Adversity

For children, ‘adversity’ has been defined as ‘the experience of life events and circumstances which may combine to threaten or challenge healthy development’ (Daniel et al., Reference Daniel, Wassell and Gilligan1999). The assumption underlying this statement is that children's developmental pathways can be adversely affected by a range of circumstances. The definition of these adverse factors can range from the narrow to the very broad including:

• • incidents of physical or sexual abuse,

• • traumatic incidents such as loss and bereavement,

• • chronic situations – such as environments of neglect,

• • experiences of bullying or racism,

• • family stressors,

• • structural inequalities and socio-economic disadvantages.

The current ‘child protection’ system in the UK, and in countries with similar welfare traditions, developed through the 1980s and 1990s. It shifted from one shaped by a ‘socio-medical’ discourse in which abuse was ‘diagnosed’ and therapeutic support was offered, towards a ‘socio-legal’ discourse with an emphasis upon the need for a statutory basis for intervention in family life and a forensic approach to the ‘investigation’ of allegations of abuse and neglect (Parton, Reference Parton, Burns and Lynch2008). The operation of the system was shaped by guidance that focused on fairly narrow definitions of the kind of adversity that could legitimate state intervention into family life – thus in England the categories used to shape multi-disciplinary decisions are physical abuse, emotional abuse, sexual abuse and neglect (Department of Health and Department for Education and Employment, 2000) and in Scotland they are physical injury, sexual abuse, non-organic failure to thrive, emotional abuse and physical neglect (Scottish Executive, 2000).

Dissatisfaction with, and challenges to, these narrow definitions come from different directions. They include, for example, concerns that the system focuses too much on parental acts of commission or omission and deflects attention away from adversity resulting from wider state policy associated with socio-economic inequality and high levels of child poverty (Baldwin and Spencer, Reference Baldwin, Spencer, Taylor and Daniel2005). Such arguments are not so much a direct challenge to the operation of the child protection system within the narrow field of adversity defined for it, as a challenge to the assumption that it is a sufficient response to the needs of children. A more direct challenge to the operation of the ‘child protection system’ comes from bodies of research into the impact upon children's development of a wider range of domestic adversity. These include attention to the impact of male violence against women upon the children in a household (Hester et al., Reference Hester, Pearson and Harwin2000; Mullender, Reference Mullender1996). This body of work has been sufficiently influential such that within the recent series of Lancet papers on maltreatment, ‘intimate partner violence’ is included as a separate category along with physical, sexual, psychological or emotional abuse and neglect.Footnote ¹ Research into the impact of emotional deprivation and chronic neglect upon brain and other physiological development has also been influential (Perry, Reference Perry and Osofsky1997; Glaser, Reference Glaser2000). It is perhaps a testament to the power of our positivist scientific tradition that the evidence of physiological harm appears to lend more ‘weight’ to arguments for intervention than evidence of emotional harm. There has also been research into the impact upon children of parental substance misuse (Barnard and Barlow, Reference Barnard and Barlow2003), parental mental ill-health (Falkov, Reference Falkov1996) or combinations of such factors (Cleaver, Reference Cleaver1999). Perhaps because they stem from the domestic arena, all these identified factors have acted to ‘stretch’ the concepts of adversity that are traditionally called upon within the child protection system; or, seen from another angle, these adversities have been pushed and pulled so that they can ‘fit’ with the categories above − often ‘emotional abuse’ appears to be the most convenient category.

Broader understandings of the adversities that impact upon children also imply different understandings of concepts of ‘victim’ and ‘perpetrator’. For example, in situations of domestic abuse the perpetrator of violence to the mother is also rendered a perpetrator of child abuse and the child becomes a ‘secondary’ victim of a crime. If the focus is upon a child's unmet needs, then attention is on parental failure, usually constructed as maternal failure (Swift, Reference Swift1995; Scourfield, Reference Scourfield2003; Daniel et al., Reference Daniel, Featherstone, Hooper and Scourfield2005). A focus on deprivation opens the potential for the state to be viewed as the perpetrator and children to be victims of state policies that fail to effectively tackle child poverty. All are a far step from the concept of the predatory sex abuser, which is the more familiar narrative of ‘perpetrator’.

A more recent phenomenon is the re-defining of the ‘child protection’ system with a view to it being able to accommodate a wider range of adversities. As the range of identified adversities has widened, the focus of concern has broadened and become generalised under the umbrella of ‘unmet needs’ (DoH, 2000; DFES, 2004). Parton (Reference Parton, Wilson and James2007, Reference Parton, Burns and Lynch2008) has undertaken a detailed analysis of New Labour's children's policies. He describes the shift in language from ‘child protection’ to ‘safeguarding’ as the focus shifts to unmet need, and he identifies the increasingly wide range of adversities it is considered that the state should be concerned about, including school exclusion, running away, involvement in crime and so on. This broadening of scope means that ‘adversity’ has spilled well beyond the operational categories defined for the ‘child protection’ system.

Risk

In the context of child protection, ‘risk’ denotes the chances of adversity translating into actual negative outcomes for children. In practice, it is used as shorthand for capturing an often ill-defined combination of issues including:

• • the chances of a child experiencing a particular adverse circumstance;

• • the chances of a child being sexually or physically abused, or re-abused;

• • the chances of the parents or carers being unable (or unwilling) to ameliorate the impact of adversity;

• • the chances of the adverse experience having a significant negative impact upon well-being during childhood and

• • the chances of longer-term negative outcomes as a result of experiencing adversity.

Like adversity, risk can also be defined narrowly or broadly:

In the narrow world of statutory ‘child protection’ activity, risk is inextricably tied in with concepts of harm, both acute forms of harm, such as physical injury, and more chronic forms, such as compromised global development as a result of neglect. It can be used both to denote the likelihood of abuse occurring and the likelihood of harm resulting from abuse.

The concept of risk brings a future focus and entails making predictions based upon weighing up probabilities. As Parton (Reference Parton, Wilson and James2007) highlights, a future focus was first explicitly encapsulated in the legislative system in England and Wales in the late 1980s with the introduction of the criterion for intervention in the Children Act 1989 as ‘that the child concerned is suffering, or is likely to suffer significant harm’ (s.31(92)(a)). Similarly, section 52(2)(c) of the Children (Scotland) Act 1995 uses the term ‘is likely . . . to be impaired in his health or development’. The word ‘likely’ has been fraught with difficulties and has been the subject of much debate. On the one hand, it is would be difficult to maintain a position that intervention on behalf of a child should be delayed until actual harm has been experienced – to prevent harm occurring has to be viewed as a decent and reasonable response (Glaser, Reference Glaser2002). Indeed, prevention and early intervention are concepts that increasingly underpin children's services policies in many countries. On the other hand, this entails the need to assess the likelihood of future ‘suffering’, which is notoriously difficult.

One response to the problem has been the development of a range of risk-assessment tools, some of which aim to incorporate a predictive element (Browne and Saqui, Reference Browne, Saqui, Browne, Hanks, Stratton and Hamilton2002). But, given that human beings are poor at gauging probabilities, Munro (Reference Munro2007) suggests that we should not expect child protection workers to be ‘fortune tellers’. She provides a detailed mathematical analysis of the accuracy (or rather, inaccuracy) of risk assessment instruments and one of the key messages from her analysis is that the rarer the event, the harder it is to predict with accuracy. The other key message is that factors associated with abuse are only moderate predictors, because of the extent to which non-abusing families share the same factors. Munro suggests that, even though actuarial methods have very poor predictive powers, they are still more useful than clinical judgement, which she suggests is even worse. The issue then hinges on the general public's tolerance for both false positives and false negatives. Some suggest that the unacceptably high levels of both false positives and false negatives in actuarial models means that their use is unhelpful (Appleton and Cowley, Reference Appleton and Cowley2004) and even unethical (Taylor et al., in press).

Organisational contexts can also shape understandings of risk, not least risks to the organisation itself and to its staff that flow from the negative consequences of apparently taking the wrong decision in respect of risk. Dalgleish's (Reference Dalgleish, Calder and Hackett2003) model for decision-making in the context of risk explicitly includes the organisational tolerance of risk as one of the factors to be taken into account. He describes the ways in which organisational tolerance can fluctuate in the light of local inquiries into child deaths and recent experiences of apparently poor decisions. Parental definitions of risk may also be very different from professional definitions (Waugh, Reference Waugh, Fawcett and Waugh2008). The risk of having a child removed may be more salient to parents than an ill-defined likelihood of harm in the future, should the child remain in their care. The range of different risks have been summarised by Cooper et al. (Reference Cooper, Hetherington and Katz2003: 15):

Munro (Reference Munro2007) explores the ways that the concept of risk has moved from one with essentially neutral connotations to one where it is associated with negative outcomes. When viewed primarily in terms of bad outcomes, ‘risk’ essentially means the risk of bad outcomes resulting from the impact of adversity upon a vulnerable child. As she points out, we do not tend to talk about the ‘risk’ of a good outcome for a child. She suggests that this association with negative outcomes distorts decision-making in relation to child protection, such that the drive to avoid ‘bad’ outcomes is inextricably linked with the assessment of ‘risk’. The focus on preventing negative outcomes means that we do not look at the aspect of promotion of well-being, rather only focus on minimising danger. Cooper et al. (Reference Cooper, Hetherington and Katz2003) also suggest that risks in human settings, such as child protection, cannot be conceptualised in the same way as in other settings. They suggest that risk management models from arenas such as accountancy have been misapplied to child protection and that the focus on diagnosing risk works against the development of therapeutic approaches that depend on relationship-building and trust.

‘Risk’ has become a term that can be defined very narrowly, but in relation to child protection is used loosely and widely to denote a range of negative outcomes for children, parents, society, professionals and organisations. Thus, it does not provide a straightforward model for adoption in adult settings. When combined with concepts of adversity and vulnerability the picture becomes even more complex.

Vulnerability

Children are treated as a discrete group for the purposes of the UN Charter of children's rights, thus signalling their need for particular attention (United Nations, 1989). Many countries have specific legislation and guidance that applies to individuals who fall within a specific age-range from birth up to a cut-off point defined as the end of ‘childhood’. The cut-off point varies according to different countries, and within countries may be different for different aspects of legislation. Nonetheless, there are protections accorded to a category considered to be ‘children’. In addition, many countries have policies that target groups of children seen to be at elevated vulnerability because of the adverse circumstances that they are experiencing. In sum, vulnerability appears simultaneously to be conceptualised broadly and narrowly with a view that all children are vulnerable, but some are more vulnerable than others.

It is the concept of children as especially ‘vulnerable’ that has legitimised the development of the forensic investigative child protection system. ‘Vulnerability’ is generally a highly contested term, and in regard to childhood in particular there are a range of possible positions about the extent to which children are conceptualised as vulnerable, merely by virtue of being children. Children may be viewed as vulnerable because of their size, their fragility when little and because they are still being ‘formed’. An adverse factor is defined thus because it potentially compromises a child's healthy development, and it is the interaction between adversity and the developmental processes that is considered to be uniquely damaging. At the extremes, vulnerability may be clear-cut: shaking an infant, for example, is more likely to inflict serious physiological damage than shaking a young adult. How, though, is vulnerability to be understood in cases where an adolescent young man physically assaults his mother? Similarly, is it possible to place the impact of sexual abuse on a hierarchy of trauma correlated with age such that it is always assumed to be more emotionally damaging to children than adults or older adults?

Perhaps the greatest challenge to the automatic assumption of childhood vulnerability comes from the sociology of childhood in which it is argued that the concept of childhood is socially constructed. Stainton Rogers, for example, suggests that ‘there is no natural distinction that marks off children as a certain category of person. Rather, the category of ‘a child’ is just that – a category’ (Stainton Rogers, Reference Stainton Rogers, Foley, Roche and Tucker2001: 27). She describes two discourses about childhood: one that romanticises childhood as a time of innocence and another that assumes that children need to be socialised and to be taught morals. The common theme of both, though, is one of concern about children with an attendant adult responsibility for children. Thus the two discourses are recast within current UK policy as a ‘discourse of welfare’ and a ‘discourse of control’, with the former running though social policy and the latter through education policy. She raises questions as to whether the conceptualisation of children as passive recipients of our concern reduces them. Practitioners operating within the child protection system frequently override the wishes of children: for example, the volume of calls to ChildLine shows the extent to which children crave the confidentiality that the statutory system does not provide. There is an interesting resonance here with the powerful statement made by Dame Butler Sloss in her report into an Inquiry into child abuse investigations in Cleveland: ‘the child should be seen as a person, not as an object of concern’ (Butler-Sloss, Reference Butler-Sloss1998: 245). The dichotomy can also be seen in the ways in which children who commit violent crimes or who are seen as a threat to social order are constructed as ‘perpetrators’ rather than ‘victims’ and as no longer being in the ‘child’ category (Piper, Reference Piper2008).

A further concern about the assumption of children as inherently vulnerable is that, in effect, it ‘problematises’ childhood (James, Reference James2008). The names of the two key policy documents in the UK – Every Child Matters (DFES, 2004) and Getting it Right for Every Child (Scottish Executive, 2005) specifically signal a concern for all children. The common principle underlying these policies is the expectation that the universal services of health and education should play a role in ensuring that the range of needs of all children are identified and met. It could be argued that these policies signal recognition of children as people in their own right who should have access to all the services they need in order to thrive. On the other hand, Parton (Reference Parton, Wilson and James2007, Reference Parton, Burns and Lynch2008) questions the direction of UK policy by suggesting that the broadening of concern to include all concerns about children's health and development is leading to increased regulation and surveillance of children as a group. He also suggests that this broadening of focus may be counter-productive by asking:

It is further suggested that our current pre-occupation with protecting children as a group is actively damaging to children. Newman and Blackburn (Reference Newman and Blackburn2002) suggest that we are pre-occupied with risk and its avoidance, to the extent that children in general now have reduced opportunities to take risks and learn importance competencies. Therefore, by trying to eliminate risk we increase the risk of other poor outcomes, including poor psychological and physical health and poor coping skills (Newman and Blackburn, Reference Newman and Blackburn2002). This argument may be influenced in part by the discourse of childhood as a time of innocence and freedom – the recent rise of concern about ‘cotton wool kids’ perhaps also captures a sense of adult concern about a lost world of freedom. Paradoxically, therefore, we can see that ‘concern’ about children as a group in need of specific attention because of their unique vulnerability can simultaneously signal a recognition that children should be accorded state protection but potentially de-humanise them and deny their role as active citizens in their own right.

Resilience

The population-based studies that link poor outcomes with abuse and neglect, coupled with the concept of children as especially vulnerable to the impact of abuse and neglect, act as powerful drivers and justification for a state protective system. However, the research into resilience poses real challenges for a system of ‘protection’ that is based upon the assumption of damage, because it shows that there are considerable individual differences in the extent of poor outcomes as a result of adversity. Resilience is defined as ‘a phenomenon or process reflecting relatively positive adaptation despite experiences of adversity or trauma’ (Luthar, Reference Luthar, Cicchetti and Cohen2005: 6).

One of the most influential studies of resilience was that of Werner who tracked the development of 698 children born in 1955 on Kauai (Werner and Smith, Reference Werner and Smith1992). A third (201) of these children were considered to be at risk of poorer developmental outcomes. By the first decade, two out of three of those considered to be at high risk were showing learning and behaviour problems and, by 18, had mental health problems, delinquency records and/or teenage pregnancies. However, a third of the children predicted to be at high risk grew into what Werner defined as ‘competent, confident and caring adults’. She and her team were particularly interested in the factors that were associated with better than predicted outcomes. For young children, school aged children and for adolescents they mapped the factors at the level of the individual, family and wider community that were associated with resilience. These include a cluster of personal characteristics such as self-efficacy, coupled with factors associated with secure attachments and relationships and resources in the school and wider community. Gilligan (Reference Gilligan1997: 12) has summed these up as:

Resilience, therefore, can be seen as an emergent property from the experience of adversity and the demonstration of relatively good outcomes. However, resilience is also a concept that is somewhat slippery, especially with regard to definitions of adversity and of good outcomes. For the purposes of empirical research, adversity is normally defined by the researchers, but different studies use different measures. Some define adversity in terms of constellations of experiences, such as poor socio-economic circumstances, poorly resourced neighbourhoods (Sameroff et al., Reference Sameroff, Gutman, Peck and Luthar2003); others focus on a particular adversity, such as parental depression (Hammen, Reference Hammen and Luthar2003) or parental divorce (Hetherington and Mitchell Elmore, Reference Hetherington, Mitchell Elmore and Luthar2003). This variation makes studies hard to compare, but it also means that studies are not routinely incorporating attention to the ways in which children and young people subjectively construct stressful events and adversity (Gore and Eckenrode, Reference Gore, Eckenrode, Haggerty, Sherrod, Garmezy and Rutter1994). One person's stressor may be another person's challenge. The concept has also been criticised as individualising what should be seen as structural adversities and as contributing to an individualistic notion that children and young people can overcome any difficulty if they try hard enough (Rigsby, Reference Rigsby, Wang and Gordons1994). Despite the criticisms, the consistent findings that different children respond differently to adversity and that negative outcomes cannot always be predicted poses challenges to a child protection system that is underpinned by risk assessment.