Social work legitimacy: democratising research, policy and practice in child protection

This article analyses the concept of legitimacy as applied to the use of power in statutory social work with children and families in the UK. It draws on literature from police studies and criminology, in which the concept is a stable one that continues to be heavily researched and analysed. Police and social workers bear comparison in respect of legitimacy because of the significant powers they use on behalf of the state with direct implications for the civil and human rights of their fellow citizens. The article defines legitimacy in theoretical terms before applying the concept to social work. Here, perceptions of fairness in the distribution of resources, the quality of treatment people receive, and the quality of decision-making are critically examined. The article then proposes a democratising agenda across the three domains of social work research, policy, and practice. Through challenging social work’s legitimacy and analy-sing its relationship to social democracy, it is argued that new ways may be found to realign practice with the values of human rights and social justice that are said to underpin the profession. Given the severe socioeconomic impact of the Covid-19 pandemic on many families, these questions acquire a particular urgency.


Introduction
This article asks if the use of power in social work can be regarded as legitimate and how concerns about its legitimacy may be addressed.

www.basw.co.uk
The article draws on literature from police studies and criminology, in which there has been an expansion in research and vigorous debate on the concept for some twenty-five years, constituting a 'legitimacy turn' (Tankebe, 2013, p.104, emphasis in original). The expansion of interest was spurred by the recognition that public perceptions of police legitimacy are key to understanding why people obey the law (Tyler, 1990). In this context, power is generally considered legitimate if it meets the three conditions of legality, shared values, and consent (Beetham, 1991). Whereas legitimacy in police studies has become a stable topic, there is, as yet, barely any sign of explicit attention to it in social work. The aim of this article is to signal the start of a legitimacy turn for social work.
Police and statutory social workers bear comparison because they each have duties and use significant powers on behalf of the state that have direct implications for the civil and human rights of their fellow citizens. Both police officers and social workers can also be understood as 'street level bureaucrats' in that they have certain degrees of discretion in their interpretation and enforcement of policy and the law (Lipsky, 2010). With parallels in social work, police work is described as being a 'tragic necessity' involving great power, practised in low visibility environments, and (mainly) encounters with groups who have relatively low social status and little power (Reiner, 2016, p.133). The significance of their power reaches far beyond the impact of decisions made at street level. Social work in child protection is an 'acute representation of the underpinning settlement between the family and the state' (Morris et al., 2018a, p.364). Similarly, questions of policing are political questions about the nature of society and the limits of state power (Jones et al., 2012). Both police and social workers are seen by policymakers as solving, on the one hand, the problem of crime and, on the other, the problem of child abuse and neglect. The search for the 'magic copper bullet' of fighting crime (ibid. p.221) may be compared with the equivalent expectation of statutory social work's capacity for 'child rescue'. In each case, however, the causes of the problems are far wider. Just as policing is 'clogged by the fallout of socio-economic and power inequalities' (Reiner, 2016, p.134), so too is social work with children and families. It is government policies outside both policing and social work that have the biggest impact on both crime rates and child welfare.
By offering this analysis, there is no intention to idealise developments that have taken place in policing, where questions of legitimacy and democratic egalitarianism are clearly far from being resolved. In July 2020, the Independent Office for Police Conduct launched its thematic focus on race discrimination, stating: 'Evidence of disproportionality in the use of police powers has long been a concern which impacts on confidence in policing, particularly in Black, Asian and minority ethnic communities'. The present article simply argues that the concept of legitimacy is potentially helpful in reconceptualising the human rights and social justice issues that the social work profession faces in the UK context. The arguments complement other appeals for change (for example, by Featherstone et al., 2018), and add a potentially powerful dimension to the long history of service user involvement in social work (Beresford, 2018).
To make the case for a legitimacy turn for social work, the focus of the article is on the evidence for disproportionality and the experience of people who are subject to the use of social work power in child protection interventions, particularly families living in poverty. Questions of legitimacy raise, in turn, questions about democratic accountability, and the article therefore also proposes an agenda for change in social work which is democratising. Democratising social work means establishing dialogue with local communities about values and being open to scrutiny in the exercise of statutory powers.
It is important to stress that proposals for change are presented with a realistic appraisal of the crisis that communities and services face, not only immediately, in the continuing aftermath of austerity policies, but also in coming years, given the severe socioeconomic impact of the Covid-19 pandemic. There are already, at the time of writing, deep causes for concern (Featherstone and Gupta, 2020). As the Local Government Association has observed, 'The challenges and opportunities ahead demand a re-thinking of public services at pace ' (2020, p.4). It is argued that it is precisely under such extreme conditions that the use of state power by statutory social workers should be subject to questions of legitimacy and it is also when a democratising agenda is most urgently required.
Three important points about the article should be noted. Firstly, the concept of legitimacy and the democratising agenda that is proposed raise many complex issues that cannot all be addressed in detail by a work of this length. The aim of the article is to establish that further indepth work in this area would be worthwhile by providing a persuasive argument that social work legitimacy is a valid focus for attention. Further work that extends these ideas will follow. Secondly, the article necessarily engages with a more negative narrative about child protection social work than might seem reasonable. This is not because the evidence necessarily points to an overwhelmingly negative picture -in fact, there is considerable room for debate about the 'true' picture in this regard. It is, rather, because questions of legitimacy inevitably must address the more critical problems and dilemmas that social work in this arena faces. Thirdly, while the article is written from a UK child protection perspective, the criteria for legitimacy in the use of power in social work may usefully be applied in other contexts and to other domains of practice where a statutory role is required.
In the section that follows, the article looks in more detail at the relationship between social work and policing, and highlights, briefly, the contrast between these two state actors relating to the legitimacy of their power and democratic accountability. The article then moves on to define legitimacy in more detail, particularly as it has been applied to policing, before focusing on how the concept of legitimacy can be applied to social work in child protection. The article then discusses what a democratising agenda for social work research, policy and practice might comprise.
Social work, the police and state power: between soft cops and the secret social service Social work and policing share tensions between what might loosely be termed their 'care and control' functions. It has long been recognised that social work has a paradoxical role as a humanising project in which social workers simultaneously act as 'soft cops' in maintaining social order (Goroff, 1981, p.1). Whilst policing has historically been overtly identified with the function of political and social control, the controlling functions of social work have manifested 'under the cover of kindness' (Margolin, 1997). This is not to suggest that social work involvement is necessarily experienced as kindness by those subject to it (as explored later). It is rather that social work's controlling functions operate alongside the imperative of 'doing good' rather than explicitly 'keeping order'. This contrast is further illuminated by understanding how police and social work powers are symbolically located differently in relation to public and private domains.
Parenting, or more specifically mothering, is designated a private activity that takes place in the home. The state's intervention in family life is thereby subject to a complex array of tensions and contradictions (Parton, 2014). This reflects the roots of social work, which lie in home visiting and casework (Parton, 1985), and more specifically the gendered nature of ordering society. The importance of gender relations can be observed in the significant aspects of policing that intersect with social work. The welfare domains of policing such as domestic abuse, mental health, child abuse and youth offending have long been compared with social work, so that 'the police remain, to this day, the secret social service' (Stanko et al., 2012, p.328, after Punch 1979. 'Mission creep' has extended police involvement in this range of activities over recent years, particularly under austerity (Independent Police Commission, 2013, p.39). However, research suggests that the emotional labour involved in these activities means they are often regarded with disdain by officers, consistent with the traditional masculine culture in the police service at large (Lennie et al., 2020). Similarly, Garrett (2004) found a feminisation of child protection work in the police, and a blurring of the social work role.
Since its earliest incarnation, policing in the UK has been explicitly associated with the public domain and thereby with specific forms of public accountability. Policing was envisaged as being by public consent, where 'the police are the public and that the public are the police'these principles from 1829 being cited by the British Home Secretary as recently as 2012 (Home Office, 2012). By implication, any departure from public conceptions of fairness in the operation of police powers is perceived as inherently suspect. The Macpherson Inquiry into the racist murder of Stephen Lawrence in 1993 sought to establish greater community involvement in policing through Independent Advisory Groups, which were to act as 'critical friends' (Dixon, 2020). Theresa May, as Home Secretary to the 2010 Coalition Government, set in train the most significant change to police accountability in decades through the direct election of Police and Crime Commissioners. In her introduction to Policing in the 21 st Century: Re-connecting the police and the people, May announced, 'a radical shift in power and control away from government back to people and communities' (Home Office, 2010, p.2). However, it should be noted that, while elections are necessary, they are not sufficient to achieve democratic policing, particularly in an era of widening inequality and in the absence of full civil and socioeconomic rights (Reiner, 2016). Writing at the time of the Black Lives Matter movements across the globe, many police forces including in the UK face a renewed legitimacy crisis.
What this brief account illustrates is the degree to which the use of state power in policing has been explicitly subject to questions of legitimacy and democratic accountability in ways that social work power has not. Before going further, it is important to clarify in theoretical terms what is meant by legitimacy in the present context.

Defining legitimacy
In his influential work on the concept, Beetham (1991Beetham ( ,2013 identifies legality, shared values, and consent as the three main criteria against which the legitimacy of power can be tested: Power can be said to be legitimate where it does not breach established rules; where its acquisition and exercise are normatively validated in terms of socially accepted beliefs about rightful authorisation and due performance; and where it is confirmed through appropriate acts of recognition and acknowledgement (Beetham, 1991, p.xiv) The legitimacy of power is thereby judged by how far those in given power relations judge as rightful the authority to which they are subject. Particular attention in this section is given to the shared values criterion, for reasons that will become clear.
The criterion of legality is concerned with the extent to which people acquire and exercise power in ways that are legal and conform to rules. However, legality alone is not sufficient. The law operates in a social context and must therefore be considered alongside community values, meaning that those who claim legitimate power act in ways that protect and promote the identity of society (Tankebe, 2014). To be judged legitimate, the acquisition and exercise of power must be rooted in normative beliefs within a population about the rightful source of authority and due performance in the exercise of power. Legitimacy is thereby authorisation (Tyler, 2003, p.308). The third criterion concerns the degree to which actions by relevant groups acknowledge and confirm the legitimacy of those in power. Such actions serve to enhance or delegitimise authority, including 'performative acts' of consent to power, such as swearing an oath of allegiance.
The criterion of shared values warrants close examination, as it is particularly relevant to the question of social work legitimacy as well as having received significant attention in studies of police legitimacy. Tankebe (2013Tankebe ( , 2014 identifies three aspects of shared values, which are the normative expectations that people hold in relation to, firstly, distributive justice, secondly procedural justice, and thirdly performance or effectiveness. These are important because they relate to the core underlying principle of modern democratic societies, which is that all citizens are deemed to be equal and should not be subject to discrimination. While these principles may fall short when applied in practice, they remain an important yardstick against which to judge legitimacy. Distributive justice is concerned with the perception that the outcomes people receive-such as a decision to prosecute them for a crime-will be fair. It is equally concerned with perceptions of fairness in terms of the distribution of these outcomes between social groupsfor example, by social class or ethnicity. There is overwhelming evidence accumulated over many years, that the police in the UK routinely underenforce the law with some groups while over-policing others, and that this poses a serious problem for democracy (Tankebe, 2013).
Procedural justice relates to perceptions of fairness in terms of the processes that are undertaken to reach a specific decision or outcome. The focus is on 'what fairness means to people in the community' (Tyler, 2003, p.352). There are two dimensions to procedural justice, one being the quality of decision-making and the other being the quality of treatment that people receive. In the rapid growth of empirical research on police legitimacy in the past fifteen years, community surveys have been a key tool (Tankebe, 2014). Many data are yielded by large-scale public surveys, such as those conducted annually by London's Metropolitan Police (Stanko et al., 2012). Research has consistently shown that people attach particular importance to being treated with dignity and respect in their interactions with the police (Loader and Sparks, 2012). These experiences 'communicate to citizens information about their standing and membership in society' (Tankebe, 2014, p.8). In short, in terms of judging the legitimacy of authorities, people are sensitive to how they exercise that authority, and this is especially important in convincing people that authorities are entitled to be obeyed.
In contrast to the police, survey evidence in relation to social work is scarce. What little there is suggests that the general population is either relatively indifferent to questions of confidence and trust regarding social services (McCulloch and Webb, 2020), or hold a confused or negative view linked to media portrayals of the profession (Penhale and Young, 2015). This difference almost certainly reflects the fact that the police are familiar public figures, and most of us will experience crime and have contact with the police at some point. In contrast (setting aside media portrayals), child protection social workers are less familiar, yet the prospect of social work intervention remains a very salient prospect for some citizens. The question of 'which public's?' view of fairness and procedural justice should be explored in relation to social work legitimacy, and how it can meaningfully be done, is therefore a vital one and is one that is addressed in the discussion section of the article.
The third dimension of the shared values criterion of legitimacy is the perception of justice regarding performance or effectiveness. In policing, as Tankebe stresses, this is not confined to the purely practical or utilitarian measure of the police 'getting results' but relates to a 'normative condition for their legitimacy'in the mind of the citizen, such as tackling local crime (Tankebe, 2013, p.112). Questions concerning the performance or effectiveness of social work are important, but due to lack of space they will be considered in a separate article. In the following sections it is the arguments about distributive and procedural justice in social work that are the focus.

Distributive justice in child welfare
Evidence on the distribution of resources in child welfare exemplifies the questions that need to be answered about distributive justice and the use of power in social work. There are large and growing differences between local authorities in terms of the numbers of children entering state care and placed on child protection plans (Bilson and Bywaters, 2020). Bywaters (2015) reconceptualised these differences as 'child welfare inequalities'. This reconceptualisation has bought into sharper focus the significance of deprivation and thereby the structural factors at play and their implications for social justice. Analysis of the variable rates of care and protection interventions has shown that it is the social and economic circumstances of families that are the main determinates (Child Welfare Inequalities Project, 2017). Children in the most deprived 10 per cent of neighbourhoods are over 10 times more likely than those in the least deprived to experience out-of-home care (Morris et al., 2018a). Analysis of data on ethnicity and socioeconomic deprivation has highlighted the importance of understanding the intersectional nature of child welfare inequalities and the complexity of risks associated with them . As Sayer (2017) argues, inequalities injure people and their negative effects are probabilistic.
In their study of large cohorts of children in England, Bilson and Munro (2019) identified an 'investigative turn' in social services, with a 35 per cent increase between 2011 and 2017 in investigations involving children under five years old. They again found wide variation between local authorities and concluded 'that the chances of a family receiving support or being split up are determined by national and local policies, resources and a growing culture of child rescue responses to family difficulties' (p.1). Out of the overall increase in the number of investigations, the study also found an increase in the proportion that proved to be unfounded. Analysis of media and political reaction to high-profile child deaths has underlined the potential importance of such events in the rise of investigative risk-based policies (Warner, 2015).
Added to the evidence of disproportionality in the risk of social work interventions is the question of distribution of resources that might support families. While demands on services have increased, the resources available to support families have decreased under austerity and are set to be limited even more severely during, and in the aftermath of, the Covid-19 pandemic. Yet we know that even modest improvements in income to households reduce the frequency of child welfare reports (Raissian and Bullinger, 2017). Socioeconomic inequality rather than standards of parenting alone is therefore an important determinant of state intervention by social workers. As Loader and Sparks have stressed, 'legitimacy is irrevocably a political concept and one that bears upon the distribution of chances, risks and wrongs' (2012, p.31, emphasis in original).

Perceptions of fairness in the distribution of resources
While there is empirical evidence about actual disproportionality in child protection, questions of legitimacy concern the degree to which these outcomes for families in poverty may be perceived as fair. This begs further, thornier, questions of whose shared values should be considered in terms of their perceptions of fairness. What do different communities feel the balance between 'child protection' and 'child welfare' should be in terms of state involvement in family life? Crucially, we do not have a Social work legitimacy 1175 sophisticated understanding of community values in relation to such questions. Based on research on police legitimacy, we know that a focus on the experiences and perceptions of populations that are 'at risk' of the disproportionate use of state power is essential to addressing questions of legitimacy. In the case of social work with children and families, that would entail gauging the views of families experiencing poverty and deprivation. The urgent need for significant further social work research into these complex questions is addressed in the discussion section of the article.

Procedural justice and social work practice
Questions of procedural justice relate to the quality of treatment people receive and the quality of social work decision-making. The issues here are again complex, and it is important to state from the outset that there is evidence for a mixed picture of experiences by service users (see Ghaffar et al., 2012;Penhale and Young, 2015;Tilbury and Ramsay, 2018, for example). Analysis of the relatively high prevalence of complaints from the public against social workers reveals a multi-faceted picture (van der Gaag et al., 2017). It is important to reiterate that the focus in the present article is on the more negative experiences of social work interventions because the question of legitimacy hinges on the nature of these rather than upon the many positive experiences.
Further, many of the issues in individual social work practice should be understood as reflecting the wider structural and institutional context. It is salutary to note in Morris et al.'s study, for example, that social workers who sought to support families reported feeling overwhelmed by the organisational pressure to focus on individualistic risk factors (2018a). It is precisely for this reason that the argument of this article is not to propose remedies at the level of individual social work practice such as, for example, more or different education and training. Instead, the focus is the wider lens of social democracy and the practical implications of such a focus for research and institutional change-what Loader and Sparks (2012) refer to as 'just ordering', as discussed later in the article.

Perceptions of the quality of treatment people receive
Research produced by the alliance 'Your Family, Your Voice' (Morris et al., 2018b) found that families felt they experienced 'cold hearted [. . .] demeaning and hurtful experiences'(p.6) Morris et al. (2018b) found that social work practice with families living in poverty was 'both deeply humane and inhumane'(p.369), and there was moral ambiguity among practitioners. Their understanding of poverty was fused with stigmatising and punitive discourses about the supposed 'underclass' (ibid.). There was a disconnect between abstracted understandings of poverty and practice, including physical recoiling from deprived localities. Yet, at the same time, the physical geographies of deprivation became a 'normative back-drop'by which the families accessing services were judged (p.367). Gibson (2020) found that social work practice in the child protection process often evoked or exacerbated feelings of shame and humiliation in the parents he observed or spoke with. Parents in another study experienced social work interventions as denying them recognition and respect through othering, a deficit-based model of practice, and a failure to understand families' situation from their perspective  Given that the value base for social work heavily endorses an empathic response to human suffering in arenas such as child protection, it is surprising to find that empathy in practice appears to exist at relatively low levels (Lynch et al., 2019).

Quality of decision-making
Given the 'moral muddle'that appears to characterise the use of social work power in the child protection system (Morris et al., 2018a, p.371), how can we be confident about the quality of decision-making? One answer is that social work brings expert knowledge. However, when the basis for this assumption is critically examined, we find a complex, discursive interaction between lay and professional knowledge (Scourfield and Pithouse, 2006). The nature of everyday social work practice-involving fluid situations, partial information, and intense pressurefavours rapid, intuitive decision-making rather than deliberation (Saltiel, 2016). Scourfield and Pithouse argue that it is lay assumptions that dictate the choice of theory and practice, with certain theoretical perspectives privileged above others. This is particularly problematic if, as Tunstill has argued, pluralist models of knowledge creation have been discarded in favour of an ideologically driven 'professional social work knowledge industry'(2019, p.73).
Important insights into the quality of social work decision-making are found in court judgements. Here, the judiciary determine the outcome of applications made by local authorities, including applications for the removal of a child from their family to place them into the care of the state. While some judgements are positive and praiseworthy, others are critical. In his judgement in Re A (Child A) (2015) EWFC 11, for example, Lord Munby described the local authority's application for care orders as, 'a tottering edifice built on inadequate foundations' (Para.11). It is in court judgements that we are given a strong sense of the explicit connection between social work decision-making and the community values that underpin democracy, at least as perceived through the judicial lens. These values are most cogently expressed through what are known as the 'threshold criteria', as expressed in this oft-quoted judgement from Hedley J in Re L (Care: Threshold Criteria) [2007] 1 FLR 2050, para.50: Basically it is the tradition of the UK, recognised in law, that children are best brought up within natural families. . .It follows inexorably from that, that society must be willing to tolerate very diverse standards of parenting, including the eccentric, the barely adequate and the inconsistent.
Tolerance for diverse standards of parenting is therefore a defining feature of a democratic society. In Re A, Munby went on to pinpoint a range of parental behaviours that may not, in and of themselves, warrant removal, including, 'people who commit crimes, who abuse alcohol or drugs, who suffer from physical or mental illnesses or disabilities, or who espouse antisocial political or religious beliefs'. (Para.15). Similarly, he pointed to the 'tens of thousands of children who are cared for in homes where there is a degree of domestic violence (now very widely defined)'(Para.16) The complexity of responding to diverse standards of parenting is made all the more so by widening inequality, globalisation, and social changes that 'expose a new and more complex map of exclusions, affiliations and frictions '(Loader and Sparks, 2012, p.22).
Crucially, the principle of tolerance for diversity should not be read simply as code for abandonment by the state. On the contrary, the UK law is equally clear and forthright on the proper role of the state in its obligations to provide appropriate services and support. Local authorities must provide 'requisite assistance and support' to enable families to continue to care for their child. These legal principles reflect the civil, political and social rights of citizenship that are equally part of a functioning democracy; welfare rights being those that flow from social and economic rights.

Discussion: a democratising agenda for social work research, policy and practice
The focus in this final section is on three domains where the questions raised about social work legitimacy might be addressed. First is the demand for new knowledge through research about shared values; secondly, the need for different institutional arrangements; and thirdly, establishing conditions for social work practice that involve a democratic orientation to community life. Lack of space here means that only indicative suggestions can be described in each domain, but the overarching theme across all three is democracy through deliberation. Deliberation reflects the 'dual and interactive character of legitimacy', involving both power-holders and audiences in dialogue and a constant state of flux (Bottoms and Tankebe, 2012). Processes of deliberation may be fraught with difficulty but, as Loader and Sparks put it, 'We cannot decline to engage with the emotions we find there ' (2012, p.32).

Researching social work legitimacy
While research on police legitimacy provides some useful models, social work legitimacy has different empirical requirements. These are linked to the task of gaining a sophisticated understanding of the complex array of community values that lie behind the state's role in supporting families and protecting children. When it comes to judging the fairness of the application of threshold criteria and the distribution of resources to families in need, whose shared values should be explored, and how might this be done? Constituencies may be categorised into critical and general groups. Critical constituencies include those with direct experience of social work interventions; for example, families where children have been taken into care. A considerable amount of evidence about the views of some of these parents has already been collected, as discussed earlier in the article. Organisations such as ATD Fourth World and the Family Rights Group dovetail research with their campaigning work on behalf of families. This approach is consistent with other major arguments about social work and social justice concerning the politics of recognition and respect, specifically in response to people living in poverty .
Another, wider, critical constituency comprises those families that are known to be disproportionately 'at-risk' of social work intervention. This includes, for example, those who are experiencing deprivation, those parents who were themselves in care and children who are looked after or who have otherwise experienced social work interventions. The salience of questions aimed at these groups, concerning their perception of the fairness of social work decision-making, is likely to be far higher than the general population. However, gauging the views of the general population (encompassing those in critical constituencies too, of course) remains vital if social work decision-making is to reflect a dialogic legitimacy with respect to community values.
In terms of methodological approaches, there is space here to focus on just three of many possible options. One is the standard survey approach, as widely utilised in researching police legitimacy. However, as already discussed, general public surveys on social work do not appear to have the same resonance for most people that questions about policing do. Surveys about values and attitudes relating to parenting itself are likely to resonate more widely. Qualitative research is likely to be productive in exploring, for example, how key concepts in parenting and child protection are constructed by a range of publics, including those of us in the social work profession. An example is Williams' (2017) research on lay constructions of child neglect, which found that participants were uncomfortable with what they saw as narrow professional definitions of the problem. The use of focus groups in Williams' research highlights the importance of processes of deliberation. The third proposed method takes this further with the concept of deliberative democracy and the use of democratic forums.
Deliberation as a research technique has three main features (Burchardt, 2014). Firstly, it aims to 'uncover the public's informed, considered and collective view on a normative question' (Burchardt, 2014, p.353). The focus is the considered judgements and underlying values that people hold in relation to the given subject. Secondly, it involves the provision of expert opinion to aid discussion, which is a challenging but important dimension of deliberation as a research approach. Thirdly, it is assumed that there may be shifts in the relevant beliefs and values of participants through their involvement in the research. Despite its potential in offering 'uniquely considered, insightful and well-defined answers', deliberative research requires careful design and planning, particularly regarding the value positions of the researchers that use it (ibid. p.353). The key strength of the approach is that it specifically facilitates a deliberative and collective form of 'slow thinking' (Stoker et al., 2016, p.3) about complex issues rather than knee-jerk, 'intuitive' responses. Processes of deliberation link closely to proposals for democratising policies, where deliberative institutions chart a course through epistocracy -rule by knowers -and emotive populism (Loader and Sparks, 2012).

'Just ordering' in social work policy
In their article advocating 'institutions of just ordering', Loader and Sparks (2012) argue for 'deliberative institutions and mechanisms that help make good the claim that the decisions of actors are legitimate in the eyes of those affected by them' (p.32). In policing, Police Independent Advisory Groups were established to address the detachment of local police from communities and to reflect the idea of 'critical friendship' (Dixon, 2020). Rather than direct public accountability, these groups were to offer a form of 'answerability' by the police to the communities they serve. For social work, a form of answerability may be achievable through an enhanced role for local authorities. Local authorities already comprise democratically elected councillors who are accountable for the delivery of services, including social work, to local people. According to polling conducted in June 2020 on behalf of the Local Government Association (LGA), the Covid-19 crisis has led to significantly increased levels of public trust in local councils. If a different democratic contract between citizens and social work is to be achieved, then the local authority will need to be a major broker. But would local authorities want a deeper role such as this? The LGA's answer appears to be that it would: We are not seeking to run everything, and our residents don't want us to, but we do have the democratic legitimacy to shape and lead a joint endeavour locally. By meaningfully empowering place leadership councils can play a central part in giving hope to communities, offering them the levers to level up inequalities and shape their future. In short, people trust local decision making more than they do national. (2020, p.4) The nature of 'empowering place leadership' in the context of the present arguments relate to the need for processes of deliberation that can engage the views of citizens and social workers on how social services should respond to the needs of the local population. This places pressure on social work to give an account of its practices 'before the eyes of the people' (Loader and Sparks, 2012, p.32) The importance of employing social workers who reflect and can represent the diversity of the communities they serve comes into the foreground here.

Street level democrats? Democratising social work practice
The relationship between democracy and social work has a long and compelling history (Toft, 2020), but in the work of Jane Addams we find detailed accounts of the connection in vital form. Far from being purely abstract, there is a pragmatic focus on the need for proximity to those living in poverty, such as through the Settlements. As Nackenoff (2009) has argued, Addams' ideas throw down a challenge to liberal middleclass conceptions of democratic engagement that are largely confined to comfortable, safe spaces (p.127). Addams articulates, 'the desire to make the entire social organism democratic, to extend democracy beyond its political expression' (Elshtain, 2002, p.15).
While Settlements may not be a practical option for contemporary social work practice, there are examples of practice, both nationally and internationally, that are more consistent with the values of democratic egalitarianism. Just one case where the potential seems clear is Family Group Conferencing (FGC) as used, for example, in the Family Valued programme in Leeds (see Mason et al., 2017). The programme involved a restorative justice model and the expansion of FGC. Family Valued engaged a wide range of partners beyond Children's Services and addressed gaps in service provision. There were statistically significant reductions in the numbers of child protection plans and looked after children. Families that participated in the evaluation study reported that their values had been respected.

Conclusion
This article has questioned the degree to which the power exercised by social workers in child protection can be considered legitimate. Questions of legitimacy lay social work open to the challenge of identifying the social democratic mandate for its decision-making. Paradoxically, by questioning social work's legitimacy at this fundamental level, we may yet find new ways to realign practice with the values of human rights and social justice that are said to underpin the profession. As Addams urged more than a century ago, we must 'place the democratic aim of social justice firmly within social work' (Toft, 2020). Such a realignment is particularly urgent, given the socioeconomic impact of the Covid-19 pandemic and its implications for a large and increasing number of families. I argue that the concept of legitimacy offers new ways for social work to articulate its values as a profession by advocating boldly and explicitly for the civil, political, social and economic rights associated with social democracy.