A Critical Discourse Analysis of the New Labour Discourse of Social and Emotional Learning (SEL) Across Schools in England and Wales: Conversations With Policymakers

This paper reports on a critical discourse analysis (CDA) of the New Labour (1997-2010) discourse of Social and Emotional Learning (SEL) in schools, and how it was understood and enacted by policymakers in England and in Wales within the context of devolved government across the UK. By SEL I mean universal school-based programs, located in the concept of children’s emotional wellbeing and concerned with developing children’s ability to understand, express and manage their emotions. Through a series of in-depth interviews with national level SEL policy actors this investigation worked to understand and identify how the SEL policy discourse worked “to privilege certain ideas and topics and speakers and exclude others” (Ball, 2008, p. 5). Through shining a critical light on the discourse of New Labour SEL policy makers this work responds to the call of Gunter et al. (2014) for critical education researchers to develop and amplify voices which New Labour discourse of Social and Emotional Learning (SEL) in England and Wales 2 challenge the current hegemonic “common sense” discourse. The findings identify that the relationship between language and political ideology in England and Wales during the New Labour years powerfully shaped the SEL policy discourse.


Introduction
Through the voices of policymakers and utilizing the lens of CDA this study demonstrates how the SEL discourse reflected notions of national identity, particularly issues of social justice (Fraser, 1997), the conceptualization of children (Watson et al, 2012), and the understanding and impact of the neo liberalization of educational policy and practice (Lakes & Carter, 2011).My thinking regarding national identity was influenced by the work of De Cillia et al (1999) and the notion that national identities are "discursively, by means of language and other semiotic systems, produced, reproduced, transformed and destructed" (p. 153).
This work was undertaken from a critical realist perspective (Bhaskar, 1998;Fairclough, 2013) in that it advocates that knowledge of reality is mediated by our perceptions and beliefs.It is a perspective that sits comfortably with a study seeking to reveal the power relations and inequalities pulling and pushing the discourse, texts and voices within a particular event.Moreover it is also interested in what was not said or omitted within the SEL discourse as well as recognising the "reality" of the material social practice (e.g.money, jobs, school buildings, and education staff) within the SEL story.
The study was driven by three research questions: how was the development of SEL policy in England and Wales framed by New Labour?What was the experience of policymakers who sought to develop and deliver SEL policy in England and Wales and how did national tradition, history and identity (including versions of childhood wellbeing) influence the discourse and policy of Welsh and English SEL under New Labour?
This paper is interested in the doing of CDA and for this purpose the process of reporting the study will take the following form.I shall start with briefly setting out, from a critical perspective, the key themes, thereby locating the study.This will be followed by an examination of my positionality and how this shaped the study.Next will be my methodological approach where I will present a robust reflection on the thinking with theory that supported this study and acted as the oil ensuring the smooth operation of the nuts and bolts of applying CDA through Fairclough's Three Dimensional Model (1992).This will be followed by a detailed and critical reflection of the study's participants and the interview process.The subsequent sections will present my analytical process leading the reader onto the findings and discussion in which I shall draw across the study's data to illustrate, in this instance through the application of interdiscursivity, the merits of approaching such a study through the lens of CDA.The paper's conclusion will place particular emphasis on how the doing and findings of this study can inform future critical education (policy) research and the field of SEL.

SEL
Social and Emotional Learning (SEL) is an umbrella term for a wide range of concepts, skills, dispositions and attitudes within UK education policy and practice.The conceptual and theoretical apparatus has been operationalized through policy programs such as the Social and Emotional Aspects of Learning (SEAL), Promoting Alternative Thinking Strategies (PATHS), Second Step, Creating Confident Kids, and the UK Resilience Program (Challen et al., 2011;Humphrey, 2013).The term SEL refers to both those programs focused on positive models (emphasising social and emotional wellbeing) and those premised on deficit or negative models that seek to repair or develop responses to socially problematic issues such as depression, anxiety and anti-social behavior.
As SEL has developed in the UK into substantive policy, the terminology surrounding it has morphed.Many early SEL programs (from 1999 onwards) were launched under the academically credible headline concepts of 'Emotional Intelligence' (Goleman, 1995) or 'Emotional Literacy' (Sharp, 2000).By the mid-2000s this had been transposed into 'Social and Emotional Learning' and in praxis the 'Social and Emotional Aspects of Learning' (SEAL).Social and Emotional Aspects of Learning was the prescriptive (taught and caught) program, managed in England by the National Strategies which delivered New-Labour-in-England's universal version of SEL to state schools.The SEAL approach used an objective list model (Seligman & Royzman, 2003) whereby a series of skills and abilities defined by experts (the SEAL steering group) became learning outcomes (42 descriptors), competencies that could be measured and assessed.At the elementary/primary school level one of the learning outcomes was, "I can make, sustain and break friendships without hurting others" (DfES, 2005, p. 93).
The SEAL program, as a government-created initiative, grew to become the dominant vehicle for SEL across England.By 2010 the program was operational in 90% of primary and over 70% of secondary schools in England (Humphrey et al., 2010).It was so dominant that for England, when we talk about SEL and New Labour (in England), we are talking about SEAL.
As SEAL was rolled out across England it was picked up by some Welsh Local Authorities (Conway for instance) and it is clear that the SEAL program was delivered in a number of Welsh schools.However it was not distributed and consumed in either the same manner or with anywhere near the same number of participants.In contrast to England, New Labour in Wales never developed a universal SEL intervention but rather linked social and emotional learning to broader issues of social justice and challenging inequalities.During the early years of devolution SEL in Wales was driven through the Learning Country (2001).This was a key policy document in the Welsh SEL story for it acted as a lens through which SEL was understood and as a location for it to reside in the policy process.

New Labour Education Policy
Westminster-based New Labour policy activity (1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010), dominated by the twin forces of competition and standardization (Apple, 2007), resulted in an incredible rise in the raw number of education interventions enacted.In 2006, Coffield remarked that the "depth, breadth and pace of change" and the "level of government activity" in education was historically unprecedented in the UK (Coffield et al., 2008, p. 2).The New Labour (in England) marketization of education, propped up and bolstered by a discourse of rapid neoliberal modernization, has been criticised by many commentators such as Fairclough (2000), Adnett & Davies (2000), and Davies (2000), who have argued that a policy model premised on the overarching value of market-like choice is "a triumph for class politics, for the power of the British middle-class to corner what is best for its children, much of it disguised as the exercise of parental choice".
If New-Labour-in-England's education policy was a continuity with the New Right (Ball, 2000) Welsh education policy used a fundamentally different discourse (Phillips, 2003).This was the period of First Minister Rhodri Morgan's Clear Red Water speech (2002).Delivered at Swansea University, Morgan's speech was a powerful and passionate public declaration of Welsh otherness.Built on a Welsh identity of progressive social democracy and Welsh socialism (with a small s) Morgan set out a clear progressive consensus for public policy in Wales focused upon a collaborative relationship between the state and its citizens whereby the state should and could intervene to support the collective wellbeing of the nation.
The actions of the Welsh Assembly Government clearly owe more to the traditions of Titmus, Tawney, Beveridge, and Bevan than those of Hayek and Friedman.The creation of a new set of citizenship rights has been a key theme in the first four years of the Assembly -and a set of rights, which are, as far as possible: free at the point of use, universal, and unconditional (Morgan, 2002).

English and Welsh Policymakers
Prior to 2001, Welsh education policy was dictated in Westminster by the UK government.After 2001, following formal devolution and the establishment of the National Assembly for Wales, a large degree of new found freedom was given to the Welsh elected representatives to shape and deliver education policy away from Westminster.As the emphasis of this study is on the New Labour administration (1997 -2010), it is worth noting at this point the following observation on educational policy in England and Wales: (New) Labour in England, dominated by the discourse of modernization, is prepared to utilize the market, selection and specialisation to achieve its aims, its colleagues in Wales use more conventional (to Wales) reference points in labour tradition such as community, locality and universality.(Phillips & Harper-Jones, 2002, p. 303) All of my interviewees were influential in the New Labour SEL discourse; essentially they were key policy actors.Drawing on Fairclough (1995Fairclough ( , 2000) ) Gunter and Forrester (2010) and (Ball, 2008), I understood policy actors as socially situated, complex and fluid (rather than fixed) individuals, organisations and networks -state and non-state, profit and non-profit -"whose identities and professional trajectories are often bound up with the policy positions and fixes that they espouse" (Peck & Theodore, 2010, p. 170).This included government officials, consultants, business representatives, intellectuals, academics, researchers, steering groups, think tanks, voluntary sector representatives, advisors and lobby groups (Ball & Exley, 2010).

Positionality
Bringing the Self to the Study I engage here with a critique of SEL policy in English and Welsh schools from the vantage point of an insider (Burns et al., 2012).I have a long professional history within this field.From 1995 to 2010, I was as an education consultant closely connected to the New Labour agenda (Fairclough, 2000).A clear rationale for this study is therefore to give voice to individuals whose job it was to concretise policy ideas and Ministerial proposals.
In my capacity as a SEL practitioner I worked on the formation of policy for both local and national governments, including work on the creation of SEAL (Social and Emotional Aspects of Learning).I specialised in SEL and conflict resolution and wrote one of England's first SEL qualifications: the Personal Employability, Achievement and Reflection for Learning (PEARL) framework.Alongside these activities, between 2007 and 2010, I worked as a policy advisor to the Welsh Assembly Government, developing a national SEL framework as part of the Demonstrating Success policy program (WAG, 2007).Engaging in these activities, building a large network of influential colleagues and having access to senior policymakers in both England and Wales, has given me a unique opportunity to witness policy and ideas in practice.
My relationship with policy actors in England and Wales meant that I knew the key players; I had relationships which granted intimate access; I had experience of insider conversations and activities; I knew the discourse and language of SEL and I had been part of the policy machinery in writing SEL materials for government bodies.Ultimately there was a period where I was at or near the centre of the SEL community in both locations (England and Wales).Yet participating in SEL alone was not enough to warrant deep academic exploration.It was through my entry into a critical education community that I came to understand and begin thinking with theory.In discovering critical theory at the University of Manchester, Critical Education Policy and Leadership Studies (CEPALS) group I found a safe space, supportive colleagues, and relevant theorizing to reflect on and explore the thinking, conversations, and activity I had with SEL policy actors.This led me to understand that what we (policy actors) talked about and what we did was part of a much deeper and powerful discourse, the New Labour neoliberal agenda and, in particular, the rise of New Managerialism (Apple, 2007) in education.
My professional experience in the field of SEL policy led to a personal and poignant epiphany, fully detailed in Gunter et al. (2014).Over time this resulted in my transition from being an ally and agent of the New Labour SEL discourse in England to being a critic, challenging the commodification of children's social and emotional states (Timimi, 2011) and questioning the processes by which the marketplace and "common sense" have dominated SEL discourse.

Unpicking Positionality
Mercer's 2007 paper exploring the challenges of the researcher investigating places they have worked in, considers that the insider researcher wields a double-edged sword whilst treading a series of delicate paths.I would concur with this but also add that my status as both an insider and an outsider further complicated this experience.On the one hand, I used my insider history, connections and knowledge to ensure that positive contact was made with the participants that would be likely to result in the interviews being granted.On the other hand, I was also aware that my growing outsider status as a critical voice in English SEL would be likely to raise challenges and barriers to gaining access to previous English SEL colleagues.Finally, I also had to tread Mercer's delicate path ensuring, for methodological purposes, I held onto a sense of outsider status, the researcher peering inside.
As a practitioner witness and active participant I believe it is important for my position as a critical voice to be clarified.I advocate strongly for SEL to have a place within the English and Welsh education systems.It is the current, one size fits all, objective list model and the hidden (neoliberal and marketization) discourses underpinning SEL policy that I object to.I have engaged with this study with my assumptions, as Kincheloe & McLaren (2005) would say, on the table.These assumptions and precepts include: the belief that there is a difference between the SEL discourses in England and Wales and the contention that this is in part due to the discrepant history, identity and value systems of these two countries.This standpoint, underpinning the study, thus sees there being quintessentially different political forces, knowledge claims, validity assertions and power dynamics at play in England and Wales.These differences "regulate" the discourses taking place in the interface between ideology, ideas and practical policy.
In bringing this paper, drawn from a doctoral thesis, to publication two particular challenges have focused my attention.The first of these is how to attain, within a journal paper, a level of depth regarding my approach and perspective alongside a thickness of analysis that will confidently respond to Antaki's (2003) critique of six analytical shortcomings common to (critical) discourse analysis.The second is the perennial question of researcher bias (Wodak & Meyer, 2001) regarding both my approach to the study and to my management, interpretation and presentation of the data.
In response to the Antaki challenge I have endeavoured, where applicable, to illuminate some of these six factors throughout this paper and detail my response within the text.However, it is important to state that although I agree with much that Antaki has to say, I have some concerns regarding the sense of a presupposed right way of carrying out CDA.I particularly question who judges what is under analysis, for example, through summary or isolated quotation.There is a danger that if this concept is pursued, in doing so one constructs a right level of analysis (Burman, 2004).This is particularly problematic; for who is to judge this and against what criteria?
In relation to researcher bias and possible claims of selecting participants and subsequent text that would support my preferred interpretation, the context of justification, I again address this where possible in the paper.However, I would ask the reader to access the full thesis and, in particular, my concluding comments where I acknowledge that in the thesis, as I understood my data and saw it reveal powerful stories, I developed a greater understanding of the complexities of how policy is formed and the contingent human emotions connected to this process.By undertaking this study over time I developed a different perspective from the policy story and experienced directly the transformative possibilities of CDA when one ensures that the analytical process is not contaminated by the position.

Critical Theory
This research is in essence a discourse of resistance grounded in critical theory.I argue that power and discourse, working through what Foucault (1995) termed governmentality, constructed and driven by New Labour in Westminster, created and operationalized a specific objective list model (Watson et al., 2012) of SEL, the SEAL program, designed to reflect and promote English neoliberal cultural values.I contend that New Labour in England, particularly in its second term, adopted SEL as a managerial tool for achieving social justice (as a sticking plaster to the excesses of the free market) as located in and understood by a neoliberal administration.Alternatively in Wales, the assembly, similarly in New Labour's second term, adopted SEL as a tool for developing egalitarian social democratic notions of social justice (located in notions of a collective community) closely connected to understandings of national identity fuelled through the devolution process.This paper thus views education as a social institution with an inherently politically active function in society.
Understanding education requires that we situate it in the unequal relations of power in the larger society and in the realities of dominance and subordination -and the conflicts -that are generated by these relations (Apple, 2010, p. 152).
Apple's succinct words comprised a clear rationale for this study's methodological predisposition and encouraged me to approach this study through CDA as it is designed to uncover power relations, highlight inequalities in society and uncover injustice via analyses of texts and subtexts (Corson, 2000).This study was therefore a political act and counter-neoliberal statement as it not only sought to reveal injustice, therefore wrongdoing, in practice, but it also sought to serve as an interruptive act (Apple, 2009).It aimed to question, critique and confront the dominant neoliberal discourse which presided over public policy and political debate in England and which has increasingly had an impact on Wales and its commitment to social democratic principles(see Emery, 2014, for more about this process).
I act in this study as a critical secretary.This is a term used by Apple (2007) to describe he/she who sheds light on the ways in which education policy is connected to issues of power, subjugation and ideational domination.The critical secretary's role is to give voice to those who challenge existing dynamics of unequal power (Apple, 2010;Au, 2012).This is a study of neoliberal power and identity.This, however, is not a simple binary dynamic of England-versus-Wales, for Wales itself is, as with anywhere, a loci of ideological conflict as well as a broadly radical opposition to English neoliberal discourse.

Critical Discourse Analysis
This work was undertaken using a CDA approach.I specifically state approach here as I see CDA, as do Fairclough (1992) and Wodak (1999), as an approach rather than a methodological "package" or "framework".CDA enables researchers to deepen their understanding of the world we live in, the notions of knowledge and truth which make this understanding, or other understandings, palatable and coherent, and how these truth/knowledge claims can be deconstructed and/or reconstructed.CDA is thus "an interdisciplinary approach to the study of discourse that views language as a form of social practice and focuses on the ways social and political domination are reproduced by text and talk" (Fairclough, 1989, p.20).
CDA is highly useful for exploring the contradictions, power structures and ideological tensions that exist within educational systems and policy development (Rogers et al., 2005).The data collected and analysed herein flags up precisely these kinds of tension and power relations.CDA enabled me to theorise SEL discourse and its relationship to the wider educational policy framework and political system.As a tool it empowered me to identify contradictions in macro and micro power dynamics, expose stark inequalities and assess the impact of cultural and historical forces on the creation and application of SEL policies in England and Wales.

Fairclough's Three Dimensional Model
This study employed Fairclough's Three Dimensional Model (1992, p. 73) of Critical Discourse Analysis.This model was specifically developed by Fairclough in order that analysts would have a standardized framework through which to undertake empirical research.According to this model every discourse has three elements: a text (writing, talking, images, symbols); a discursive practice (the production, distribution and consumption of the text for example the SEAL policy paper) and the social practice (the social events and activities taking place in the society which the discourse represents).

Figure 1 Fairclough's three dimensional model
In relation to the methods I employed there is much debate within the CDA community about the value and appropriateness of utilizing standardized or systematic tools of analysis (Rogers et al., 2005).Some commentators contend that a structured approach poses challenges to the epistemological principles and universalistic urges of CDA (Bucholtz, 2001), whilst others focus on the absence of a systematic approach (Martin, 2000) and the lack of attendant objective rigor in the analysis, using these particular factors to justify claims that analysts "search their data for examples of what they are trying to prove" (Rogers et al, 2005, p. 379).
In utilizing Fairclough's Three Dimensional Model I believe I have presented a systematic approach that brings together the macro and micro analysis of language use in context.Based on knowledge and experience, the approach I undertook was the most appropriate method to best examine the problem under investigation.It supported me in applying an explicit analytical framework to the data thereby facilitating better data management of a large data set and repudiating claims of simply finding what I originally sought for in the data.This position also responded to Rogers et al (2005) call for CDA researchers to present "clear analytic procedures outlining the decision making of the researcher" (p.387).

Application of the Model
Fairclough's model has no traditional starting point with some authors beginning with the social practice and others entering at the text stage.Indeed one could state that in defining the need for a study to be undertaken in this subject area one is already making use of the third dimensionsocial practice.It is a desire to address an issue of social practice (the neoliberal commodification of SEL) that drove this research.

Text
At the practical level, textual analysis requires the researcher to look at the linguistic features of the text (my interview data) or the inside elements (the words and structure).I did not have the space or time to undertake an in depth micro linguistic analysis; and since this would not be in the tradition of CDA, I placed the focus on four elements of the text.Firstly, nominalisation, turning a verb or other word into a noun; secondly, collocation, a sequence of words that occur more often than expected or by chance, one word juxtaposed with another; thirdly, representation, the relation between two entities, how x is different from y, and finally identification, personal undertaking, commitment or judgement, how the speaker commits to the text.This approach is supported by Jorgensen and Phillips (2002) and other commentators who note it is not necessary to use all the methods.

Discursive Practice
According to Fairclough (1992) the stage discursive practice "focuses upon processes of text production, distribution and consumption" (p.71).This stage formed the main body of the analysis and through interdiscursivity is the element reported on in the Findings and Discussion section of this paper.Discursive practice is particularly concerned with how texts (both verbal and written) are interpreted, used and reproduced; one could call it the journey of the discourse from creation to public interpretation.Integral to this analysis was the role of intertextuality and interdiscursivity (how different discourses are related and connected).Both of these factors needed to be explored and understood in order to capture what Fairclough (1992) calls the historicity of the text, "historically transforming the past -existing conventions and prior texts -into the present" (p.85) and what discourses they draw upon.For my purposes this part of the analysis required taking the interview data and identifying examples of where the respondents were either a) talking about how the SEL discourse was produced, distributed and consumed (their interpretation) and b) how they contributed to the production, distribution and consumption (their activity and my interpretation).

Social Practice
Social practice is the final element of the three dimensional model.Jorgensen and Phillips (2002) identity two particular components to this activity: firstly the relationship between discursive practice and the order of discourses taking place and secondly mapping the non-discursive social, political and cultural structures and factors that create the wider context for the discursive practice.Fairclough (1992) calls this the social matrix.This stage is commonly referred to as the explanatory stage for it allows the analyst to draw conclusions on how the discursive practice both constitutes the social world and is constituted by other social practices; texts shape and are shaped by social practice.Issues of ideology, power and governance can therefore be illuminated through this process for as Janks (1997) notes this level of analysis allows the researcher to explore how the text is positioned or positioning.Whose interests are served by this positioning?Whose interests are negated?And what are the consequences of this positioning?(p.329).

The Participants and the Interview Process
The Participants As this work was always going to be as much a reflection on my journey as it was the story of policy actors and the SEL discourse it was clear from the start that as a qualitative, critical theory, study I would be listening to these policy actors and myself.The sample was therefore purposive (Ritchie et al., 2013).I chose to talk with policy actors with whom I had some form of a connection, ranging from close and friendly to distant and mistrustful yet still a connection.England and Wales were chosen as this is where I had worked and had participated in the SEL discourse.In discussion with colleagues I drew up the following criteria for policy actors' participation: 1) Had been a senior national level policy actor/leader of a policy network that was instrumental (according to the literature and my insider knowledge) to the SEL discourse in England or Wales during the New Labour period 1997-2010.
2) I had either a previous working relationship with them or they were aware of my work in England and Wales and could be contacted through networked relationships.
The SEL policy actor community across England and Wales was not a large pool, and as I had been working in this field since its inception, my knowledge, contacts, colleagues and the literature indicated the parameters of the sample.I was aware from the outset that although the pool of potential participants could be up to 30, in order to be effective for this in-depth CDA study ultimately it would need to contain around 16 names (in expectation that one-third would not take part) and ideally equally divided between England and Wales.It was this group of 16 (nine in England/seven in Wales) who formed the first sample for this study.This initial list of contacts was checked by both my supervisor and expert SEL colleagues in order to ensure that selection bias was not taking place and a representative sample of SEL policy actors were being approached.Interestingly the majority of the names and organisations that I approached, unbeknown to me at the time, have been highlighted in Ball & Exley's (2010) New Labour Policy Networks paper as key policy participants.
From a pool of 11 willing respondents, the final number of people interviewed was nine, as one participant each from England and Wales proved unable, due to diary commitments, to agree an interview time.From this number of nine, the final interview count was eight participants.The ninth interview (English) was undertaken but omitted from the study in order to ensure a sense of parity for England and Wales.This parity in numbers was repeated in regard to range and status with the four interviews in each country featuring a Government Minister, Head of a relevant Civil Service Department (Children, Health, Education), Senior Government Advisor (Academic) and Head of a national SEL campaigning group.All English participants had been directly connected to the national Social and Emotional Aspects of Learning (SEAL) program, the dominant English SEL policy vehicle (Humphrey, 2013).In Wales all participants had been members of the national Demonstrating Success, SEL working group, or in the Minister's case the person this group reported to (Watson et al., 2012).

Interviewing Elites
If one accepts Lilleker's (2003) notion that elites are "those with close proximity to power or policymaking" (p.207) it is evident that my participants fell into this category.It should be noted that I have some problems with the term "elite" particularly, in its current usage, the simplistic way it is often applied in a powerful/powerless binary and as an unproblematic category.The idea that "elites" can be neatly defined and are static does not concur with my own thinking.I would advise the reader to explore the writing of Smith, (2006) in order to understand my position better.
However, I do recognise and acknowledge that my potential interview participants did reflect many of the qualities attached to the term "elites", and at the level of fluid power all my candidates did hold this at some point.This, of course, raises a number of particular issues including access, power dynamics and control for it is likely, due to their connections, profile and status, that my candidates were well experienced in the interview process and managing of it.A situation that Bateson and Ball (1995) note as the elite's ability to skilfully employ tactics in order to "weave narratives of justification" (p.208).A literature has developed in relation to dealing with these issues and in particular the interview process itself (Woliver, 2002) which I drew on when both seeking and contacting my participants and developing the interview schedule.It should also be noted that I was not an inexperienced researcher, looking upwards (Smith, 2006), but rather one who with my insider history had both an understanding of the issues relating to elites and also extensive interview and media training.

The Interviews
The interviews took place between autumn 2011 and spring 2012.All interviews were conducted at a location decided by the participant.Each participant was sent the themes of the questions to be asked beforehand and a two-hour schedule was agreed.All participants were informed that the interview would be digitally recorded and that they would remain anonymous, and were made aware of the confidentiality guidelines.Each of the interviewees was sent a final full transcript of the interview for member checking.
It is important to note that I came to the interviews at a point, where for the first time, SEAL in England was being publically critiqued (my paper in BERJ, the work of Ecclestone and Hayes, Carol Craig).In Wales the new Education Minister, Leighton Andrews, had published his contentious 20 point plan (Andrews, 2015) moving education in Wales into the standards and performance agenda.These are two important points to note, as these actions were challenging to and critical of what had gone before, particularly in relation to SEL.For my participants one could argue that the fluid forces of power and the discursive genres were shifting again and this was a period, alongside the introduction of the UK coalition government, of policy change.

The Research Questions
The creation of the research questions followed the thinking of Su et al. (2010) and were built from "an attention to what others have said and are saying; attention to what others have written; and attentiveness to one's premonitions" (p.93).One could say that the questions were a process born from my experiences, thinking and discussions as an active member of the SEL community since 1995.The research questions were a reflection of my own journey and identity (Dunne et al., 2005).

How was the development of SEL policy in England and Wales framed by New Labour?
This question was concerned with examining, interpreting and explaining how SEL was imagined, enacted and put into the service of New Labour politicians in England and Wales between 1997 and 2010 (the frame setting (Lakoff, 2004)).As it is concerned with development, the main focus was placed on England between the years 1997 (when New Labour came to power) and 2007 (when SEAL was firmly established in state schools across England, (Humphrey, 2012)) and in Wales between 1999 (when devolution was enacted) and 2008 (when Welsh SEL policy took a distinctly different direction (Emery, 2014)).

What were the accounts of experience of policymakers who sought to develop and deliver SEL policy in England and Wales?
This was focused on identifying and exploring accounts of the policy making experiences of the interviewees during their time as key players in the SEL conversation.Here I wished to gather data in order to illuminate how the accounts of the interviewees reflected, or otherwise, the geography of the land they were operating in (Keep, 2009).

How did national tradition, history and identity (including versions of childhood wellbeing) influence the discourse and policy of Welsh and English SEL under New Labour?
The third question was designed to develop understanding and identify the influences of national tradition, history and identity on the SEL communicative event in England and Wales.My thinking here, supported by the work of Fairclough (1995) and Wodak (1999) was that at the social practice level, discourses serve the construction and reconstruction of national identities in part through which version of the nation and its attendant cultural, social and political components is talked about.In illuminating this process as a CDA activist my intention was also to reveal and question the configurations of power at play in the construction of concepts of nation particularly in regard to Welsh devolution.

The Interview Schedule
May (2011) contends that, "interviews yield rich insights into people's biographies, experiences, opinions, values, aspirations, attitudes and feelings" (p.120).As this was a qualitative CDA study, I chose to undertake semi-structured interviews (Denzin & Lincoln, 2005).Semi structured interviews sit comfortably with the researcher who is looking for meaning derived from social phenomena and social interaction (Sarantakos, 2005;Talmy, 2010).That is, how did the policy actors interviewed make sense of the world they operated in?For as Marston (2000) states understanding how policy happens and how it is experienced by policy actors is under researched.My interview approach followed that of Potter and Wetherall (1997) who consider that the art of such an interview is ….keeping to the schedule enough to ensure that the topic is dealt with by each participant, but at the same time letting the conversation flow and following up interesting lines of talk as they happen (p.84).
The interview questions were built around the themes associated with the study's research questions and were open ended in order to allow for flexibility and to build rapport between myself and the participant.Pursuing the thinking of Kvale and Brinkman (2009) the interview questions followed a three-stage process of introductory, general and specific.

Interview Thinking
Much has been written on the generation of interview data (Gubrium & Holstein, 1995;Rapley, 2004) with post-modern commentators concluding that the contemporary interview is more about performance, reflexivity, power and identity than it is about delineated roles, objective truth and concepts of neutrality.Of particular interest to me was the co-construction process in regard to how the data was spoken or said from the perspective of the different roles or voices present in the room.
The talk the interviews produced was jointly constructed around firstly our previous experiences, identity and relationship (where it existed) and secondly (influenced by the first) the mutually co constructed dominant interests or common ground that we pieced together as the interview progressed.I have already noted my insider/outsider status with the interview participants and do not wish to revisit this.However, what has not been made explicit is that my status (connected to experiences, identity and relationships) from my perspective seriously shaped the talk that took place, the co-construction process.Flick et al. (2004) identify two types of attitude that the interviewer may have (influenced by their insider/outsider status) towards the interview process.The first is termed a feeling of exploitation, the second a feeling of happy coincidence.In the first instance the feeling of exploitation derives from the interviewer carrying a latent guilty conscience that the interview is being undertaken for their own personal benefit (completing a thesis, project, work activity etc.).In this case Flick et al opine that the interviewer is likely to be wary of pushing the process through asking challenging questions, taking too much time or getting close to the interviewee, which may result in a "thinner" engagement and resulting data.In contrast, the alternative approach or feeling of happy coincidence considers the interview is a beneficial coincidence whereby the interviewer's curiosity meets the desire of the participant to have their voice heard through an enriching experience.In this instance it is more likely that openness, engagement and a greater depth of data will be generated.
These categories clearly have value to understanding the early formulation of the interview activity.However, I am cautious about the somewhat simplistic, binary nature of the terms.One can certainly see that the different shades of my insider/outsider status and how this was interpreted and understood by the interviewee and expressed by myself impacted on the interview activity.For the purposes of giving greater clarity to this process I therefore propose a third category to Flick et al.'s attitude types, the feeling of happy exploitation.I would categorize this attitude as one where the interviewer's (my) curiosity met the desire of the interview participant to be heard but I still carried, due to different shades of the insider/outsider relationship, certain elements of the exploitation process.Of course, one has to be careful here as I believe my insider/outsider status was constantly ebbing and flowing (very rarely a static thing).However one can see across the relationships particular elements that were more dominant indicators of my status to the participants.This third position can, I believe, be applied in my case to the interviews where I knew the participant professionally (insider) but we were both aware that my position in the SEL community had moved from one of insider to an outsider.

Situated Nature
Finally, all of the interviews are situated in a particular context(s).Holstein and Gubriom (2003) state that "Meaning is not constantly formulated anew, but reflects relatively enduring local contingencies and conditions of possibility" (p.149).It is these local contingencies and conditions of possibility that I wish to consider in the light of the Welsh/English element of my interviews.With the Welsh interviews I was asking the interviewees to talk about a period in Wales that was strongly driven, particularly in the early post devolution years, by a "Wales is different" discourse (Hawker, 2009).This concept of difference was understood and developed through Wales's relationship with and to England.All of the Welsh interviewees talked about England, generally in the form of the "other" (physically, politically, and socially) a "them" and "us" in binary opposition.This usage of "England" and how Wales was articulated in relation to this was a powerful local contingency in the talking process.None of the English interviewees talked about Wales or any other country.Fairclough (1995) identifies genres within discourse which he contends are "a socially ratified way of using language in connection with a particular type of social activity" (p.14).Genres are, in effect, a form of acting taking place within the discourse.There are many genres in operation and often one can identify multiple genres taking place at the same time.For example, within a university department staff/student supervision session, the range of genres could include conversational, educational, political, managerial, medical etc.It is important to note this for clearly different genres can result in different activity, understanding and outcomes.The order of discourse is the sum of the genres in use and is vital in setting limits on what can and cannot be said (Fairclough, 1992).This thinking is developed and expanded upon by Keep (2005) who in exploring New Labour and the policy narrative contends that Stories that underlie policy serve to structure the geography of the 'land of the possible' wherein policy makers come to determine and then promulgate what they can and cannot seek to do in this field (p. 1).
A final point to note here is the issue of another language, Welsh, being present in the interview talk.All of the interviews in Wales were conducted in English and one must consider whether this approach alone shaped the interview talk in Wales.In approaching the interviews in Wales in this manner, I question whether I was simply propagating an English colonial view of Wales (Day, 2006).I assumed and utilized English as the dominant language yet two of my Welsh interviewees were fluent Welsh speakers and greeted me in Welsh as well as using common Welsh phrases in the talk (phrases that they knew I was conversant with).Examples of this included "bore da" (good morning) "diolch" (thanks) and "te neu coffi" (tea or coffee).Did my English language approach challenge, conflict, or deny their identity?Was the use of simple Welsh language words an attempt to draw their own language into the interview frame or was it done in order to make me feel welcome and an insider?

The Analytical Process Stage 1 -Managing the Data
The transcribing of the interviews was a slow and complex process.Building on guidance published by Wetherell et al. (2001) and Jorgensen and Phillips (2002) I also had much discussion with colleagues regarding the process of transcription and, in particular, issues regarding passages of conversation that were hard to understand due to recording issues, background noise, etc.Where this happened, the transcription was marked as inaudible and a time noted in order that I could return to this brief distortion to try and decipher it as the analysis developed.At this stage I undertook what would be considered a detailed and full transcription.All of the conversation, the to-ing and fro-ing, including utterances, pauses and identifiable punctuation were typed up in the manner of a vertical style, screen play, script.The transcripts averaged just under 10,000 words each.Once the final transcription was completed I read and reread each interview in order to make myself familiar with the conversation.It was this full transcript that was sent to the participants for member checking.
At this point it was apparent to me that the notion that all of this data could be analysed and fed into my findings was unachievable.Therefore, as the analyst, I made a choice based on my theoretical approach and the objectives of the study about which detail should be carried forward.

Stage 2 -Applying the Research Questions
At this stage, I took the final transcripts and reviewed them for data (excerpts of speech) that I believed was relevant to the study's objectives; answering the research questions/themes.This is an approach supported by O' Connell and Kowal (1995) and a number of CDA practitioners (Fairclough, 1995;Rogers, 2004).Data that was relevant was highlighted and dropped into a new file.This was a complex process as I wished to make sure I was not censuring or presupposing the relevance of what the interviewees stated however as with all conversations there were passages that had in my experience less bearing on the research questions.Examples of this would be conversation regarding my travel arrangements, lunch/tea/coffee, etc.Some may consider that my initial management of the data took an a priori position as I vetted the transcript in order to remove what I considered to be data not directly relevant to the research questions.This process did, of course, result in possible informative aspects of talk being lost at an early stage of the analysis process (Antaki et al., 2003).In defence of this position, I draw the reader to the following points.Firstly, this study was undertaken from a critical realist perspective and although it acknowledges a complex reality, its critical essence goes a step further by claiming that the acceptance of differences of perception as equally legitimate does not take into account the factors that privilege one version of reality over another (Mertens, 2010).
As a trained CDA practitioner with extensive knowledge and experience in the SEL field I approached and managed the data by bringing my intellectual and pedagogic skills to the question of what was important data.My intention was to fulfil one of the key tasks of a critical education researcher (Apple et al., 2011), that is to reconstruct knowledge to serve progressive and transformative social needs.Secondly, some of the omitted data I did, in fact, return to when thinking critically about the interview process.My thoughts and interpretations at this level are captured in my earlier reflections on the Welsh interviews as well as in the full thesis.Thirdly, the lived reality as a lone, part time, PhD researcher was that I had to employ data reduction techniques in order to successfully manage a large qualitative data set.

Stage 3 -Introducing the Three Dimensional Model
Taking firstly the English and then the Welsh transcripts I coded the data from each transcript through the following lenses: discursive practice (the production, distribution and consumption of SEL); text (grammar, use of nouns, genres and discourse orders used by the interviewees) and the social practice (broader social/cultural/economic/political events).It should be noted that certain aspects of data fell within and across multiple categories.On a practical level to help me with this process I took each interview in turn and color-coded the data from the transcript that illuminated the lens I was applying.
On completion of this editing task, I then cut out the relevant data and placed the paper slip (labelled with the participant's name code for example E (England) 1) into one of three boxes each representing one dimension of the model (text/discursive/social).It should be noted that this physical process was built upon longer sentences and paragraphs rather than atomised individual words.Where a slip was placed in one box but could also be related to or placed in another box then the initials of that second box were written on the slip.This resulted in each interview being literally physically broken down and placed into and across the three dimensional boxes.At the end of this process I had three boxes for the Welsh respondents and three for the English respondents.
The physical breaking down of my data could, I acknowledge, lead to claims of attempting to objectify subjective experiences.This is a claim worth examining as it is representative of one of the great tensions sitting at the heart of CDA, particularly when approached from a critical realist perspective.From my perspective I did not objectify subjective experiences but rather enacted a blurring of the quantitative/qualitative dichotomy (Apple et al., 2011) in order to reveal and offer not a single given truth but rather a contingent socially situated meaning.As Bunge (1993) notes for a critical realist the perception of facts depends on our beliefs and expectations.
Critical discourse analysis is controversial in that it is undertaken in the service of revealing problems, inequality, power dynamics and contradictions (Fairclough, 2000;Wodak 1999) all of which are contestable.It is therefore for the practitioner to make the decision as to what the inquiry shall be and reveal the thinking supporting such an endeavour.Drawing this level of subjectivity down to the level of data management I would contend that decisions made by the analyst, in my case to physically break down the data, were undertaken with the a level of self-analysis and critical thinking commensurate to the task in hand.I was not seeking to turn my data into numbers but rather to reveal the social ordering, shaped and mediated by power relations, and within this the specifically dominant ordering above and beyond the variety of orders enacted.Rather than extracting the data from its context I would contend that I was, in this activity, taking the data on a journey in order to reveal another context to the data.This blurring of the traditional constructivist and realist positions is as Schouten et al. (2007) note not uncommon within a critical realist CDA study.

Stage 4 -Fairclough's Model Directly Connected to the Research Questions (RQs)
I applied each RQ in turn to the three boxes applicable to each country.In effect each RQ was considered against six deposits of data (three for England and three for Wales).My job here was to identify what each element of the CDA process (for England and then Wales) had to offer in regard to answering the research question applied to it.Although each lens had something to say, certain lenses proved to be more applicable to particular questions.For example, the social practice dimension for revealing meaning in the accounts of policy makers' experiences and the discursive practice dimension, particularly through the examination of intertextuality and interdiscursivity, when enquiring how New Labour framed SEL in England and Wales.Ultimately, however, each of the lenses or dimensions were interconnected and reliant upon one another in order to understand the whole.
Supporting this structured analysis each RQ was also summoned to rigorous critical discussion linking back to and drawing on the supporting theoretical literature and utilizing the thinking and theory of commentators such as Apple (2010), Gunter (2011) and Ball (2008).This process was particularly powerful when focusing upon the relationship between discursive practice and the broader social practice, thereby supporting the study in reaching its final conclusions (Jorgensen & Phillips, 2002).

Interdiscursivity
The following section draws directly from the full doctoral study in order to illuminate how the CDA approach and specifically, for this paper, how interdiscursivity (a series of connected powerful whispers) proved an effective tool for unpicking the range of forces shaping the policy makers conversations about SEL in England and Wales.I draw the reader's attention to interdiscursivity as a form of discursive practice for as Fairclough notes discursive practice acts as a bridge linking the text to the wider social practices, thereby revealing in whose service and interests the discourse operates in.For this study locating the interdiscursive mix was a key task in clarifying the relationships and forces at play in England and Wales between language and political ideology.Direct quotes are taken from the policy makers and these are presented in italics with the participant's location defined by either an E for England or W for Wales.

England
My data indicated a complex and high degree of interdiscursivity across the English interviewees.A wide range of discourses were drawn upon and at the higher level all English participants could be categorized as presenting a new interdiscursive mix that combines elements of neoliberal free market business thinking (Hill, 2001), new managerialism (Farrell & Morris, 2003) (particularly performance management), therapy discourse (Ecclestone, 2007) and New Labour in England's education standards and achievement discourse.This new discursive mix employs a high level of nominalisation (vagueness and lack of specificity) and modality (commitment to the truth of a claim) (Fairclough, 1995).One could conclude that ultimately the mix in England was a hybrid managerial discourse concerned with shaping the "other" in the image of itself.This highly interdiscursive mix is illuminated by the statement below.In terms of modality one can see an authoritative categorical assertion of truisms (for example, "I think those white, working-class people who were left behind in those areas would totally lack confidence").
We now have three full generations in the same area without the support, now I haven't sort of got evidence for this, but I think those white working class people who were left behind in those areas would totally lack confidence.I think we felt it was our [government] job to give them [state school children] ambition and to give them the qualifications so that they had choice.E1 Behind this higher level of English interdiscursivity sits a range of lesser but still powerful discourses combining the individualistic, promotional discourse of New Labours' third way alongside a discourse of moral panic (Humphrey, 2013) whereby children are at risk or damaged (a deficit model of childhood (Watson et al., 2012)) and in danger of losing (economic) opportunity when understood in a neo liberal achievement agenda.This hybrid blend of interdiscursivity is exemplified in the following statements from the English interviewees.E3 combines a deficit model discourse with features of the neoliberal "other", framing SEL through a measurement discourse (albeit one built on an assumed truth) that judges "they" don't have what the speaker possesses, in this case social skills.A further neoliberal discourse of individual selfpromotion is also at play whereby poor performance resides not in social, political, economic and educational institutions, but rather in the child, and by extension, the family (Lubeck & Garrett, 1990, p. 327).Both modality and nominalisation are very strong across this statement.In regard to modality an opinion (that there is a lack of social skills) is presented as a categorical fact thereby making it sound authoritative whilst nominalisation is employed through the vague terms "teach them" and "social skills".We are left to populate these terms ourselves.Interestingly one could consider that the interviewee recognises the nominalisation taking place with the "as it were" strategically placed after the first usage.I don't like that word feral and it's … but there is an issue about the lack of social development that some children from perhaps non confident families, I don't think it's to do with poverty it's to do with parenting and I think working class parents are well able to bring up their children but some parents lack that ability or that motivation.E1 The above statement from E1 again reflects the self-promotional, performance driven, achievement and deficit discourses blended together to form a specific, New Labour in England, interdiscursive mix.Strong nominalisation and modality are in evidence with the speaker relating an opinion built on a series of vague and undefined processes (social development, non-confident families, parenting, etc.) as a categorical fact.The dominant discourse here is built around the managing of the "other", this time giving the "other" a name "feral" and locating them as the children of (some of) the working class (Squires, 2006).One could conclude as Lowe (1993) does that this form of discourse is essentially policy or politician as parent.The working-class parent is objectified as feckless and in need of moral guidance (McCaig, 2001) for the deficit in the feral child is connected to poor (or perhaps non-middle-class) parenting skills rather than any wider social economic factors.This is a position and process that Gewirtz (2001: 366) identified as New Labour's resocialisation program, the determination to "make all families like middle-class families".

Wales
In Wales the interdiscursive mix, until the appointment of Leighton Andrews as Education Minister in 2009, was located in and of service to New Labour in Wales, particularly Morgan's Clear Red Water thinking.However this is superseded and itself shaped by the significant and deep change that took place in Welsh politics with the introduction of devolved government in 1999.This was the first time since the 1536 Acts of Union, that Welsh politicians had (limited) direct control over the democratic process and policy making.This change is at the heart of the interdiscursive blend for the Welsh interviewees which brings together a range of transformative, social democratic, national identity, socio-political and social justice discourses.This is an interdiscursive model that places the citizen and the state together as one collaboratively building a new society, grounded in children's rights, on the memories and shadows of past failings and historical disenfranchisement.It is a collaborative discourse concerned with equality of outcome rather than equality of opportunity (Phillips & Harper Jones, 2002).
They [the Welsh Assembly] were absolutely as one in terms of the sort of society they wanted to see… To build for Wales that it is one in which people have the opportunity to achieve their potential that everyone has what are regarded as the basics of life which is they don't live in poverty.W2 The above quote from W2 captures the heady interdiscursive mix operating across the Welsh interviews.A transformative discourse can be seen in the active declaration to "build" a particular form of society, a position that presupposes this vision is not currently in place, thereby connecting to the Welsh otherness discourse.This vision through the use of "for Wales" is wrapped in a nation building discourse.Alongside these discourses sits a social democratic political discourse (opportunity, everyone, society and people) that is both universal and collaborative.Finally, one can see the social justice discourse at play as this journey will result in the removal of poverty.Modality is strong with the speaker, by drawing together a range of discourses, creating a passionate, almost rallying call.There is a clear sense of commitment from the speaker to the statement, for example the use of absolutely, which results in what Fairclough (1995) would term obligation modality.Interestingly, nominalisation, in comparison to England, is weak, with the process of change being connected to specific (albeit slightly vague) tasks and a clear outcome.This creative mix of discourses reflects what Bradbury and Andrews (2010) have called civic Welshness, a perspective that Egan and Marshall (2007) believed placed collaboration, community and the citizen at the heart of New Labour in Wales's policies.I think there was a very strong belief that people's social and economic circumstances determined their life chances… so that was a social justice issue I think , equally if people were to improve the life chances then Wales as a country would prosper as well.W1 This interdiscursive mix in the service of civic Welshness is evident again in the above statement by W1 where the interviewee blends together a transformative, social democratic, collaborative (note the lack of the "other") and social justice discourse wrapped within a nation building frame.
Modality is strong as witnessed by the use of "strong belief" whilst nominalisation is low as the process, building an equitable Wales, is given specific form through the identification of social and economic factors.Essentially the message is that poverty and social exclusion reduce people's quality of life.This is a message acknowledged by Drakeford (2010) who remarks that "When things go wrong in the lives of children and young people, the Welsh focus has been on trying to put right flaws in the system on which they depend, rather than on focusing on the deficits in young people themselves" (p.141).
Interestingly when the talk turned to the later post-2008 period of the Welsh Assembly (The PISA performance crisis and appointment of Leighton Andrews as Education Minister (Andrews, 2014;Evans, 2015)) and the SEL conversation, a different blend of interdiscursivity appeared.This was a discourse utilizing the language of standards, outcomes and measurement (in the thesis I have termed this neoliberal lite).This was a period of questioning and change from a different perspective, one that was critical of what had gone before and as Elaine Edwards (General Secretary of the Welsh Teachers' Union, UCAC) notes resulted in a white knuckle ride for education (Evans, 2015).This was clearly a significant and challenging experience for the interviewees.
I think that probably you know we were doing er ... we thought we were doing had all the right policies but I think up until things like [2006 PISA assessment]  There are no quick wins in the things that that we're talking about.It's generational and to a great extent I think that was understood by that first cabinet.Er so maybe they weren't sharp enough about asking for results early enough and Leighton Andrews has come in education wise and is looking for quicker wins now.W2 One can detect in the above quotes hesitancy, pauses and a sense of reflective questioning in the interviewees.The modality regarding the social justice discursive practice is no longer strong and categorical but in Fairclough's terms at the median to low end.This weak modality indicates a more subjective truth commitment (Fairclough, 2013) to the social justice/SEL conversation thereby creating a new space.I would contend that for the interviewees this space is a construction of the neoliberal standards discourse exerting its power on the land of the imagined.As the interviewees reflect further on the introduction of a standards and outcomes discursive practice and the introduction of Leighton Andrews as Education Minister, one can see that two factors are occurring.Firstly, we get the sense that something is over and secondly, the subjective space is squeezed in a different direction as a series of critical truth commitments are made regarding the process driven, transformative, social justice discourse that shaped SEL .One can sense that the interviewees are attempting here to articulate the experience of being on the receiving end of a particular power.
I don't know why I've never spoken to him about it I mean … you know he sort of just said you you know … it's purely about educational performance which I think it is it is it's going back to the fact that you know you need some sort of delivery er.W1

Conclusion
One of the key and consistent challenges throughout this study, one so concerned with power, language, identity and the emotions, has been the realization and management of these said forces in regard to my own journey.This study at the personal level, as a CDA practitioner, took me on a transformative journey that demanded I critically reflect on my own identity, experiences, emotions and actions and develop a willingness to allow the process to reveal and shape me as it revealed and shaped the data.Key to successfully navigating this journey was my constant checking and lived engagement with the critical theory supporting this study.The time spent thinking with the theory was crucial to my application of CDA for it provided a robust and credible space to which I could retreat, repair and progress both myself and my research practice when the lived experience of doing the analysis hit obstacles.
The manner in which I applied the three dimensional approach and particularly the detailed analytical procedure , including particulars of the linguistic forms associated with the relevant dimensions (normalisation, modality, etc.), presents a level of analytical depth and reflexivity sometimes absent (or unreported) in previous CDA studies (Rogers et al., 2005).In doing so this responds directly to the application critiques of Widdowson (1998), Martin (2000), and others.However I must also acknowledge that in ensuring all dimensions of Fairclough's framework were systematically pursued I placed very challenging time and resource implications on my work.Looking to and across all of the dimensions, ensuring interconnectivity was greatly enriching for understanding and applying the data yet it was a considerable task both mentally and physically.This task requires a great deal of time and space both in the mental and physical sense.Moreover in undertaking this process one can understand why so many journal articles and book chapters present a thin sliced version of CDA.To give a full and broad picture would require publishing space very rarely offered outside of a doctoral thesis.
In bringing CDA to the SEL policy conversation this study explicitly demonstrates and builds on Fairclough's (2000) belief that that in current times policy is done in and through "discourse driven" ideas of social change.In both England and Wales, although SEL policy took different routes, my study reveals that the foundation in both countries was a limited version of social change constructed through a specifically located discursive practice.The discursive practice (here illuminated through interdiscursivity) of neoliberal, self-managing, entrepreneur in England, and social democratic, collaborative, citizen in Wales, intimately entwined language, knowledge and power, in the service of the dominant state actors.This dominant discursive practice was consistently repeated by the policy makers interviewed who in turn produced and distributed this practice through both text (SEL programs such as SEAL) and talk (media, papers, presentations, i.e., Clear Red Water).This process was in England and Wales powerful enough to get inside people's heads (George, 1999) and my findings illuminate this process and the power of governance through the identity forming, limited, language employed in the SEL policy text and process.
For my policy makers in Wales there is strong evidence that the SEL communicative event was in essence part of a wider nation building journey, a position that reflects the findings of much of the policy analysis published (Drakeford, 2007;Phillips & Harper Jones, 2002;Reynolds, 2008).In contrast to the limited land of the imagined operating in England at the time the imagined land in Wales, during the early years of the assembly, was driven by the power of devolution and built on ideological notions of a made in Wales inclusivity (Bradbury & Andrews, 2010).This ideology created a space where for my policy actor's the rhetoric and reality, for a short moment in time, were as one in a land of Clear Red Water.It was only when powerful social forces attached to the global neo liberal agenda (PISA, national standards, measurement) gained traction in the Welsh discourse and the producers of the system were no longer given the freedom to run it (Evans, 2015) that this positive, intimate, experience changed shape and become one of confusions and uncertainty.From being in a position of innovative, process, producers my Welsh policy actors found themselves having to operate in a new land where outcome rather than process mattered and the producers' voice had less power.
This study has revealed the dominance of a particular performative (Ball, 2008) version of neoliberal thinking on the English SEL imagination and the power this thinking has had in shaping the English SEAL program and reframing the Welsh SEL approach.I would contend that a key challenge for future SEL research is for the community to investigate how SEL could look different?In answer to this I propose the following tasks.The gaze of SEL professionals and practitioners needs to critically focus on UK and European based initiatives.It needs to engage with and support locally based SEL initiatives and for the imagination to be open to forms of SEL that are not built on the objective list model (Watson et al, 2012) and defined through measurement and commodification of the emotions.This study recommends that educational research explore minoritarian, fluid, understandings of SEL that are contextual, subjective and relational (Emery, 2014).
In advocating an alternative approach I call on the SEL community to build on the collaborative social justice model and children's rights strengths of the Welsh experience.In essence I encourage the SEL community to locate their understanding and application of SEL within a transformative frame (Apple, 2009) grounded and shaped by the communities they are serving.My own work is actively promoting these emancipatory aims through direct practice in English schools and further education institutions and a pan European joint working SEL project (school pupils aged 6-19] with the University of Helsinki, Lithuania (Ministry of Education), Latvia (National Centre for Education), University of Milano-Bicocca and the Hellenic Ministry of Education, Research and Religious Affairs.
Finally, I would contend that as Fairclough (2005) considers discursive practice to be the bridge between text and social practice this study is itself a discursive bridge, one produced and constructed to mediate between the SEL community (and its associated conversations) and the field of critical education policy.education policy analysis archives editorial board I really started saying you know what are we what are we doing what are we doing?These kids they had teachers in the you know and they were trying to teach them but they actually lacked the social skills as it were, they lacked the social skills because they'd never been taught these.E3

Table 1 -The interview participants
came out that er well you know and sort of showed that well you're not actually delivering on the outcomes.W1 And because we had never extended from Jane Davison's you know no league table stuff … we hadn't developed a sufficiently strong school improvement agenda… We just didn't do it and and I don't know why.W3