The philosophy of critical realism and childhood studies

Abstract

Critical realism is a philosophy of social science that analyses and aims to remedy current problems and gaps. Basic tenets of positivist and quantitative research tend to contradict those of qualitative and interpretive research, and critical realism proposes ways to resolve the contradictions. Vital themes in childhood research that are reviewed in this article include a comparison with feminist research, critical realism, being and thought, transitive and intransitive, theory/practice consistency, agency and structure, closed and open systems, micro and macro in the global/local nexus, four planar social being, facts and values, and transformative change through the four-stage MELD dialectic. Critical realism aims to understand the world in order to be able move from coercion towards creative liberating power.

Introduction: Philosophy and practical research methodology

This article considers how philosophy is integral to practical social research methodology. During 30 years of researching with children and young people, I became increasingly puzzled by philosophical splits and contradictions in social research and gradually learned how critical realism (CR) helps to resolve them. CR is a philosophy of all the social and natural sciences, not an alternative sociology (Bhaskar, 1998, 2010). It challenges traditions in 2500 years of Western philosophy and science and proposes alternatives. CR helps researchers to reduce and avoid confusions, to clarify their chosen theories and methods and to justify, validate and strengthen their research findings in order to inform policymakers, practitioners and the general public.

A few critical realist concepts will be reviewed later in this article: being and thought, transitive and intransitive, theory/practice consistency, agency and structure, closed and open systems, micro and macro in the global/local nexus, four planar social being, facts and values, and transformative change through the four-stage MELD dialectic. First, connections between philosophy and social research will be considered.

Philosophy is often taken to be a separate discipline from social science, one that social researchers can safely ignore and which only raises unnecessary complications in social research. However, philosophical questions pervade all research, whether researchers address them directly or not. To avoid philosophy can leave unresolved confusions and hidden unquestioned assumptions that distort research. Efficient research methodology therefore involves examining basic philosophical questions. These include the following: What is the nature and purpose of the research? How can we be sure that the things we observe (people, objects, events, relationships, outcomes) really exist? How can we check the meaning, accuracy and validity of our observations? These questions are key philosophical ones about meaning (epistemology) and existence (ontology) and later sections will review how CR helps to answer them.

Philosophy assists clear, logical, critical analysis, which appreciates and organises different perspectives, and works to justify conclusions. The philosopher Mary Midgley (1996) compared philosophy to plumbing, saying that often both are ignored until things go wrong, when there are leaking pipes, blocked drains and flooded floors – or incoherent, confused and blocked flow of ideas. Then, the central importance of plumbing or philosophy is recognised. Philosophy helps to connect the pipes (the coherent, organised flow of ideas) and clear the drains (blockages, irrelevances and contradictions).

Philosophy at work in feminist research and childhood studies

The need for clearer philosophy about the meaning and purpose of much childhood research will briefly be shown in a comparison with feminist research. Feminism has definite meanings and purposes. These include to promote equality between men and women; to show how, far from realising its claims to be unbiased and inclusive, sexist sociology has excluded, ignored and denigrated women; and to demonstrate how women are central not marginal members of society and therefore that feminist research can provide more comprehensive and reliable social accounts and analysis of women and men than one-sided sexist research can do.

Some researchers hoped that childhood studies would similarly emancipate children, as feminist research has helped to promote women’s status and inclusion. However, among many other childhood researchers, Horton and Kraftl (2006: 139) are concerned that childhood research is not ‘more useful, more engaged in policy and weighty contemporary issues’; more relevant, purposeful and applied; and less ‘self-referential, endlessly re-citing an all too familiar body of work … a rationale for itself … a comfort zone … cosy and unchallenging’.

These problems connect, I suggest, to unresolved philosophical questions. Unlike feminism, childhood research has no decisive purpose of promoting equality between children and adults because of uncertainties about children’s and young people’s competence and maturity (although adults’ competence also widely varies). By respecting children, childhood research may be more unbiased and inclusive than adult-centric research. Yet, whereas feminism addresses all adults, childhood research is less inclusive when it tends to ‘focus on children’, on their specific, or different, or unique concerns, largely separate from adults.

Undue ‘focus on children’ has serious disadvantages. It risks sidelining many of the most vital concerns that affect children as broadly and deeply as they affect adults, such as politics and economics. Children’s citizenship, for example, is still mainly researched as the education and preparation of children, seldom as active political agency (Hayward, 2012). Much research gives partial thin accounts of children in institutions and in their personal and semi-private lives. Effects on children of economic policies may be measured, but the actual policies are seldom critically analysed from a childhood perspective in the way they constantly are from adult-centric perspectives (Harvey, 2012; Wacquant, 2009, among countless examples). And the main positioning of children into an age group displaces the much more important grouping of children alongside adults into their socio-economic classes.

Children are doubly excluded from expert research analysis. First, childhood researchers seldom have the weighty expertise that informs complex ‘adult’ social research, such as in political or economic theory or jurisprudence. Instead, there is useful but less theoretical concern with welfare, protection and education. Second, mainstream research ignores children, much as it used to ignore women, and academic texts seldom mention children. Women even claim that they are ‘half the world’, deleting the one-third of people aged under 18 years.

Do these exclusions matter? They raise questions about the nature and purpose of childhood research. Since the 1970s, feminist research and activism have brought great advances in women’s rights and social inclusion. Yet from around 1990 while childhood research has expanded, high and rising rates are reported globally of many children’s and young people’s poverty, debt, stress, mental illness, social exclusion, unpaid or very low paid work and of ‘zero-tolerance’ schools, with rigorous testing that countless children fail. Children’s school test scores are among the massive databanks that so much research time and funding are devoted to analysing, as part of the growing childhood surveillance industry. Meanwhile, armed conflict and economic and ecological crises are multiplying the problems younger generations endure and are inheriting, with lack of the critical forward planning based on intergenerational justice that childhood research could do far more to analyse and promote.

Although research cannot directly transform lives, philosophical questions related to the indirect effects of research include the following: Cannot the inclusive aims, methods and impact of feminist research be applied more fully in childhood research? Does a ‘focus on children’, partly isolating them from the complex society they share with adults, provide misleading findings? Does it possibly inadvertently reinforce the social exclusion of children? What deeper beliefs underlie our research about the nature and purpose of being a person, child or adult, individually and in larger society? To address these questions, this article considers both children and adults, and if an age is not specified, that means children are extensively involved too.

Being and thought

Western philosophy and science tend to reduce existing things into thoughts, ontology into epistemology (Bhaskar, 1998, 2008, 2010), such as real children into abstract concepts or statistics of childhood. Phenomena and their existence and function can then seem abstract, tenuous, distant and unreal. Researchers then face the problem of how to demonstrate that the subjects in their reports (actual people, events, relationships) really exist. A young girl’s view when she was interviewed in an English secure training centre (youth prison) will be used to illustrate this:

I think we should be able to hug our families when they come and visit us because it is bad enough being away from them for a year and a half. It is awful. (Willow, 2015: 170)

Quantitative researchers tend to take such responses at face value and reduce them into preset formats, such as measures that grade young prisoners’ views from ‘very satisfied’ through to ‘very dissatisfied’ with family visits. Formats might measure young people’s reported needs or behaviours, or compare their views during detention with later outcomes, including through multivariate analysis.

Useful as all this research can be, there is a hyper-realism (over-belief in reality), when complex personal views are treated as if they are objective, stand-alone, yes/no facts that fit binary analysis. There is a sense that whoever observes, records, reports or reads about the response sees the same facts, which have an essential, lasting reality, unchanged across time and space. Positivist social researchers partly share natural scientists’ aims: to discover general laws and replicable findings and to predict likely outcomes based on seemingly powerful correlations and statistical probabilities. Quantitative researchers might concede that the girl could be exaggerating or lying, but be confident that involving hundreds or thousands of participants can balance out such problems, and give power and validity to the total data. Faith in numbers guides this reducing of phenomena such as diverse living experiences (ontology) into thinking and measuring (epistemology).

Qualitative researchers tend to be more reserved about facts and reality. Many prefer a cautious hypo-realism (under-belief in reality). They may be wary about the accuracy of interviewees’ accounts and perceptions and memories and of researchers’ own abilities to record and verify these. Younger interviewees and those with a criminal record may be treated with extra scepticism. Researchers may see accounts as contingent and socially constructed, varying according to the nature and purpose of the interview and the relationship between the speaker and listener, hardly expressing independent, real, transferable facts. Ethnomethodologists may suspend or bracket off interest in the possibly misleading, consciously presented content of accounts (the girl’s experiences and reported perceptions). Instead, they search for more reliable data in the format of how interviewees, often subconsciously, present themselves (Silverman, 2010). The girl may be claiming a moral account of herself as a loving daughter who wants ‘normal’ relationships, and her actual relationships may be seen as irrelevant.

Researchers often triangulate young people’s accounts with observations and with other reports. Young people in detention centres reported feeling very hungry (Willow, 2015), views that were confirmed by the minimal catering budgets, poor quality of the meals and service and critical official Inquiries. Pre-packed breakfasts delivered to cells in the evenings might be eaten 18 hours before the next meal at midday. However, to collate these different forms of supporting evidence depends on some belief in the reality of real food and real hungry bodies, just as ‘hugging’ relatives during visits spoke of real embodied relationships and longings.

Yet, interpretive research methods tend to treat phenomena as if they are constructed by our subjective perceptions and negotiated interactions within specific social contexts and cultures. Existing things then seem to have little or no essential, inherent ontology and independent, lasting truth or reality of their own that could transfer intact across time and space. Being and doing are reduced into thinking; phenomena are not seen to exist independently ‘out there’ in the world, but only through the social institutions and cultures that give them meaning, for example, in the social constructionism promoted by James and Prout (1997).

This can be liberating, such as when forms of childhood and childcare are recognised to be highly varied and local, not inexorable. And it can encourage children and adults to have more agency and choice in deconstructing seemingly fixed realities, and potentially in changing oppressive traditions into liberating ones.

Yet much social research, including anthropology, aims to observe and report neutrally by respecting each tradition in its own context, describing and theorising but avoiding critical questions or comparisons or plans for policy changes to alter the observed world. Such neutrality suspends moral judgements of customs and how these may harm or benefit children and adults. This unreal value-freedom (Sayer, 2011) can merge into relativism and irrealism, when values become abstract notions separated from real moral embodied emotions of joy, suffering, fear and hope, and from mental-physical hunger, constraint, pain or pleasure.

When meaning is detached from direct living (epistemology from ontology), connections between research data and conclusions and between recommendations and later policymaking may look like tenuous constructions. This reduces the potential of childhood studies to promote social justice.

A further great problem is that unresolved contradictions between social researchers’ reliance on hyper-realism or on hypo-realism can undermine the reputation, validity and practical influence of social science. If social scientists cannot convince one another, how can they convince the public or policymakers?

CR analyses the underlying theories and seeks to resolve the contradictions. Bhaskar (2008, 2010) traced how Plato identified true reality in unseen abstract enduring Forms (such as the idea of a universal child), which transcend the countless everyday transient varieties of children’s experiences. Psychological research interest in statistical norms of child development is one legacy of Forms, as if the abstractions are more scientific and significant than all the diversity of existing children. Descartes took the abstraction further by doubting the existence of everything except his own thinking mind, which he took as proof of any existence, cogito ergo sum. Similarly, social constructionism reduces reality into thought when it takes reality to be constructed or invented by our contingent perceptions, without an independent existence to be discovered.

This brief article can only allude to lengthy CR analyses (such as Bhaskar, 2010; Hartwig, 2007; Norrie, 2010) of thousands of years of the epistemic fallacy (collapsing being into thinking) and of realist alternatives. The next sections review two versions of the epistemic fallacy, transitive/intransitive and theory/practice interactions.

Transitive and intransitive

CR helps to resolve hyper-reality and hypo-reality contradictions. First, CR recognises our transitive subjective thoughts and perceptions as separate from intransitive reality, which exists independently before and after we are aware of it. Neither can be wholly reduced into the other. Although our thoughts can affect real people and events, they do not create those realities. And real children and their practical daily lives cannot wholly be reduced into concepts of childhood, although they are affected by the concepts. When researchers skip the necessary philosophical task of distinguishing ontology from epistemology, they confuse and collapse real existing children into theories of the developing, or resilient, or helpless child. The theories may partly illuminate, but they also limit and fragment, research about children’s complex lives.

Second, CR turns dualisms into interactive dialectics. Children are perceived in varying ways by the adults and peers who know them (transitive epistemology), and they also exist partly independently of these varying views (intransitive ontology), though influenced, and ‘reconstructed’ by, and interacting with, the perceptions on many levels.

Theory/practice consistency

Another way to avoid the epistemic fallacy and to validate reality is through CR’s concept of theory/practice consistency. Researchers show an irrational inconsistency if, on one hand, in research theory they assume a constructed contingent hypo-realism and deny any independent enduring transferable reality, but, on the other hand, in everyday practice they rely on the printed words of their reports being replicated and understood anywhere. Everyday practice relies on an independent enduring transferable reality which the theory denies. CR sets the same consistent standards of realism in theory and practice, in research as in life.

Levels of reality

CR understands reality at three levels. The empirical level recognises our transitive subjective and greatly varying perceptions and experiences. The detained girl’s feeling about family visits, described earlier, is an empirical example. Level 2 involves actual objects, people, interactions and events, such as the girl and her family sitting in the detention centre. Level 3, the real level of causal reality, may be invisible, not provable and only known in its effects. Deep powerful causal social realities include inequalities of class, family income, gender, ethnicity and generation besides global politics such as deregulated capitalism.

As these real social powers are largely invisible and often linked to politics and social justice, they tend to be doubly dismissed by ‘objective’, empirical scientists who concentrate on the empirical and actual levels. Yet, this can be like observing many falling stones, looking for obvious causes of falling within the stones or in patterns between them, perhaps using multivariate analysis, but distracted away examining deeper explanatory causal and structural power. With the stones this power is gravity. With childhood research, the describing, narrating, measuring or correlating methods tend to stay at the empirical and actual levels and stop short of CR analysis of real underlying social causes. Causal problems may then overtly or covertly be identified within the children and adults observed and their behaviours, but not also within underlying structures, such as the economic and penal contexts of imprisoned young people.

Currently, research funders tend to favour evidence and proof, diverting researchers’ attention away from unseen social causes and towards empirical effects. Yet, the levels of causes and effects are vital in both the natural and social sciences. CR shows how the social and natural sciences are much closer than is often supposed. Far from one science dealing with physical things, and the other with social ideas, they both deal with solid things and abstract ideas, with realities and theories, sharing a unity, though not uniformity, of methods (Bhaskar, 1998). In the natural sciences, causal realities such as gravity and evolution can never directly be known or proved and might one day be disproved, but they have infinite, enduring effects that appear to prove them consistently. The next section looks at complications in proving causes.

Open and closed systems

CR contrasts open systems when different forces compete with the very rare closed systems where only one force exists. Causes can be proved in closed systems, but the relative influence of multiple causes is hard to discern in open systems. Scientists therefore aim to obtain clear proof by constructing closed systems such as through randomised controlled trials (RCTs). Randomisation aims to ensure that the only difference between, say, all the groups of patients is the type of medications they are given so that outcomes can be compared.

However, natural scientists do accept proof from competing causes in open systems; the different forces of wind currents, bird flight and jet engines enable objects to defy gravity. But they are not thereby assumed to disprove the power of gravity.

Like gravity, social structures such as class and generation may be active, or counteracted, or unseen latent and potential forces. Because social causes work in multiple, complex, interacting and emergent open systems, social effects cannot be directly attributed to a single, determining, structural cause. Yet, like the mixed (not 100%) results of gravity or of medication in an RCT, socio-economic forces such as poverty may have such highly consistent effects that they too can be accepted as causes. This CR analysis helps to validate complex analysis of social causes and predictions. They are further complicated by human agency as considered next.

The nature of agency and structure

Social research tends to ignore or downplay children’s agency in two main ways. Quantitative surveys emphasise social structures and variables in thousands of anonymous lives (for example Hansen et al., 2010). And qualitative, interpretive methods may question and ‘decentre’ individuals’ agency, seeing agency as ‘an effect brought about by the assembly of heterogeneous materials’ and diverse resources ‘through which children’s agency is (or is not) produced’ (Prout, 2000: 17).

There is ambiguity here about who produces children’s agency, as in Prout’s use of actor network theory. This sets people and objects on a similar level of agency and overlooks the conscious agency unique to human beings. Oswell’s erudite analysis of agency concludes that ‘the individual child’ myth, which ‘few believe’, falsely sees children as possessing capacities individually. Instead, Oswell contends, ‘agency circulates around children’ impersonally, and exists ‘only by virtue of their relationship with others’ as facets of each local situation. ‘Agency neither starts nor finishes with any individual agent’. It is orchestrated within narrative structures, character is not ‘real’ and there is no single author but always multiple authors of agency. Agency is never a property and is always relational, in-between, dispersed (Oswell, 2013: 264–270). His view is supported in many chapters in Esser et al. (2016). While partly acknowledging these complexities, CR takes each child and adult also to be a distinct, conscious, embodied, unique individual, possessing real, though limited, agency.

The possibility of individual agency may be denied if agency seems to merge too much into the social context. Yet, this denial could mistakenly assume all or nothing: agency is either pure and separate from all context or is dispersed in social contingencies. The reality lies between the extremes. Agency and structure can only exist in dialectical relation to each other, separate but interacting and partly overlapping entities. Without the values and resources from structures that enable us to make sense of experience, agency would be either empty or else chaotically, meaninglessly overwhelmed.

The critical realist Douglas Porpora (1998, 2007) identified four theories of social structure. First are patterns stable over time of aggregate behaviour by atomised individuals, the model used in economics. Second, law-like regularities govern, determine and explain the somewhat mechanistic behaviour of social facts such as in birth cohort studies (Hansen et al. 2010), structural functionalism and child development theory. Porpora believes this model is too virtual, static and random and takes too little account of agency and diversity. CR accepts neither the first over-free voluntarism nor the second over-determining structures.

Third, structures are seen as rules and resources, such as in Giddens’ (1979) structuration theory. Archer et al. (1998) criticise this as over-emphasising voluntary agency. While they accept that social structures, unlike natural ones, can only operate through human agency, they believe that structures cannot be reduced into agents’ conscious conceptions of them or into agents’ activities and use of structures, rules and resources.

Fourth, Porpora’s and Archer’s preferred model sees social structures as tensed systems of human relations among social positions. These constantly change through interactions with agents and between structures in the transformational model of social activity (Bhaskar, 2008: 154–160) and the morphogenetic social process (‘morph’ meaning shape and ‘genetic’ meaning origins) (Archer, 1995, 2003). Rather than simply being underlying or external rules to be invoked, structures ranging from childhood and the family to financial systems are powerful and enduring systems and positions in and through emergent material and social relations. Over time, structures far precede and outlast individual agents. Although individual agents do not construct social structures, they constantly reproduce and resist, modify and interact with them, and are shaped and reshaped by them through social processes in time and space. Structures may work to exploit or oppress agents, and they include modes of production, domination, competition and inequality. The broad concept of powerful structures allows for agents’ widely varying individual reactions, interactions and resistance within open systems.

Interacting structure and agency

Social structure is a necessary condition and medium for reproducing and transforming social forms. CR sees individual and collective agency as neither over free and voluntary nor over determined and mechanical. Instead, structure and agency are seen to interact dialectically over time: agents draw on and are driven by a range of pre-existing structures in open systems; structures may cause but do not determine agent’s activities; and agents shape and reshape structures in tensed sequences.

Loic Wacquant (2009) asserted that vast, recent economic and urban changes, with the rise of the penal workfare and prisonfare state, were caused neither by individual agents nor by mysterious, forceful social structures. Instead, they emerged from numerous interacting influences and struggles. Many combined forces work as if to open certain doors and close others. The forces expand the power, momentum and hegemony of some policies, while they restrict and deter others. Alternatives do not become impossible, but they become much harder to imagine or achieve, changing the balance of powers in open systems.

Very many young people in prison have mental illness and learning difficulties and have suffered severe abuse and neglect. But if these influences are mentioned to partially explain youth crime, a common response is that not all young prisoners have these disadvantages, and many young people who do have the disadvantages do not commit crimes. It is implied that therefore everyone has the free will, choice and agency to rise above circumstances. This sets the power of agency above the power of structures, and implies there is a single overriding power in a closed system (such as human free will) instead of complex multiple pressures in open systems. Social structures of childhood abuse, neglect and poverty have immense, though not absolute, power, just as natural structures do. Some young people overcome severe disadvantage, but they do not thereby disprove its great power to shape many lives.

Margaret Archer (2003) analysed in great detail the time sequences when (adult) agents draw on and then reflect and relate to structures through their internal conversations and later act on them. Children’s knowledge, judgement, foresight, freedom of choice, control and agency are all very limited, but so too are adults’ capacities. At all ages, human agency is constantly constrained by structures and by other agents, by resources and chance, as shown in the limited agency of supposedly powerful politicians.

The critical realist Christian Smith (2011: 317) contended that ‘a thick [complex] notion of persons is essential for rightly understanding what social structures are and why and how they come to exist and change’. The characteristics he gives for social structures fit childhood: actively sustained by human agency and bodily practices, having specific dynamic historical existences, always implicating inanimate material objects (from toys to schools), constituted in part by cultural mental categories (such as the ignorant, dependent child) and always involving normative and moral dimensions and normalising sanctions (especially relevant to children’s education and socialisation) (Smith, 2011: 322–329). Smith believed that social structures usually encourage passive acquiescence if not agreeable adherence. Persons and structures exist and interact in stratified reality (at the empirical, actual and real levels), and they are emergent. They develop and emerge from prior entities, but cannot be reduced back into them anymore than water can separate back into oxygen and hydrogen molecules.

Micro and macro: Troubling the global/local nexus

CR assists global childhood studies to examine micro-local studies and the broad sweep of international concerns and to organise and inter-relate all the levels in between, even in small, short projects. Small studies can be nested into background literature reviews that show how they affect and are affected by much larger arenas. CR’s four planar social being analyses all aspects of human life (Bhaskar, 2008):

1. Material relations with nature in the physical reality of bodies and of the natural world;

2. Interpersonal subjective relationships between individuals and groups;

3. Broader social structures;

4. Inner being, personal subjective agency and ideas about the good life and the good society.

The four interacting planes are understood in relationship, reshaping one another in time and space, history and nature. In the example of the girl in the detention centre, they inhere, first, in her physical presence and her embodied needs and activities and, second, in her relationships with her family and friends, the staff and the other young prisoners. Third, broader structures include penal systems and social inequalities, which result in prisoners disproportionately being from deprived and ethnic minority backgrounds, and having learning difficulties, mental illness and interrupted schooling. Local, national and international economic policies influence whether societies concentrate on education and crime prevention or on intensive policing and punishment. Plane 4 attends to personal processes of change within young people as they become reformed, or alternatively become more involved in crime. There are also political and social change when penal systems become more punitive and concerned with retribution or else more concerned with reform, re-education and helping young people to improve their life chances. Childhood exists on all four interacting planes of social being in diverse ways, best revealed through interdisciplinary research that challenges and troubles the global–local nexus (Alderson, 2013, 2016).

Facts and values

A dominant philosophy following the 18th-century philosopher David Hume separates facts from values, and therefore ‘ought’ from ‘is’. For instance, although the ethnographer Martyn Hammersley (1995) acknowledged that ‘value argument … is essential to governance and to everyday life’ (p. 14), he contended,

what should or should not be done [cannot] be logically derived solely from social science evidence about the matter … Neither the social scientist nor the philosopher has any superior expertise or authority in judging appropriate action and disagreements … [They] should strive to be value neutral or objective [in their own work]. (Hammersley, 2009: 5–6)

However, although researchers may not have more moral expertise than other people, they may not have less, while their well-collected data and careful analysis give them some authority to engage in moral analysis. Social facts, such as the girl’s wish to hug her family, do not necessarily dictate which values to follow, but they raise unavoidable questions and choices about the values which are integral to social relations (Archer, 2003; Bhaskar, 2008; Midgley, 2014; Sayer, 2011). Attempts to split values, such as respect, kindness, justice and relief of suffering, away from children’s everyday lives distort any understanding of childhood. Researchers need not decide which ‘ought’ to favour, but they can analyse a range of options starting from the childhood perspective or standpoint, which is so often missing from mainstream work.

Ironically, Hammersley invokes the morality he wishes to avoid when he advises that researchers ‘should’ be ‘value neutral’ or amoral. Even the effort towards ‘value-free’ objectivity springs from values about truth and excellence in research. Theories and values that are assumed or covert (such as illusions that research can be value-free, adults are constantly benign and rational, children are inevitably volatile and irrational) can misdirect and distort research more powerfully than explicit and openly questioned theories and values do. CR works to reveal and unravel such social and moral fallacies, seeing social research as a moral endeavour to benefit societies.

MELD

CR examines the dynamic moral process through dialectic. The traditional dialectical search for truth is the three-part thesis, antithesis and final closure in synthesis. CR expands this into more practical and open-ended four stages, denoted by the acronym MELD (1M, 2E, 3L and 4D) (Bhaskar, 2008).

1M first moment begins with absence, ‘the simplest and most elemental concept of all’ (Bhaskar, 2008: 239). It involves non-identity, alterity, sheer other-being and irreducible differences between beings, and between our transitive perceptions and the intransitive reality observed but never wholly understood. Like anthropologists, critical realists stand back to try to understand the meanings held within the intransitive subjects of research, instead of trying to identify and collapse these meanings into the researchers’ own prior transitive understandings. 1M differentiates between the empirical, the actual and the real, knowing that we can understand very little of the vast unseen real. 1M aims to move beyond anthropism (human-centeredness and the Cartesian epistemic fallacy), which so seriously distorts our understanding of the cosmos and our place in it among the elements and other species. CR Childhood research also aims to move beyond the adult-centrism that misunderstands, or ignores, or excludes children from mainstream political societies.

1M aims to overcome dualisms such as individual-general, agency-structure, mind-body, which underlie coercive power2 relations in the epistemic fallacy. 1M studies what is absent, and the awareness of omissions moves forward to 2E, second edge. Here, the concern is with active intervention to negate or absent the ills and needs noted at 1M. Powerful groups tend to start at 2E, such as when lawmakers assume that young criminals must be punished, to negate their past failings. They omit the vital 1M stage of searching for the original problems and causes of crime that need to be redressed at 2E.

Practical 2E is followed by another reflective stage, 3L third level, which considers larger structures and totalities. Penal systems and crime prevention are connected to many national and international socio-economic contexts, for example, poverty and inequality, migration into cities of strangers, which lose the former small community relationships that deter crime within the larger global–local nexus of economic and political systems that increase violence and unrest (Alderson, 2016; Harvey, 2012; Wacquant, 2009). CR’s aim is less to excuse crime than to understand its deeper and larger causes in order to plan really effective reforms at political and not only personal levels.

Whereas traditional dialectics concentrated on thought and logic (epistemology), CR is also concerned with practical social change (ontology), and therefore 3L involves partial but open synthesis which leads to 4D fourth dimension. In the dialectic of movement and change, the MELD impetus seeks to move from oppressive coercive power (the CR concept of power2) towards creative personal and political freedom, the CR concept of power1, and the flourishing of each person that depends on the flourishing of all in freedom and justice. Alderson (2016: Chapter 12) discusses how this involves children.

Conclusion

This article has briefly outlined a few ways in which CR can help childhood researchers to conduct more philosophically informed, critical and realistic research, with greater theory–practice consistency, which is practical and relevant to policymakers and professionals and to all generations of the general public. More detailed application of CR to interdisciplinary childhood research is considered in my two recent books (Alderson, 2013, 2016).

I have suggested that childhood research could adopt more of the inclusive aims and methods of feminist research. Instead of a ‘focus on children’ partly isolated from complex mainstream society, we could do more to promote children’s social inclusion, first, by expanding socio-economic-political themes in our own work and, second, by prompting our mainstream colleagues to pay far more attention to children and young people. These aims involve analysing philosophical beliefs that underlie our research about the nature and purpose of being a person, adult or child, individually and in larger society.