Introduction: Why should the concept of resilience be employed and what is the value added of such a resilience perspective?

In the social sciences, research concentrated for a long time on negative outcomes and processes, mostly employing the so-called deficit model (Boyden and Cooper, Reference Boyden and Cooper2007). The work of resilience, however, directs the attention of researchers to positive outcomes for individuals and investigates the factors for succeeding in overcoming adversities (Howard et al., Reference Howard, Dryden and Johnson1999). Luthar and Zelazo advocate this expansion of the scientific mindset and adopt an asset-focused model: the resilience perspective aids ‘scientists to adopt fresh mind sets, leading them to consider strength among groups usually thought of in terms of problems and failure’ (Luthar and Zelazo, Reference Luthar, Zelazo and Luthar2003). Engaging in resilience research further adds important aspects when examining various issues. Take for example the prevention of social exclusion; here both the work on factors that exclude groups but also factors that build up their resilience are important research areas to inform policy (Edwards, Reference Edwards2007). Such a holistic perspective is likely to result in better-designed and thus more effective policy interventions – especially prevention programmes. The shift from a deficit model may also facilitate better cooperation with targeted groups for intervention, as a positive connotation is attached – e.g. teachers and parents are often more enthusiastic when aims of the programmes are framed in an optimistic way (Howard et al., Reference Howard, Dryden and Johnson1999).

Historical development — origins of the resilience concept and how it developed

The word resilio originates from Latin and means ‘to jump back’ (Manyena, Reference Manyena2006). However, it is contested which discipline can be credited for using resilience as a concept for the first time. Some claim that the term resilience was first employed in ecology, where resilience refers to the ability of ecosystems to bounce back after major shocks (Brand and Jax, Reference Brand and Jax2007; Adger, Reference Adger2000). Others state that the concept has its origins in physics and engineering where it captures the ability of materials to ‘bounce back’ after shocks and resume their original condition. From physics, the term found its way into medicine where it was used to describe the physical recovery of patients (Boyden and Cooper, Reference Boyden and Cooper2007). However, most of the literature cites social psychology and psychiatry as the place of origin for the concept where it dates back to the 1940s (Bartley, Reference Bartley2006; Howard et al., Reference Howard, Dryden and Johnson1999; Luthar et al., Reference Luthar, Cicchetti and Becker2000; Luthar and Zelazo, Reference Luthar, Zelazo and Luthar2003; Manyena, Reference Manyena2006). Some of the most influential empirical studies were conducted by Norman Garmezy and Michael Rutter who conducted a long-term survey of 200 children in the US (Garmezy and Rutter, Reference Garmezy and Rutter1983) and by Emmy Werner and Ruth Smith who followed over 600 people in Hawaii over a time period of 30 years (Werner and Smith, Reference Werner and Smith1988).

Garmezy and Rutter more specifically studied both the coping strategies of children living with mentally ill parents (Garmezy and Rutter, Reference Garmezy and Rutter1983; Werner and Smith, Reference Werner and Smith1988; Howard et al., Reference Howard, Dryden and Johnson1999; Masten and Garmezy, Reference Masten, Garmezy, Lahey and Kazdin1985) and what protective factors prevented adverse outcomes. They found that the majority of children who experienced high adversities did not develop the expected negative outcomes (Rutter, Reference Rutter1985, Reference Rutter1987, Reference Rutter, Rolf, Masten, Cicchetti, Neuchterlein and Weintraub1990, Reference Rutter1999). So the prevalent research question was what factors and processes lead children and adults to succeed despite adversities? This early focus on children – especially in the psychology literature – is still visible today as childhood resilience is one of the most extensively researched areas in the field.

What started in the 1950s with the research on assets and strength of individuals and communities instead of risk and resilience led to a real paradigm change from risk research to resilience research in the 1980s (Gore and Eckenrode, Reference Gore, Eckenrode and Haggerty1994; Rhodes and Brown, Reference Rhodes and Brown1991; Boyden and Cooper, Reference Boyden and Cooper2007; Howard et al., Reference Howard, Dryden and Johnson1999). Furthermore, today it is employed by a wide range of scientific disciplines, from developmental psychology, ecological economics, development studies, social policy, genetics and disaster studies (Boyden and Cooper, Reference Boyden and Cooper2007; Bartley, Reference Bartley2006; Brand and Jax, Reference Brand and Jax2007; Condly, Reference Condly2006).

The research focus of resilience broadened beyond psychological adversity and risk factors and covered social and economic stressful experiences as well. Garmezy's work on poverty is a classic example (Garmezy, Reference Garmezy1991). Today, the themes in the resilience literature are widespread and it is only possible to give a short and limited overview of the diversity in the resilience research. Regarding the adversity caused by parental characteristics that children face, the problems ranges from alcohol and drug abuse (Bancroft et al., Reference Bancroft, Wilson, Cunningham-Burley, Backett-Milburn and Masters2004), parenting style (Bartley et al., Reference Bartley, Head and Stansfeld2007; Armstrong et al., Reference Armstrong, Birnie-Lefcovitch and Ungar2005), mentally ill parents (Gladstone et al., Reference Gladstone, Boydell and McKeever2006; Tiet et al., Reference Tiet, Bird, Hoven, Wu, Moore and Davies2001; Tebes et al., Reference Tebes, Kaufman, Adnopoz and Racusin2001), divorce (Greeff and Van Der Merwe, Reference Greeff and Van Der Merwe2004; Hetherington and Elmore, Reference Hetherington and Elmore2003; Sandler et al., Reference Sandler, Wolchik, Davis, Haine and Ayers2003) and socio-economic disadvantages and unemployment (Kim-Cohen et al., Reference Kim-Cohen, Moffitt, Caspi and Taylor2004; Criss et al., Reference Criss, Pettit, Bates, Dodge and Lapp2002; Blanden, Reference Blanden2006; Beck et al., Reference Beck, Wagener and Grix2005).

Another branch of literature deals with the impact of social capital, peer group relationships and neighbourhood characteristics on resilience of individuals and families (Walter-Ginzburg et al., Reference Walter-Ginzburg, Shmotkin, Blumstein and Shorek2005; Armstrong et al., Reference Armstrong, Birnie-Lefcovitch and Ungar2005; Pedersen et al., Reference Pedersen, Vitaro, Barker and Borge2007; Laursen et al., Reference Laursen, Bukowski, Aunola and Nurmi2007; Delsing et al., Reference Delsing, Bogt, Engels and Meeus2007; Michael and Ben-Zur, Reference Michael and Ben-Zur2007; Kan and McHale, Reference Kan and McHale2007; Gatti and Tremblay, Reference Gatti and Tremblay2007; Bellair, Reference Bellair1997; Pinkerton and Dolan, Reference Pinkerton and Dolan2007; Hughes and Evans, Reference Hughes and Evans2007; Criss et al., Reference Criss, Pettit, Bates, Dodge and Lapp2002).

In the area of health and education, the following topics are a selection identified: chronic illnesses (Bellin and Kovacs, Reference Bellin and Kovacs2006), child mortality (Garenne and Gakusi, Reference Garenne and Gakusi2006), mental and physical disabilities (Retzlaff et al., Reference Retzlaff, Hornig, Muller, Reuner and Pietz2006), stress (Joiner et al., Reference Joiner, Wingate, Gencoz and Gencoz2005) and poor health in old age (Walter-Ginzburg et al., Reference Walter-Ginzburg, Shmotkin, Blumstein and Shorek2005; Gattuso, Reference Gattuso2003). Learning environment (Theron, Reference Theron2006) and early school leavers and the return to education (Sackera and Schoon, Reference Sackera and Schoon2007) are further areas of research.

And, lastly, the literature on violence and disaster is increasing and issues such as war and refugees studies (McAdam-Crisp, Reference McAdam-Crisp2006; Doron, Reference Doron2005; Riolli et al., Reference Riolli, Savicki and Cepani2002; Hosin, Reference Hosin2001), rape (Edwards et al., Reference Edwards, Sakasa and Wyk2005), terrorism and its impact on mental health (Moscardino et al., Reference Moscardino, Axia, Scrimin and Capello2007; Bonanno et al., Reference Bonanno, Rennicke and Dekel2005; Bonanno et al., Reference Bonanno, Galea, Bucciarelli and Vlahov2006; Fredrickson et al., Reference Fredrickson, Tugade, Waugh and Larkin2003; Kendra and Wachtendorf, Reference Kendra and Wachtendorf2003), earthquakes (Perez-Sales et al., Reference Perez-Sales, Cervellon, Vazquez, Vidales and Gaborit2001), community violence (O'Donnell et al., Reference O'Donnell, Schwab-Stone and Muyeed2002) and discrimination of ethnic minorities (Spencer-Rodgers and Collins, Reference Spencer-Rodgers and Collins2006; Flores et al., Reference Flores, Cicchetti and Rogosch2005; Warren, Reference Warren2005; Todd and Worell, Reference Todd and Worell2000) are addressed. Recognising this wide range of research topics within the study of resilience, it is useful to study how the term resilience can be defined to be used in all these research areas.

The concept of resilience – definitions

Resilience can be defined as the ‘dynamic process encompassing positive adaptation within the context of significant adversity’ (Luthar et al., Reference Luthar, Cicchetti and Becker2000), or, as Edwards puts it in the context of developmental psychology, ‘Resilience . . . is a capacity for adaptation along appropriate developmental pathways, despites disruptions such as family breakdowns’ (Edwards, Reference Edwards2007).

As exemplified above, most definitions emphasise two elements as crucial for defining resilience (Luthar and Zelazo, Reference Luthar, Zelazo and Luthar2003; Luthar et al., Reference Luthar, Cicchetti and Becker2000; Masten et al., Reference Masten, Best and Garmezy1990; Rutter, Reference Rutter1987; Werner and Smith, Reference Werner and Smith1988). First, an input perspective: the exposure to risk and adverse circumstances, which can vary from moderate to extreme risks environments. Risks can be conceptualised as follows: ‘it is used to predict vulnerability to a range of negative life outcomes including school failure and or dropping out of school, drug abuse, failed relationships, delinquency/criminal activities, unemployment, ill health and early death’ (Howard et al., Reference Howard, Dryden and Johnson1999). The second element of a resilience definition is in respect to an outcome perspective, studying whether coping mechanisms lead to outcomes within or above the expected range. This should be assessed by comparing the outcome to a context specific reference group (e.g. same age group, social and cultural context, etc.) (Masten and Powell, Reference Masten, Powell and Luthar2003). The definition further emphasises the aspect that resilience is a process rather than a static concept or an individual characteristic. There is an interaction and adaptation process that occurs after the individual is exposed to an adversity. This process perspective often requires the dynamic assessment over time which makes it necessary for researchers to follow the same individuals – for example by employing panel data. As Rutter Reference Rutter1987 emphasises, such a dynamic perspective helps to avoid misunderstandings of the concept of resilience as a fixed personal characteristic: ‘resilience cannot be seen as fixed attributes of the individual. If circumstances change, the risk alters’ (Rutter, Reference Rutter, Rolf, Masten, Cicchetti, Neuchterlein and Weintraub1990).

Other important conceptual issues raised in the literature are so-called protective factors, which help individuals to cope with adversity. There are three broad areas of protective factors that frequently appear in the literature (Howard et al., Reference Howard, Dryden and Johnson1999; Evans and Plumridge, Reference Evans and Plumridge2007). Here, the literature on childhood resilience is used to exemplify some protective factors. First, internal protective factors, which are the attributes of the child, are frequently mentioned. These range from problem-solving skills, aspirations, social competences, intellectual/cognitive skills, self-regard and self-esteem (Garmezy and Rutter, Reference Garmezy and Rutter1983; Werner and Smith, Reference Werner and Smith1988). Second, family characteristics as a factor external to the child can be a protective factor. These characteristics can include effective parenting style that employs adequate and persistent role models, harmony between parents, quality time spent with children and the socio-economic resources of the parents (Howard et al., Reference Howard, Dryden and Johnson1999). Third, external institutions such as the school or community of the child are important aspects. Here the strength of social support networks and the ability to extend these networks, having positive school experiences and opportunities to engage in social life are most often cited as protective factor (Barnes and Morris, Reference Barnes and Morris2007; Werner and Smith, Reference Werner and Smith1988; Manyena, Reference Manyena2006).

The concept of resilience – critique

One of the most prominent critiques in the resilience literature concerns the ambiguous definitions of resilience and central terminology, together with the limits to operationalise and measure key concepts (Boyden and Cooper, Reference Boyden and Cooper2007; Howard et al., Reference Howard, Dryden and Johnson1999; Luthar et al., Reference Luthar, Cicchetti and Becker2000). It is important to note here that resilience is never measured directly but inferred from its defining elements such as risks and adaptation (Luthar and Zelazo, Reference Luthar, Zelazo and Luthar2003). So, for example, a positive outcome of adaptation can be the reduced anxiety of a child under stress caused by war or natural disasters. However, what the resilience perspective makes necessary is for there to be a pre-existing understanding of risks, which already involve value judgements and assumptions. The same is true for the term of positive outcomes or positive adaptation (Boyden and Cooper, Reference Boyden and Cooper2007). While Luthar, Cicchetti and Becker do not deny this, they claim that this critique is only partly valid. They argue that for any scientific construct, a variability in research methods and terms is necessary. Indeed, they say such variability enables exploration of the research domain in the first place – especially if the research domain is interdisciplinary and tries to capture multi-dimensional phenomena. However, they do concede that the different disciplines should establish more precision in terminology and in spelling out the context and protective factors (Luthar et al., Reference Luthar, Cicchetti and Becker2000).

Additionally, there should be care with the unreflected use of values to define what positive outcomes in resilience are. Positive outcomes and success differ between groups (e.g. gender, ethnicity, socio-economic status). Definitions created by, for example white, middle-class, male academics in the Western World, have their limits in their applicability to other countries – especially developing countries (Howard et al., Reference Howard, Dryden and Johnson1999; Boyden and Cooper, Reference Boyden and Cooper2007). There is the question as to who is asked to define the concept. Further, there is a critique that resilience research lacks culturally sensitive definitions, as well as definitions that include what the target groups themselves understand as resilience. For example, Howard et al. argue for the inclusion of a child perspective on risk and resilience as an important component to understand and tackle the issue of child resilience (Howard et al., Reference Howard, Dryden and Johnson1999). This is where more participatory and culturally sensitive definitions of resilience could be explored. This could also reveal some of the behaviour that is usually not captured by a normative approach, as the latter tends to overlook the fact that successful outcomes may equally be due to maladaptive behaviour. Luthar et al. cite the example of inner-city youth who gain important protective factors such as self-esteem from peer acceptance through non-conforming behaviour (Luthar et al., Reference Luthar, Cicchetti and Becker2000).

Another critique of the resilience concept concerns the very shift from a deficit model to an asset model. The dangers inherit in a focus solely on preventive factors can misinform policies, and an over-reliance on prevention programmes could be the result. Further, it is important to make sure that the notion of resilience is not abused as a euphemistic description of otherwise too negative or depressing situations. Protective factors should add value to the literature of risk, but often they are just the opposite of risk variables (Howard et al., Reference Howard, Dryden and Johnson1999; Little et al., Reference Little, Ashford and Morpeth2004).

Another critique that reflects the established debate in sociology of actor versus structure is also valid for resilience research. Resilience research is accused of being too actor centred, ignoring any structural forces. This can de-politicise efforts such as poverty reduction and emphasise self-help in line with a neo-conservative agenda instead of stimulating state responsibility (Boyden and Cooper, Reference Boyden and Cooper2007). This individual−focused perspective can also lead to blaming the victim, especially when resilience is mistaken as a personal trait rather than a process (Luthar and Zelazo, Reference Luthar, Zelazo and Luthar2003). Related to this is an under-representation of research on community influences, social capital and networks (Barnes and Morris, Reference Barnes and Morris2007; Prior and Paris, Reference Prior and Paris2005; Beirens et al., Reference Beirens, Hughes, Hek and Spicer2007; Howard et al., Reference Howard, Dryden and Johnson1999; Edwards, Reference Edwards2007).

Gaps in the literature – Further Research

There are several gaps that can be identified when reviewing the literature on resilience. First, while there is extensive coverage of children and adolescents in the resilience literature, there are relatively few studies investigating resilience among adults and old people. Second, complex adversities – multiple events occurring simultaneously (e.g. unemployment, low income, poor housing, bad health and family breakdown) − are rarely analysed together. One of the potential reasons is the limits of quantitative models. The complexity of research, including a wide range of outcomes, potential risk factors, protective factors and processes, places limits on quantitative models (Barton, Reference Barton and Ungar2005). Third, regarding the timeframe of research studies, while there are a few pioneering exceptions, such as the studies by Werner and Smith (Reference Werner and Smith1988, Reference Werner and Smith1990), most use short-term timeframes. Fourth, as already outlined in the critique of the resilience concept, the majority of resilience research has its focus on the individual level. Therefore, more research with different levels of analysis such as networks, organisations and policies is important. And, finally, there is a lack of comparative studies – e.g. different welfare state regimes (such as UK and Sweden) − with one of the few exceptions being Jones et al. (Reference Jones, Burström, Marttila, Canvin and Whitehead2006). All these perspectives present important and exciting areas for further research to better understand how people cope with adversities.