Dynamic Risk Factors, Protective Factors and Value-Laden Practices

Abstract

Dynamic risk (and to a lesser extent protective) factors are the foundation of correctional practice; the assumption that they exist, can be measured and are able to change is at the heart of what forensic practitioners do. However, there has recently been a surge in interest and debate around what these constructs are and how they relate to offending. In progressing this debate, we shift the focus from risky characteristics, behaviours and contexts (e.g. antisocial attitudes, associates, drug abuse), to the practices (i.e. goal-directed actions) to which these descriptions refer. Embedded within practices are values (i.e. priorities, motivators, norms), and underpinning them human capacities. Identification of these capacities and relevant contexts (i.e. norms, opportunities) can inform rehabilitation, which strengthens them to support healthier and less harmful functioning. We offer examples of risk and protective factors for sexual offending, although the ideas are not limited to this type of behaviour.

Introduction

It is generally accepted that the criminal justice system is based upon ethical or moral value systems. Collectively we decide that certain actions are wrong or harmful, and seek to restore justice by punishing the perpetrator and ideally rehabilitating them in order to avoid further victimisation (Ward & Salmon, 2011). Laws exist to protect the public from potential harms, and in this sense they are based upon social and cultural values that denote the nature and source of harm; who is protected and from what or whom varies across time, location and contexts (e.g. age of consent or culpability, drug legislation; Tadros, 2016). There are behaviours that are widely accepted as wrong and harmful (e.g. murder, rape), but even these are considered more or less wrong in certain contexts where harm can be justified (e.g. war, capital punishment, self-defence). An objective of the criminal justice system is to reduce further incidences of harm by those who have already offended. Accordingly, the field has placed increasing importance upon evidence concerning ‘what works’ to reduce crime (Craig, Gannon, & Dixon, 2013; McGuire, 2013). Over the past four decades this commitment to evidence has prompted the development of empirically informed rehabilitation theories containing lists of ‘dynamic risk factors’ (DRFs) and ‘protective factors’ (PFs) to guide practice (Andrews & Bonta, 2010; Bonta & Andrews, 2017; de Vries Robbé, Mann, Maruna, & Thornton, 2015).

DRFs and PFs can be broadly defined as features of individuals and their environments that indicate increased (DRFs) or decreased (PFs) likelihood of reoffending, derived from research tracking large groups of individuals convicted of crimes (i.e. they are statistical predictors). DRFs, and to a lesser extent PFs, are central to correctional practice; they are used to make decisions concerning level of risk, access to and goals of treatment (e.g. criminogenic needs; Andrews & Bonta, 2010), and whether an individual is sufficiently rehabilitated to reintegrate into the community. Thus correctional practice is also based upon epistemic or scientific values concerning the nature and quality of the evidence available in the field – for example, what empirical data mean in terms of risk level, what counts as ‘success’ in treatment outcome studies and the role of empirical findings in explanations of crime.

While the role of values within concepts such as justice, harm, crime, punishment and responsibility is relatively undisputed, less attention has been paid to their influence upon the process and content of correctional rehabilitation initiatives. We have discussed the value-laden nature of correctional rehabilitation interventions elsewhere (see Ward & Heffernan, 2017), key issues being the ‘failure to appreciate the pervasiveness of values in the generation of knowledge . . . ’ and that ‘ . . . theoretical and ideological allegiances may distort the detection and explanation of phenomena’ (p. 50). Assumptions concerning what phenomena cause crime and, perhaps most worryingly, interventions that contain implicitly value-laden, and at times competing, goals (e.g. community safety versus promotion of individual wellbeing) create problems for practitioners. In other words, when researchers and practitioners overlook the role of values in informing risk (and need) judgements, they run the risk of designing and implementing treatment programmes that prioritise collective ethical and social values (e.g. reduced crime) over the values of individual participants (e.g. meaningful relationships, work, autonomy). In addition to the ethical issues arising from the use of risk-avoidant treatment programmes, there is concern that individuals may fail to be motivated by programmes that are inherently aimed at reducing negative characteristics for the benefit of others, without the expectation of better outcomes for participants themselves (Ward & Maruna, 2007).

In this paper we utilise the concept of practices to explore the extent to which DRFs and PFs are value laden and, perhaps most importantly, discuss what this means for correctional practice. We begin by outlining the influence of values and norms upon practices and subsequent judgements about the degree to which they manifest in offending behaviour. We then present a well-established set of DRFs and PFs for sexual offending, which guide the rest of our discussion. It is acknowledged that variation exists both between offence types and within individuals who commit this category of offence, but for our purposes it is most useful to focus on one well-established set of DRF and PF domains. These represent features observed across large samples of individuals with detected sexual offences, and so they offer a good starting point for researchers to think about potential causes and solutions for this norm-violating behaviour (Mann et al., 2010; Ward & Fortune, 2016). We use case vignettes to explore two DRF and PF domains with respect to relevant normative practices and the values that influence them. For example, the practices constituting ‘intimacy’ refer to objects of intimacy or attraction (i.e. types of people), the contexts they exist within (e.g. time, place, social group) and the behaviours enacted (e.g. communication, sex). The normative commitments inherent within intimacy practices spell out what is widely considered desirable, acceptable and healthy behaviour in intimate relationships. We also offer some preliminary suggestions regarding the sorts of capacities involved in these practices (i.e. those that facilitate ‘healthy’ functioning). Finally, we turn our attention to potential ways in which consideration of values can be beneficial both for theoretical research and in correctional treatment.

Values and practices

In the context of everyday activities, values are embedded within practices. Practices are the application of practical knowledge within goal-directed action sequences, governed by ‘a structured body of norms’ (Wallace, 2009, p. 11). Norms are evaluative in nature and spell out whether or not an activity is done properly and if it meets the socially accepted relevant standards. For example, whether or not the actions constituting football-related practices are in accordance with the rules, reflect skilful performances or whether the game is fairly officiated. Because of their focus on successful action, practices have both causal and normative dimensions. Practices are goal-directed cognitive and behavioural activities that are intended to address specific tasks such as problem solving, planning, explaining and justifying action, establishing and maintaining relationships, regulating emotions, engaging in sexual activity, and so on. Practices are underpinned by causal mechanisms and the capacities they constitute. For example, a set of causal mechanisms might create difficulties in one’s capacity to infer mental states in another person, resulting in harmful intimacy-related practices (e.g. sexual offending) in certain contexts (e.g. when experiencing loneliness and/or intoxication). If a crime is committed, then the hypothesised causal condition(s) or practice(s) involved are given the status of DRF. For example, the mechanisms thought to underpin ‘intimacy deficits/problems’ or the practice of ‘substance abuse’ are only DRFs if retrospectively observed to precede a criminal offence (i.e. an illegal practice). This means that DRFs can come in and out of existence contingent on changes in the law or in ethical standards. Thus, what was once thought to be a DRF may cease to be one (e.g. changeable factors predicting homosexuality), and what was once considered to be benign or even beneficial might now be viewed as harmful (e.g. sex with early adolescents 300 years was socially acceptable whereas now it is a crime). This fact underscores their partly normative status, and thus their reliance on values (Ward & Heffernan, 2017).

We would like to stress here that our argument that DRFs, PFs and their manifestation (i.e. illegal practices) are value laden does not mean that scientific inquiry plays no role in their meaning (or in the referents of these concepts). They are hybrid constructs that contain both factual components (i.e. observations) and normative ones (i.e. value-based judgements). The purpose of examining the degree to which they are value laden is to point out that normative discussion is an indispensable part of correctional research and practice. Figure 1 highlights the crucial role of values and their associated normative commitments both in shaping and in evaluating practices. The danger for forensic practitioners lies in focusing risk detection and subsequent treatment upon enduring personal capacities (often in combination with risky contexts), and adopting the default assumption that this is a value-free process, guided solely by facts and their evidence. In reality, values (personal and professional sets of normative commitments) not only shape professional judgements, but also provide the context within which offences occur. We are unable to remove values from the process of assessment and treatment, because values of various types (e.g. ethical, social, empirical, personal/prudential) are present within the practices that we are trying to change (e.g. reoffending, desistence), as well as within the practices and objectives of the criminal justice system as a whole.

Figure 1.

The role of values in professional judgements concerning risk.

When conducting an assessment for the purpose of sentence planning, informing treatment or managing risk, practitioners have various sources of information to draw upon. Firstly, they are often provided with details of the offence, sometimes just the category name (e.g. unlawful sexual connection or rape) but most often police summary of facts and/or a judge’s sentencing notes from the individual’s trial as well. Sometimes this is supplemented with notes or reports containing the observations and professional opinions of other practitioners (e.g. previous probation officers, custodial staff and psychologists) and/or victim impact statements. Secondly, in most cases, the practitioner will also speak to the individual concerned. This typically includes (but is not limited to) first-person reports of their developmental history, the background to their index offence and details of the offending. Thus assessments are jointly informed by first-, second- and third-person perspectives, combined to give a more comprehensive picture of the offence and its putative causes. Finally, practitioners are provided with some sort of template that specifies the structure of the interview, with spaces to fill and boxes to tick, and, perhaps most importantly, lists of DRFs (in some cases alongside PFs) to identify as present (and relevant) or not. In the case of structured risk assessment, this is often accompanied by a manual specifying what should be considered evidence that a particular factor is present, details concerning how to reach a final score, and what this score actually means in terms of risk level and targets for intervention. Thus assessments of various types are informed by multiple sources and perspectives, but the output is most often a list of DRFs identified, and recommendations for action to address these. This prescriptive approach can result in case formulations that look incredibly similar across individuals, as practitioners attempt to fit individuals’ experiences within a well-established set of DRFs or ‘rehabilitative needs’ (Ward & Fortune, 2016).

An important factual element of these professional judgements concerns the type of illegal practice concerned and the individuals’ responsibility for the crime. This information is provided by first kind of source, third-person descriptions of the crime and judgements of guilt, often by police or the courts. Although the individual may deny their crime (i.e. conflict between sources), it is not the role of the forensic practitioner to determine guilt; the assumption is that the individual has committed a crime. The rest of the judgements are made by the practitioner, based upon other information given (i.e. self-report of the offence), as well as their understanding of the causes of offending. For example, did he intend for this crime to occur? Did he plan it? Why does he think he did it? How does he feel about it now? Does he think it was wrong? Will he act differently in the future? Perhaps the most important judgement is about the personal characteristics or circumstances of the individual that caused his behaviour, and as such should be corrected in order to stop it from happening again. These judgements are prompted by lists of DRFs, which can be thought of as facts about the problems observed within and experienced by large groups of individuals who have committed an offence. The practitioner, possibly in collaboration with the individual concerned, makes professional judgements about the presence or absence of each factor based upon their understanding of the DRF, whether it applies to the individual, and whether it could plausibly cause the offence. Often the assumption is that ‘problems’ experienced by large groups of people who have broken the law in general (or a specific type of law) not only are relevant to this offence, but also have caused or contributed to it.

What Figure 1 is intended to illustrate is that there are a number of influences upon professional judgements that are largely implicit and rarely acknowledged. Throughout assessments practitioners identify the contexts, subjective psychological states, actions and outcomes that constitute the illegal practice. However, this is only part of the picture; there are at least two valuable sources of information missing. Firstly, the wider context contains overarching values (i.e. what is right and wrong, healthy and harmful) and more concrete norms (i.e. what should people do or not do), which influence judgements about the nature of offending-related practices. We argue that in order to understand offending fully we must look at the valued outcomes toward which it is directed (i.e. what normative tasks are they aimed at?), as well as the collective values/norms they violate (i.e. do they cause harm?). For example, the sexual abuse of a child may be aimed at the experience of emotional connection (i.e. an intimacy-related practice), but it violates collective values and norms concerning sex and relationships, child and adolescent development, vulnerability, maturity and capacity to consent. Secondly, while the identification of DRFs points towards broad domains of functioning, it stops short of describing the capacities that they contain and explaining the mechanisms that underpin (i.e. cause) these (Ward, 2017; Ward & Fortune, 2016). To make this clearer, values and norms provide the context within which practices occur (i.e. standards determining their success/appropriateness), and psychological mechanisms cause or enable certain practices within this context. Awareness of the scope of information informing these judgements is a preliminary step in the right direction when it comes to understanding the causes of illegal practices. We argue that this acknowledgement, paired with a deeper understanding of human functioning and the mechanisms underpinning practices, will lead to more comprehensive and individualised case conceptualisations.¹

Dynamic risk and protective factors for sexual offending

There are a number of DRFs and PFs frequently identified in the sexual offending literature. These are labelled slightly differently depending on the source, but converge within the same set of functional domains. We have chosen to organise these well-established DRFs and PFs into the following five domains: sexual, interpersonal, emotion management, self-regulation and attitudes. It is acknowledged that there are other ways to categorise and label these factors, but due to significant overlap and heterogeneity within categories, we settled on the above labels as the most useful way to link DRFs and PFs with normative practices. Each domain contains a number of specific factors, which are grouped together under an umbrella term (see Table 1). These domains are consistent with assessment tools and literature concerning sexual offending, but the specific wording and examples used below are mainly borrowed from recent work concerning the DRFs with the most empirical support, and those considered most likely to be causes of offending. Mann, Hanson, and Thornton (2010) outlined a core set of these DRFs for sexual recidivism, and conceptualised these as potential ‘psychologically meaningful risk factors’. This means that they are both empirically supported and considered plausible causes of sexual offending; however, Mann et al. (2010) acknowledge that significant work remains in determining their causal status.

Table 1.

. Domains of risk and protection and their associated practices.

Domain	Risk factors	Protective	Examples of practices	Examples of capacities	Examples of values
Sexual	Preoccupation Deviant arousal/sexual interest	Moderate intensity sexual drive Sexual preference consenting adults	Sexual acts, emotional intimacy and communication, seeking sex, frequency/number of partners	Sexual drive, arousal, attraction (preference), acceptance of sexual identity, sexual scripts/schema	Pleasure, reproduction, safety, connection, mastery, inner peace, fidelity
Intimacy	Emotional congruence with children Lack of emotional intimacy adults Lacks concern Negative social influences	Preference for intimacy adults Capacity for lasting emotional intimacy Secure attachment Care and concern Social network External control	Partner choice, communication, emotional connection, establishing and maintaining bonds Peer choice, interests, influence	Interpersonal skills, preferences for intimacy, capacity for sexual and emotional connection Interpersonal skills	Pleasure, connection, relatedness, fairness, honesty and fidelity, harmony, support, mastery Support, friendship
Self-regulation	Lifestyle impulsivity Resistance to rules/supervision General self-regulation problems Poor cognitive problem solving	Goal-directed living Self-control Accept rules Effective problem solving skills Intact cognitive functioning Intelligence	Seeking employment, managing finances, leisure activities, self-care, compliance with rules/law, problem solving, planning and goal setting	Motivation, self-control, skills relevant to context (e.g. attitudes, conflict resolution)	Autonomy/agency, mastery/success, creativity, safety/stability, contribution
Emotion management	Dysfunctional coping Grievance, hostility	Functional coping	Identification, tolerance and communication of emotions, coping	Coping, emotion recognition, interpretation, control	Inner peace, comfort, pleasure, health
Attitudes	Offence-supportive attitudes Machiavellianism Hostility toward women	Attitudes support respectful age-appropriate sex Recognise others’ rights Adaptive schema Trustful/forgiving Positive attitudes toward women Motivated/optimistic toward desistence	Representing reality, causal reasoning, interpretation/attribution, explaining and justifying action	Memory, causal reasoning Theory of mind, flexibility, interpreting input, accuracy (i.e. based on evidence)	Knowledge accuracy/utility, predictability, creativity, mastery/success

Note: The examples of practices, capacities and values listed are intended to illustrate the kinds of practices impacted by each domain and the values and capacities linked with each; they are not assumed to be fixed within a certain domain or by any means exhaustive.

Subsequently, de Vries Robbé et al. (2015) put forward a list of plausible PFs for sexual offending, as well as discussing the various conceptualisations and functions of PFs (e.g. bi/unipolar, propensities and manifestations, internal and contextual, etc.). Due to the lack of empirical research on PFs, their list consists of the healthy poles of psychologically meaningful DRFs (Mann et al., 2010), factors related to desistence and PF assessment tool items (i.e. statistically linked with reductions in risk). They present their list of PFs within eight domains, which are able to be loosely linked with the Mann et al. (2010) DRFs, and which we have collapsed into the five risk domains in Table 1. While we acknowledge that the category label PF is not well defined, and that it suffers from the same conceptual limitations as DRF (see Fortune & Ward, 2017; Heffernan & Ward, 2017), we stick to these labels in our analysis due to their widespread use in the field. It may be useful from this point on to consider DRFs and PFs as broad categories that contain a mixture of descriptions of markers (i.e. predict more or less likelihood of reoffending), and rudimentary explanations that appeal to personal characteristics and contexts that could plausibly cause some of the variation observed in offending trajectories. In other words, we define DRFs and PFs as: capacities (i.e. putative causes), their manifestation (i.e. practices) and contexts that are associated with increases or decreases in rates of reoffending.

We now explore two DRF and PF pairs: those influencing intimacy and self-regulation practices. We lack space here to focus systematically on all five domains, but suggest that sexual functioning, emotion management and attitudes can be broken down in a similar way. For each domain we present a case vignette and outline the normative practices that subsequent risk-based judgements rely on and the values that underpin these judgements. In other words, we spell out the assumptions concerning what each type of practice should look like in pro-social, non-offending individuals, and therefore how deviance is identified and labelled as a DRF in this context. This will allow us to describe each type of practice in terms of its necessary capacities, each spanning a normative range of functioning, as well as identify deviant behaviour. Our expectation is that the capacities underlying the practices that DRFs and PFs refer to exist upon these continuums, and that their location is ecologically sensitive and dynamic, rather than being fixed.

Intimacy-related practices

Case vignette: Sam has never had a long-term emotionally intimate relationship, he has had several short-term dating relationships with women, and other purely sexual encounters. When in a dating relationship he has trouble committing to his partner and sharing his thoughts and feelings, eventually driving them away by either ceasing contact or seeking out other women. When asked about past relationships he tends to place the blame on his partners for being ‘cold’, ‘judgemental’, or ‘manipulative’. He states that relationships are a waste of time and that he is happy to play the field, but also reflects that he should settle down at some point (he is almost 35 years old) and find a ‘good woman’ who is worthy of his trust and attention. He would like to have children, stating that they fun and easy to be around. Sam was recently at a party where he failed to strike up a conversation with any of the adult women present. Feeling rejected, he decided to have a few drinks, and ignore the other adults altogether. He went outside where his friend’s 14-year-old daughter was smoking, he decided to join her and while the conversation began in a friendly manner, it ended with him sexually assaulting her.

Intimacy practices are composed of one’s interpersonal attitudes and skills, preferences for partners (i.e. objects of attraction), the nature of these relationships (i.e. the practices involved) and the contexts they occur within. Intimacy requires a certain level of honesty and trust, acceptance, commitment and companionship, emotional connection and sex and/or physical connection, as well as mutual support and caring (Fletcher et al., 2013). The aspects of a relationship that can be considered healthy or not include the number and length (stability) and their quality in terms of agreement and conflict. There is general consensus that the object of desire within intimate romantic relationships should be an age-appropriate partner who is able to consent to and participate equally in the relationship. Thus healthy interpersonal functioning also relies on accurate expectations of and beliefs about different types of relationship. For example, Sam’s beliefs about women and his own entitlement are likely to cause problems in the context of romantic relationships with women; these problems manifest in short-term relationships characterised by conflict or emotional avoidance.

Mann et al. (2010) identified a lack of emotionally intimate relationships with adults as a potentially ‘psychologically meaningful’ DRF. A ‘lack of’ can mean both the absence of enduring relationships and relationships involving ‘repeated conflict and/or infidelity’ (Mann et al., 2010). These definitions highlight the fact that healthy intimate relationships between adults have relatively low levels of conflict and are monogamous; however, without an understanding of what ‘normal’ conflict should look like it is difficult to make a distinction. Similarly, a widely supported DRF for sexual offending is emotional congruence with children (Mann et al., 2010). Emotional congruence involves feeling that relationships with children are more satisfying, feeling that children are easier to relate to or more understanding than adults, and possibly identifying with being a child, for example being emotionally immature. This feature is not evident for all men who have committed sexual offences, but those that do report an emotional congruence with children often speak about their offending as if it occurred within the context of a reciprocal intimate relationship (Mann et al., 2010). For example, Sam’s expectation that children are less judgemental than women played a part in his decision to approach the girl at the party.

Relatedly, de Vries Robbé et al. (2015) describe the healthy pole of intimate interpersonal functioning as a preference for and capacity to have enduring emotional intimacy with adults. They describe this capacity and preference as ‘a propensity to form and maintain emotionally close and satisfying relationships with other adults’ (de Vries Robbé et al., 2015, p. 26). Associated PFs include: a trustful and forgiving orientation, positive attitudes towards women, honest and respectful attitudes, and care and concern for others (de Vries Robbé et al., 2015). These personal attributes or beliefs seem necessary (or at least beneficial) in the practices of seeking prospective romantic partners, dating and maintaining a relationship long term. In addition, secure attachment in childhood is considered a developmental PF for both sexual and general offending (de Vries Robbé et al., 2015); theory has long been concerned with the existence of attachment problems and how this might lead to sexual offending for some individuals (e.g. Beech & Mitchell, 2016; Ward, Hudson, & Marshall, 1996). This highlights the developmental aspect of this domain, as we learn about interpersonal functioning and intimacy from our early experiences, both in bonding with caregivers and in observing others’ relationships. Perhaps Sam’s beliefs about intimacy were formed within the context of cold or harsh parenting styles, or earlier relationship experiences that left him feeling rejected and worthless.

Practitioners routinely undertake the difficult task of measuring intimacy preferences and capacities at a single point in time, or detecting changes in these after completion of treatment programmes. Judgements concerning whether intimacy-related practices are indicative of higher or lower risk of sexual reoffending often rely upon the manifestation of capacities within the context of a long-term stable relationship. For example, risk assessment tools often give a maximum risk score for no current relationship, intermediate scores for co-habiting relationships with problems (significant enough to cause concern to either party) or for a current stable dating relationship, and a score of zero if the individual is currently living with a partner without obvious problems (e.g. the Sex Offender Need Assessment Rating (SONAR); Hanson & Harris, 2001). Sometimes these judgements are paired with arbitrary time frames (i.e. living with partner for two years or more), which may or may not be normative for certain generations and cultural groups.

In summary, norms regulating the practices constituting intimate relationships specify what a healthy (and appropriate) relationship should look like: intimate, romantic relationships should only occur between consenting adults, ought to be reciprocal, should include personal disclosure of fears and needs, ought to incorporate caregiving and sexual components, should contain shared activities and responsibilities, and so on. The terms in italics indicate the normative and value-based elements. Examples of values central to intimacy-related practices include; relatedness, connection, pleasure and the nature of romantic love and relationships (i.e. it is good to be sexually and emotionally intimate with one person for a long time, and without too much conflict). Sam appears to be experiencing problems with both preference and capacity for intimacy with adults. He is fearful and avoidant of emotional intimacy with adult women (although he is sexually attracted to them), and he lacks the interpersonal attitudes necessary to engage in and commit to this type of relationship. Perhaps this is due to problems with inferring the mental states of women (i.e. they are unknowable or dishonest), or his expectations of intimate relationships.

We suggest that the core human capacities that enable engagement in healthy intimacy practices include (but are not limited to): interpersonal skills and self-regulation, attitudes towards the self, others, and relationships (e.g. social roles and expectations), preferences for romantic and sexual relationships, and capacity/desire for emotional intimacy. In this sense, intimacy-related practices rely heavily on influences from the other four domains of functioning. Impairments in any of these capacities can cause problems in intimate relationships, resulting in high levels of conflict, dishonesty and/or avoidance of intimacy. In some cases, these impairments and their outcomes (e.g. distress, loneliness, and rejection) can lead an individual to sexually offend against a child in order to meet intimacy needs. However, in other cases they may lead to other more or less healthy coping responses (e.g. dishonesty, promiscuity, substance abuse, self-harm, violence). In other words, many individuals experience problems in seeking and maintaining intimacy, but most do not consider sexual contact with children to be a viable strategy. It is necessary to look to other domains of functioning, as well as offending contexts, to construct a comprehensive explanation of the causes of sexual offending.

Self-regulation practices

Case vignette: Tom is 40 years old, married, and works as a tennis coach with early adolescents. In his mid-20s, and following a sports injury which ended his tennis career, Tom had a problem with gambling and abusing prescription medication, resulting in a large debt and conflict in his marriage. His problems with attaining long-term employment and a tendency to obtain loans for impulsive purchases caused marital conflict. Despite this, things had been generally improving for him until his wife got a new job and began spending long periods of time away from home. Feeling bored and lonely, he began a friendship with a young female who he coached. She was struggling at home, and he enjoyed feeling like he was helping with advice and support, he gave her his phone number in case she needed anything. One evening after having a fight with his wife, the girl called him upset and he picked her up from her house. They drove to a park where they sat in the car and talked for a while, he kissed her and touched her breasts against her will. She asked to go home and he became angry, accusing her of leading him on. He dropped her off and they continued the conversation via text message. Her mother saw the text messages and contacted Tom’s boss. When interviewed about the incident later he reported that he hadn’t thought about her like that before but that he ‘couldn’t stop himself’ once he started. He stated that he was upset after the argument with his wife, and that he thought she understood and wanted it.

Self-regulation has long been considered relevant to sexual offending; this is reflected in theory and practice targeting (temporary or enduring) deficits in self-regulation abilities (e.g. multimodal self-regulation theory of sexual offending, Stinson, Sales, & Becker, 2008; the self-regulation model, Ward, Hudson, & Keenan, 1998). Self-regulation is ‘the ability to modulate emotions, thoughts, interactions, and behaviours effectively’ (emphasis ours, Stinson, Becker, & McVay, 2016, p. 103), and as such encompasses all five domains of functioning. Self-regulation is primarily concerned with self-control (i.e. behavioural inhibition), problem solving, planning and goal-directed action. It is essentially related to the capacity for agency: individuals’ ability to intentionally engage with their environment in order to meet their needs. An ability to regulate behaviour (including cognition) is reflected within everyday practices and environments. The choices made, as well as the opportunities afforded to individuals, largely shape their lifestyles and personal identities. Norms concerning the sorts of life people should have include central features such as employment and leisure activities, accommodation, relationships and meaningful participation in society (i.e. engaging in normative practices). In this sense, ‘good’ self-regulation is the ability to live in accordance with the norms and expectations of others, as well as meeting one’s own subjective needs. This involves complying with rules and laws (e.g. not having sex with children), as well as norms (e.g. having a job, a home, relationships), and personal values (e.g. academic, sporting, or vocational success). Thus, whether or not an individual is judged to have adequate ability to self-regulate is based upon values spanning multiple levels: ethical, social and cultural, as well as personal preferences. Perhaps the most relevant aspect of self-regulation for offending is the temporary or enduring ability (and desire) to comply with dominant ethical and cultural values and norms based upon the perceived harmfulness of certain practices.

This ability to comply is reflected in the identified correlates of sexual reoffending under the umbrella term self-regulation deficits. DRFs relevant to this domain include lifestyle impulsiveness and poor cognitive problem solving (Mann et al., 2010). Mann et al. (2010) describe lifestyle impulsiveness as low self-control, instability (e.g. employment and accommodation), irresponsible decisions, lack of meaningful daily structure and problems with long-term goals (e.g. limited, unrealistic). These underlying problems can manifest in a parasitic or chaotic lifestyle, including interpersonal conflict, rule/law breaking, substance use, unemployment and lifestyle instability generally. Unsurprisingly, this DRF is indicative of increased likelihood of reoffending generally, and is probably more applicable to some types of sexual offence than others (i.e. some are well thought out and planned, whereas others are more opportunistic; Smallbone & Cale, 2016). In our example, Tom did not seem to experience lifestyle instability at the time of his offence, as he was employed and had managed to maintain a stable relationship. However, his past problems (e.g. gambling, poor financial management) suggest a propensity to engage in impulsive behaviour in certain states or contexts.

Poor cognitive problem solving involves impairment in generating and selecting appropriate or effective solutions to life’s problems (e.g. interpersonal conflict, financial hardship, unemployment), as well as making everyday decisions that affect the future. Examples of poor problem solving include avoidance, rumination, poor option selection (e.g. not considering probable negative consequences), lack of creativity (i.e. a limited selection of solutions to choose from) and an inability to recognise and conceptualise problems accurately as they arise. In addition, an individual may experience difficulties in problem solving when experiencing distressing emotions or while intoxicated, even if problem-solving skills are effective in general (Ward, Hudson & Marshall, 1995). This highlights the flexibility and ecological sensitivity of self-regulation ability; problems can reflect enduring deficits or temporary impairments. In the case of Tom above, he seems to have some trouble regulating his behaviour across different situations, as seen in his past employment instability and financial problems. In addition, when in a state of distress, it is likely that it would be more difficult for him to control his behaviour. De Vries Robbé et al. (2015) list goal-directed living and good problem solving as two of their proposed protective domains. Another two examples, being engaged in employment and/or constructive leisure activities and sobriety, seem to be manifestations of the ability to regulate behaviour effectively in various contexts. Additional factors listed within these domains include: self-control, enhanced sense of personal agency, stronger internal locus of control, living circumstances, financial management, life goals, intelligence and coping (de Vries Robbé et al., 2015). These are in line with the themes above, exercising control or agency, being capable of goal-directed action, and making rational (in terms of being consistent with one’s attitudes and goals) decisions – both in daily life and in demanding or problematic situations. The self-regulation domain covers personal capacities and contexts that support a range of normative practices, as well as the practices themselves, which are manifestations of these. For example, having goals, good problem-solving skills, a sense of autonomy, self-control, intelligence and coping skills facilitate goal-directed living, sobriety, employment, financial management, and so on. Having a balanced, structured, and healthy lifestyle is a manifestation of the capacity for normative goal-directed behaviour (agency) in various contexts (i.e. those that support or offer opportunities to meet goals). Having stable employment and accommodation, social support and living a ‘good life’ (i.e. successful, meaningful) are all dependent on one’s capacity to delay gratification, follow rules and solve any problems that arise in the pursuit of goals.

The second of these PFs, good problem solving, is defined as a capacity for managing day to day problems as they arise, without becoming emotionally overwhelmed and resorting to unhelpful behaviours. It involves complex cognitive tasks such as combining and evaluating various sources of evidence, considering competing viewpoints, generating numerous possible courses of action and evaluating the expected consequences of those actions in order to select the best option. It has been suggested that these evaluations are based on three types of expectation: rewards, norms and competence (Fishbein & Ajzen, 2010). In other words, persons select the most attractive option based upon expected positive outcomes, how other people are likely to perceive behaviour, and their confidence in their ability to be successful. It is likely that this process is supported by healthy coping mechanisms and self-control (i.e. not reacting emotionally), particularly when emotionally distressed. In the example of Tom above, the presence of healthy coping strategies or the ability to form accurate expectations about the outcomes of his behaviour could have prevented him from picking up his student.

In summary, the practice of self-regulation is primarily concerned with engaging with one’s context and responding to situations in a way that is aligned with one’s goals, values and intentions, as well as the norms and expectations of society overall. There is an expectation that individuals will obey the law and rules more generally (i.e. behaving in a way that is in line with dominant values, not causing harm to others), as well as engaging in normative activities based upon shared values (i.e. having long-term goals, participating in society). Values within these normative practices include: mastery of professional and leisure activities, achievement and success, as well as safety and stability in society. People ought to be engaged in meaningful employment and leisure activities, to have goals and be disciplined in working towards these, and to live mainly within the boundaries of the law – to control their behaviour and do what is expected of them by society. These expectations spell out what it means to be a valuable and productive member of society, and they are reliant upon both personal capacities and the presence of opportunities within one’s environment. In other words, persons have varied (internal and external) resources with which to engage in normative and personally meaningful self-regulation practices (Ward & Maruna, 2007).

The internal capacities required for self-regulation include (but again are not limited to): motivation, action selection, self-control (e.g. inhibiting unhelpful responses) and particular skills relevant to the goal or value in question (e.g. attitudes, conflict resolution, conscientiousness, etc.). In a sense this domain represents the intersection of the others (see Table 1), as attitudes, emotion management, sexual functioning and interpersonal skills all come into play within the process of various types of goal-directed practice. These practices can be more or less congruent with one’s long-term goals and over-arching values, and may be judged by others as appropriate or not. It is important to note that the relationship between self-regulation capacities and sexual offending is complex. Self-regulation capacities can be enduring and relatively stable, but they can also be suspended in certain states (e.g. intoxication, stress), and they can be used to meet needs in various ways. Good self-regulation does not ensure prosocial behaviour and vice versa. At one time or another most people will act in ways that are not in accordance with their own goals and values or those of their cultural group, but usually the outcome is not a sexual offence.

Implications of a ‘practices’ approach to risk

An important task when engaging in theoretical exploration is to consider how changes or additions to our understanding might have an impact on people in the real world. In a forensic context we are concerned with how our understanding of DRFs and PFs influences the assessment, management and treatment of individuals who have committed offences. Firstly, we propose that simply acknowledging the extent to which DRFs and PFs are value laden is a step in the right direction, and that this is not currently evident in routine practice. In addition, conceptualising risk assessment, case management and treatment targets in terms of the goal-directed practices that they refer to, and the values and norms that form the context for these practices and their evaluation, may open fruitful avenues for new developments in the process and content of interventions. For example, rather than simply concluding that a participant needs to address their ‘offence-supportive attitudes’ (e.g. children are not harmed by sex), practitioners can explore the values or goals underpinning these statements (e.g. inner peace, avoiding cognitive dissonance, pleasure), the norms that they violate (e.g. age of consent, the nature of harm), and the practices in which they manifest in (e.g. intimacy seeking, offending, explaining/justifying). These broader considerations provide additional information about the context within which harmful practices occur, the function of the particular feature in the individual’s life, and eventually a more complete understanding of the capacities that are required for less harmful practices. This can help practitioners to ascertain whether and how a particular DRF or PF is likely to increase or decrease risk of sexual reoffending for an individual, and in which contexts. The implications of this include more person-centred and holistic assessment, case conceptualisation, treatment and management of individuals who have committed sexual offences.

According to the view outlined in this article, treatment should focus on restoring (or enabling) the individual and their environment to a healthy (personally meaningful) level with respect to key areas of functioning, and the goal-directed practices which they underpin. Therefore, an important task for individualised treatment is to locate the source/s of impairments. For example, it may be that intimacy-seeking strategies are normative, but that they have been directed towards an inappropriate target (i.e. a child) so they are labelled as ‘grooming’ rather than being seen as adaptive intimacy practices. In other cases, the individual might be directing their attention towards appropriate targets generally, but in certain states or contexts they might act in ways that are inconsistent with their beliefs and values. In the first example we see a clash in values and transgression of relationship norms, where the individual views children as viable partners although this perception is not shared by the rest of society. In the second example we see impairments in self-regulation and control; the problem seems to be acting in ways that are incongruent with one’s own values (as well as being against the law). The targets for treatment will be different in each case. For example, exploring the discrepancies between certain value systems, versus strengthening capacities required for effective self-regulation and coping with distress.

Another consideration relevant to forensic practice is the theoretical and functional relationship between DRFs and PFs. There are questions concerning whether DRF and PF are distinct constructs, and also whether or how they could co-exist and interact within individuals (de Vries Robbé et al., 2015; Heffernan & Ward, 2017; Polaschek, 2017; Ward, 2017). Our conceptualisation of DRFs and PFs as markers or aspects of practices that are more or less harmful and illegal suggests that they are different aspects of the same things. These ‘things’ are the domains of functioning that contain human capacities and underlying causal mechanisms, as well as the contexts that support or obstruct goal-directed practices. In other words, DRF are aspects of emotion, attitudes, relationships, sexuality and self-regulatory practices that often accompany offending behaviour. Relatedly, PFs are often observed alongside a shift towards more pro-social practices (i.e. the process of desistence). We propose that due to the vague and composite nature of both constructs, it does not make sense to talk about DRFs and PFs interacting within individuals. However, once they are broken down into the capacities required for healthy practices in various domains of functioning, some of these may plausibly interact (i.e. compensate for or exacerbate other areas of weakness) or at least co-exist. Our argument here is that before we can reason about the possible functions of DRFs and PFs, we need to add depth to our understanding and explanations of these concepts by appealing to human capacities and psychological mechanisms. In addition, we need to understand these within the context of the values, norms and practices that they refer to, and to be aware of how practitioners’ own perspectives shape judgements concerning risk.

Of concern for dynamic risk assessment is the subjectivity of ratings based upon practitioners’ judgements concerning self-report and behavioural information (Cording, Beggs Christofferson, & Grace, 2016). This is of equal relevance for treatment planning that is based upon assessments of risk/need. For example, despite the use of structured assessment manuals, there is room for disagreement in whether – and to what degree – a factor such as ‘impulsivity’ is present. The default assumption is that risk factors are present, which results in homogeneity of risk profiles (i.e. all or most possible risk factors are identified) and little information about how to address the causes of sexual offending (i.e. impairments in capacities and the contexts within which they manifest). A better understanding of the practices that DRFs and PFs refer to has the potential to facilitate theoretically informed explanations that apply to the individual and their values, rather than relying solely upon lists of correlates derived from aggregate data. A fuller picture of the influences upon these practices can guide interviews, help practitioners complete case formulations based upon shared and divergent values, and facilitate treatment, which strengthens the capacity for practices that do not harm others.

One way for practitioners to recognise potential biases prospectively is to consider where they sit with respect to relevant values – for example, harm, consent, monogamy, respect, health, relatedness, sexual pleasure, success, and so on. Also relevant is where other practitioners who have supplied their own professional opinions sit; for example, colleagues may vary to the degree to which they share personal and professional standards (see Ward, 2013). Perhaps most important when working with individuals is the ability to recognise any discrepancies between one’s own values and normative standards, and those of the individual being assessed, treated or managed by the justice system. If practitioners fail to consider the preferences and perspectives of the people they work with, it is less likely that they will be able to engage them in treatment to reduce the likelihood of further offending (Ward & Brown, 2004). In addition, by encouraging participants to identify their own values and sets of relevant norms, and to consider how they manifest in harmful or unhealthy practices, practitioners may engage participants in collaborative treatment where they are the experts in their own lives, rather than being told that their goals (just like everyone else’s) are to reduce their ‘offence-related attitudes and emotions’. It is expected that participants would experience this sort of treatment as more engaging, and that this would be reflected within their motivation to complete treatment and remain offence free in the future (Ward, Melser, & Yates, 2007).

Conclusions

We have argued that value-based judgements (i.e. the identification of DRFs and PFs) should take into account the relevant social/cultural contexts (i.e. normative practices and their perceived success) if they are to inform explanations of offending and the reduction or management of risk. We have demonstrated that two DRF and PF domains, intimacy and self-regulation, contain goal-directed practices governed by integrated sets of norms and values. Values drill all the way down to everyday behaviours and situations, including sexual offending. Norms concerning viable romantic relationships, healthy emotional functioning, adaptive attitudes, appropriate sexual interests and behaviour, and adequate self-regulation are derived from biological, social and cultural models and ideals. It is to be expected that culturally diverse groups (i.e. ethnicities, gangs, sub-cultures) are likely to endorse different practices and normative standards, and that such differences ought to be taken into account when (a) developing risk assessment and psychological assessment protocols, and (b) constructing treatment programmes. One advantage to conceptualising DRFs and PFs in terms of practices is the ability to locate offending behaviours within their cultural context, and to suggest ways of achieving valued but legal outcomes. The process of correctional rehabilitation then becomes one of restoring or developing healthy human functioning by strengthening capacities. It is a process of learning and building strengths, rather than simply punishment and risk management. De Vries Robbé et al. (2015) suggest that ‘we know very little about what those who have offended sexually value, what makes them happy, and what skills and strengths are related to their desistence from offending’ (p. 30). In our opinion, an important potential outcome of a shift in focus towards adaptive practices and their necessary capacities is that it may prompt further research into the origins and composition of these strengths. It also highlights the fact that the prudential values motivating offending practices are likely to be the same as those of non-offending individuals (e.g. pleasure, relatedness; Ward & Maruna, 2007). This is exciting for both forensic and clinical practice, because it means that practitioners can identify and ameliorate impairments in a strength-based way.

In summary, whether a particular feature of an individual or their environment is related to their risk of recidivism is partly a value judgement. These judgements are based upon facts (e.g. the occurrence of violent behaviour, substance use), inferences about that individual’s character/capacities (e.g. impulsivity, attitudes, personality) and norms of society overall (e.g. does it fit with the dominant values? Is it healthy? Does it cause harm?). Furthermore, identifying a DRF to target in treatment involves assigning it causal status, concluding that it increases an individual’s likelihood of offending and needs to be corrected. DRFs and PFs are defined in relation to the practices that they manifest within – various types of offending and associated behaviours. These practices are goal directed (i.e. aimed at obtaining prudential values) and evaluated against social/cultural values and norms. Therefore, we cannot explain DRFs and PFs without reference to the capacities that underpin practices, the goals they are directed towards and the norms and values against which they are evaluated. We will have a better chance of understanding offending behaviours if we look beyond surface-level descriptions of ‘antisocial’ characteristics and contexts, and instead see the whole person, their values and the normative context within which they strive to grasp these.

Declaration of conflicts of interest

Roxanne Heffernan has declared no conflicts of interest

Tony Ward has declared no conflicts of interest

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.