to the seriousness of the offense (Husak, 2008: 83). Noting that seriousness in this debate consists of a combination of harmfulness and culpability, the assessment of severity might supply an empirical gauge for the former that would stand independently of prior policy decisions, public perceptions, and political views (Adriaenssen et al., 2020). The assessment might also lead to better alignment between the intended harm of the punishment and the harm of the crime that prompted it.
Strengthening remediation-oriented programs
A systematic, empirical assessment of the harms of crimes to different classes of bearers, including individuals, institutions, and the environment, might help strengthen victim programs, restorative justice, and international criminal justice. It could expand the reach of victim programs and restorative justice beyond traditional crimes and individual victims (e.g. Letschert and van Dijk, 2011) and, in all three fields, better acknowledge harms of private and public institutions and harms to nonindividual bearers (e.g. Walgrave, 2021; White, 2013). In restorative justice and international criminal justice, the assessment could provide a basis for aligning restoration and reparation, respectively, with harm (Daly and Proietti-Scifoni, 2011). It could also address restorative-justice scholars’ (e.g. Duff, 2003) concerns about proportionality.
In addition, an assessment that takes the perspective of an “impartial spectator” (Sen, 2009: 44–46, 124, 404, citing Smith, 1976[1790]), or attempts to, insomuch as possible, through a systematic, empirical process, could help policymakers identify and address a fuller range of the harms of a conduct. It could do so even if victims do not report harms, but other evidence exists, and avoid privileging the harms recounted by the most vocal victims.
Evaluating policies and interventions
Our framework can contribute to evaluations of current policies and interventions to address crime and feed into deliberations about future policies and interventions. It could be used to assess and compare the impact, including the unintended consequences and distributional effects, of different types of policies and interventions, and to inform a notional benefit-cost analysis of the alternatives. We say notional, because our framework does not require quantifying let alone monetizing benefits (i.e. the gains from reducing harm). However, we do not rule out quantifying some benefits (e.g. material) as is typical in the cost-of-crime literature (Greenfield and Paoli, 2022: 71–74). Moreover, conducting a full comparison of alternative policies or interventions, would necessitate considering implementation costs, some of which could be more readily calculable than benefits.
Building on Greenfield and Paoli (2012: 11) we suggest four stages of analysis:
Establish whether a policy or intervention reduces harms on balance, after accounting for the positive and negative consequences, by comparing conditions with and without it.
Identify winners and losers under the policy or intervention, by examining the distribution of harms across society with and without it, and consider options for compensation.
Assess the overall effect of the policy or intervention on the balance and distribution of harms in relation to the resources needed to carry it out.
Repeat the process and compare the results for any alternatives.
The first stage allows for the possibility that a policy or intervention can engender harms of its own that could negate its value and might suggest dismissing the policy or intervention and looking for a better alternative. Moreover, as implied in the third stage, we would not count spending on policy implementation, including law enforcement, or on protection as material harms, but as costs to weigh against the benefits of reducing harms. On that basis, a policy could reduce harm on balance, but present unacceptable costs. In the United States, which has long required benefit-cost analysis for significant regulatory measures, recent guidance has required consideration of the distributional consequences (US Office of Management Budget (OMB), 2023).
Even if much of the analysis is qualitative, it can still provide a systematic, empirical basis for policy evaluation. National Academies of Sciences Engineering Medicine (NAS) (2018) and OMB (2023) provide illustrations and discuss qualitative methods of analysis.) For example, we could build on our prior assessment of coca cultivation and processing in Colombia (Greenfield and Paoli, 2022) to conduct a notional benefit–cost analysis of drug control and related policies in that country. The analysis might also suggest less obvious yet potentially harm reducing courses of action. Quimbayo Ruiz (2008: 1) has suggested furnishing coca cultivators and processors with training on using fertilizers, herbicides, and other chemicals safely. Alternatively, we might focus on the benefits, costs, and unintended consequences of a specific means of criminal policy, such as imprisonment.
Insomuch as we shift our gaze to policies and interventions, we are adding them to the bundle of attributes that defines harm. Specifically, we are including particular policies and interventions as a possible source of harm and an alternative to particular activities.
In summary, we believe that a harm-based lens can help rationalize criminal policy and ensure that the means of criminal policy, including punishment, are used to advance justice, and do not become objectives in themselves or yield excessive harm. Our case rests on three related claims, that is, that harm is an instance of injustice, that reducing harm can advance justice, and that harm assessment can facilitate or lead to harm reduction. Our confidence in the third claim is founded not just on theory, but also on the results of our and others’ applications. To illustrate, we first turn to the evident value of investigating causality, which is the final step of our framework (Figure 1), and then to others’ initiatives.
In our investigations of causality, we have found that a harm can often be traced back to the context, including the policies and interventions, in which a crime occurs and that failing to address the root cause of the harm can reinforce negative outcomes or yield its own damage. For example, in our assessment of coca cultivation and processing in Colombia (Greenfield and Paoli, 2022), we learned that most harms to the environment resulted from the risks of cultivating the crop, largely a consequence of its illegality, and the specific method of eradication. While the risk of involuntary eradication by any method can encourage environmentally degrading land management practices, the extent and form of the damage can differ by method. In the case of fumigation, the chemicals may have lasting toxic effects and aerial spraying, a once-common practice, can yield further harms through non-selectivity and migratory pressure. Thus, to say that “coca cultivation harms the environment,” without considering policy’s part in creating harms, and then to demand more eradication—or a return to aerial spraying—in response to cultivation might serve to further degrade the environment. In the case of human trafficking, we also found evidence that the larger policy context, particularly restrictions on migration, was contributing to harm (Greenfield et al., 2016). Such information on causality can contribute to all phases of criminal policy, but certainly to decisions about criminalization, strategy, and tactics and in policy evaluations.
Although our focus has been on evaluating criminal policy, here too we see applicability to non-criminal concerns, possibly edging closer to broader concerns about social harm. For example, researchers and policymakers might use our framework to weigh the consequences of policies on development, land tenure, or migration directly, taking the policies, not a criminal activity, as the starting point for the assessment. Borrowing again from guidance on benefit–cost analysis (e.g. OMB, 2023), they could evaluate the policies first in relation to their intended objectives and then in relation to their ancillary benefits and collateral consequences, which might include their effects on criminal activities and associated harms.
Finally, harm assessment is making inroads outside scholarly pursuits, notably in the United Kingdom, where policy and policing agencies have been working to conceptualize and operationalize “harm” for strategic and tactical purposes. In 2008, the UK’s Serious Organised Crime Agency (SOCA) stated that “the overarching aim of the [Organised Crime] Control Strategy” was “to achieve a tangible and lasting reduction in the harm caused to the UK by organized crime” (UK Serious Organised Crime Agency [SOCA 2008]: n.p.). Although the SOCA has since dissolved, interest in harm in the United Kingdom has persisted. SOCA’s successor, the National Crime Agency (see, for example, Management of Risk in Law Enforcement, 2017) and local policing agencies have developed indicators of consequences to assess and rank the harm or seriousness of different crimes. Most recently, the UK’s Office of National Statistics (ONS, 2022) has used our framework to develop its own “crime harm framework” and a publicly available “crime harm interactive database” for collating and organizing data on “the wide-ranging impacts crime can have on individuals and society” and “[t]o achieve the most comprehensive and detailed categorisation of the available data . . .” The ONS has gone the farthest to implement our vision for harm and demonstrate its value, but harm has also gained attention in Australian, Canadian, Chilean, and other policy circles.
Conclusion
Kitcher (2001: 145), a philosopher of science, asks all scientists: “What is the collective good that we want inquiry to promote?” and, as we have written previously, our answer is “advancing justice” (Greenfield and Paoli, 2022: 260). We are convinced that the systematic, empirical assessment of harms can—and should—contribute to a thorough diagnosis of injustice, especially but not only criminal, to identifying ways to reduce it, and, thus, to advancing justice (Sen, 2009: vii, ix). If harm, as we have defined it, limits the capabilities of individuals, institutions, and the environment and presents obstacles to attaining a certain quality of life, a mission, or sustainability, then reducing harm would represent a step toward justice. Social scientists such as us cannot—and should not—alone choose the course of action in a democratic society, but we can play a part. If democracy is “government by discussion” (Sen, 2009: 324, referring to Walter Bagehot and John Stuart Mill), we believe that concern for harm should steer that discussion and that rigorous harm assessment should inform it.
We conclude with our final recommendations for solidifying criminology’s and criminal policy’s connection to justice:
Expand criminology’s mission to embrace the study of harm and position harm assessment as a core task to more-fully support a re-centered criminal policy. The assessment should pay close attention to both the amount and distribution of harms among the individuals, institutions, and environment that comprise a society.
Target harm reduction as a central aim of criminal policy, drawing insight from a criminology that embraces harm. Perhaps most ambitiously, we would make harm reduction, not crime reduction, the end goal of a criminal policy that accounts for how much harm society is bearing and which segments of society are bearing it. Here too, we would look to systematic, empirical assessments of the harms of crime for guidance in policy analysis, development, implementation, and evaluation.
In a recent review of Greenfield and Paoli (2022), Tombs (2023: 1618) refers to our work as scholarly, rigorous, and path-breaking, and then calls for even broader applications of our framework: . . . perhaps the next, crucial, step—a challenge for all of us, not necessarily or in fact even for Greenfield and Paoli—is to transcend the fascination with crime harms and to determine the harms that really blight the lives of children, women and men on a daily basis: those actions, omissions, policies, processes, conditions, states of affairs, assumptions, ways of doing things and dominant structures which contribute to the absence of or distortions in human self-actualization.
Tombs’ proposed transcendence may not require a huge leap. Although we developed our framework to assess the harms of crime, we have shown that some of the most significant harms associated with the criminal activities in our applications emerge from policy and interventions, and, as early as 2012, we advocated for assessing the harms of policy, per se (Greenfield and Paoli, 2012). Although we do not share all the ontological and epistemological foundations of either critical criminology or zemiology, we have argued for reaching beyond criminal justice to justice, and we are interested in studying broader, non-criminal harm (Canning and Tombs, 2021: 65), even if we do not intend to leave criminology behind. Notwithstanding our past focus on crime and criminal policy, we believe that our framework can, as Tombs suggests, bear on policy, itself, in and outside the criminal domain, and perhaps address Kotzé (2018: 97) plea for an integrated approach to understanding criminalized and un-criminalized harms. We will need to think carefully about what will and will not “work” in our framework, but we believe, given our experience with past applications, that most of it will be “workable.”
As we look to the future, we will strive to heed Tombs’ call and we hope that others will join us, including those who have long sought to reduce the harms of crime and improve criminal policy and those committed to the study of social harm. We look forward to seeing zemiologists and other scholars seeking to advance justice use our framework to empirically and systematically examine more of the harms that blight society daily.
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.