The Discursive Direction of Welsh Devolution : a critical analysis of Welsh anti-poverty agenda narrative

This paper contributes to the field of social policy and focuses on the terms 'independence' and 'sustainability' within key policy documents produced by the Welsh Government as part of its anti-poverty agenda in Wales. These are cross-referenced with key events in Welsh devolution from 1999 to 2013 and qualitative data from nonpolitical actors involved in the anti-poverty agenda. The aim is to highlight the wider significance of engaged, committed and systematic policy discourse critique with a view to intervention in political practice (Fairclough, Mulderrig and Wodak, 2009; Farrelly, 2010). The paper also forms a critique of assumptions around the meaning of the key terms in the context of the Welsh anti-poverty agenda. The paper begins by contextualising Welsh devolution, the anti-poverty agenda in Wales and outlining Critical Discourse Analysis (CDA) as an analytical tool. It then goes on to highlight disparity between political actions and how they are represented through discourse. Using CDA (Fairclough, 2003; Farrelly, 2010) the paper unpicks political narratives representing milestone events in the Welsh devolutionary trajectory. These events include the 1997 referendum on Welsh Devolution, the 2004 Richard Commission Report calling for increased power to the Welsh Government and the 2011 referendum on primary policy making powers for the National Assembly in Wales. Primary qualitative empirical findings, gathered as part of a PhD research project on devolution and deprivation, are used to expand on these in a three-tiered analytical approach. The paper concludes that, in practice, 'independence' and 'sustainability', as they are defined in Welsh Government anti-poverty discourses, are not mutually compatible within the initial programme structures of the Welsh anti-poverty programme, Communities First, to which their CDA practice refers (Fairclough, 2003). This finding is correlated with events punctuating constitutional change in the Welsh Government status from a largely executive to a legislative governing body.


Introduction
This paper uses Critical Discourse Analysis (CDA) to further investigate findings from data collected through 32 in-depth qualitative interviews in three Communities First case study areas, as part of a PhD research project on devolution and deprivation in Wales, completed in 2012.It argues that the subtle shift in meaning of 'independence' and 'sustainability' by the Welsh Government in the context of social justice indicates a wider change in the legislative structure, power and purpose of devolution in Wales.Namely a shift from an executive governing body seeking justification for its existence, to a legislative one established and with primary policy making powers.The way this shift has taken place over the past two decades has come under criticism for creating a less effective form of governance in Wales and has been termed: A highly flawed process of constitution building… [resulting] in government structures and processes that have repeatedly been proven inadequate for the task in hand.(Jones and Scully, 2013: 55) In the fields of anti-poverty and social justice from 1999 the Welsh Government was an executive body effectively functioning as a large local authority until March 2011 when it achieved primary policy making powers.For the purpose of clarity both the executive Welsh Government and the legislative National Assembly will be referred to as 'the Welsh Government' in this paper unless explicit reference is made to one or the other.Before 2010 the Welsh Government focused on micro-geographies and holistic approaches to governance and implementation.From 2011 onwards the focus on tackling deprivation became more strategic moving towards devolving implementation at local authority level.This could be indicative of the Welsh Government 'raising' itself up a strategic level.The change poses questions regarding the process of devolution, both from Westminster to the Welsh Government and from the Welsh Government to institutions within Wales: a process previously termed 'double devolution' (Milliband, 2005).The institutions in question have changed from local level partnerships and conglomerates to local authorities in an escalation of power and responsibility affecting multiple, complex and interlinking 'levels' of governance (Armstrong and Wells, 2005).While this is a simplification of what is a complex governing process, it seeks wider understanding of what causes the repercussions of structural change in the relationship between state and society.
In 2001 Welsh Government policy discourse advocating independent and sustainable community partnerships were core facets of community-led regeneration to tackle deprivation in Wales' poorest areas.The terms 'independence' and 'sustainability' are put in context here using CDA in the field of social policy and relating to social justice (Farrelly, 2010;Fairclough, Mulderrig and Wodak, 2009).The primary data was gathered over two years, 2008 to 2009, in Wales as part of a PhD research project on the impact of Welsh devolution on political approaches to deprivation in the country's poorest areas.The project used predominantly qualitative research methods in a case study approach to examine Welsh Government's area based, flagship regeneration programme, Communities First.Findings from this research showed the two words clearly used to mean different things by different actors representing different 'levels' of governance, making a strong case for further exploration into governing practices reflected in policy discourse.
While much has been written on the emergence (Jones and Scully, 2013), economic processes (Bristow, 2008), law (Chaney, 2002), geography (Harris and Hooper, 2006) and evaluation (Kay, 2003) of Welsh devolution; further investigation is required into the discourse surrounding key events under the anti-poverty agenda.This paper aims to contribute to debates and investigations into the past, present and a future shape of the Welsh Government's approach to poverty through discursive analysis.

Context
Since its establishment in 1999 the Welsh Government has changed from a largely executive body, to a body with both an executive and a legislative arm holding policy making powers over 20 devolved areas.Key milestones in this development began with a referendum and 'No' vote for Welsh Devolution in 1979.'No' vote campaign messages from inside and outside Callaghan's ruling Labour party were somewhat contradictory as they focused on simultaneous domination of a Welsh administration by middle-class Welsh speaking extremists and South Wales socialists.In the interim between 1979 and 1997 many interlinking factors encouraged the reemergence of devolution on the political agenda.These include: the rise of regionalism in Europe resulting in structural changes that allowed more space for nationalist movements to function (Keating, 1998); a clause on the Treaty on European Union allowing regional ministers to represent nation states and the Charter of Regional and Minority Languages in 1992; the forming of the Committee of the Regions in 1994); and the Framework for Convention on National Minorities in 1995..This context, coupled with a determination for independence in Scotland contributed to a swell of campaigning for a second referendum (Jones and Scully, 2013).Within Wales a growing dissatisfaction with a 'quango state' and the normalisation of Plaid Cymru contributed to the reassertion of policy narratives on devolution.Other significant factors include Margaret Thatcher's approach to deindustrialization and the resulting Miners' Strike in 1984 and 1985.Finally, the Shadow Secretary of State, Ron Davies, proactively pursued the devolution cause within Westminster.
When the 1997 referendum for Welsh devolution took place the socio-political climate in Wales was primed for a 'Yes' vote.This followed 20 years of Welsh Plaid Cymru and Welsh Labour Party cross-party collaboration and the influence of key political leaders and members of the Trades Union Congress.The publication of the Richard Commission Report in 2004 called for policy-making powers in Wales and the subsequent 2006 Government of Wales Act instigated a separation between the executive and legislative arms of the government.Most recently the 2011 referendum and 'Yes' vote for policy-making powers has led to the National Assembly legislation over 20 Measures, or areas of policy.During this time Wales has been ruled by a Plaid Cymru and Labour coalition government, consolidated in the policy paper One Wales (2007a) and is currently under Labour Party rule.The purpose of this summary is to pin-down a series of practices that can be reflected in discourse and better understood through discursive analysis.
Since 2001 the Welsh Government has implemented its anti-poverty agenda through the Communities First programme, aimed at tackling deprivation.The programme function has changed over the 12 years of operation to date, in tandem with the process of devolution at higher strategic level.
Taking a community-focused approach, the programme was established in 142 ward, sub-ward and Lower Super Output Areas in 2001 and changed shape and purpose considerably over the following years.In November 2012 the then Minister for Communities and Local Government, Carl Sargeant, announced 12 'cluster areas' comprising an agglomeration of smaller partnerships in Wales with just above £19 million approved for each up to March 2015.This number represents around one quarter of the full programme implying that there will be around 30 clusters in total and around 150 partnerships.All partnerships are working in the most deprived areas of Wales based on the Welsh Index of Multiple Deprivation (2012).
In 2001 Communities First partnership aims were structured according to eight 'Vision Frameworks' which covered the eight domains of deprivation.From April 2013 each partnership became part of the geographically defined 'clusters', mentioned above, and governed by an overarching partnership board (often a local authority or a partnership between local authority and a voluntary or community organisation) covering a local authority area.This clustering is indicative of a movement to widen the geographical remit of partnerships while focusing their deprivation remit from eight to three areas.Partnerships receive core funding from the Welsh Government which is administered through a grant recipient body, in most cases the local authority or a voluntary organisation and, in a small number of cases, the partnership itself.
In 2001 the programme was defined as: … the Welsh Assembly Government's flagship programme to improve "the living conditions and prospects for people" in the most disadvantaged communities across Wales.(Welsh Government, 2001) The programme is now defined by the Welsh Government as: ...a Community Focused Programme that supports the Welsh Government's Tackling Poverty agenda. (Welsh Government, 2013) Partnership structures were decided in 2001 and, in some cases, local authorities led the decision making process on recruitment of partnership Boards.In terms of partnership sustainability and independence, using existing geographical levels of governance has had significant repercussions, and was felt by some to detract from the bottom-up aspect of the programme (Adamson, 2009).
The research revealed the symbolic significance of partnership independence from local authority structures and the function of the Welsh Government in relation to Communities First as an executive rather than a legislative arm.More specifically it showed disparity between discourse at partnership, local authority and Welsh Government level.This was of particular relevance in relation to two key words 'independence' and 'sustainability'.The key terms were predominantly used by participants in reference to partnership status but with very different understandings of their respective meaning.'Independence' for Communities First partnership staff and volunteers often meant a degree of financial and political autonomy from their local authority counterparts, a status that would lead to 'sustainability' for the partnership as a self-sufficient functioning entity.Partnership 'independence' was not referred to by Welsh Government representatives, reflecting a widely significant absence of discourse; however, 'sustainability' of partnerships was taken to mean increased collaboration and interdependence between partnerships and local authorities.
One case study Communities First partnership received its funding through a local authority grant recipient body from 2001 until 2004 when it became its own grant recipient body, receiving its budget directly from the Welsh Government.The sequence which led to this change began with a series of conflicts between local authority officers and the community partnership.Discussions about these problems took place between the partnership chair, vice chair and treasurer who took the decision, based on consensus from within the partnership, to pursue for a change in grant recipient body.The partnership is now a Development Trust and company limited by guarantee.
The impact of becoming its own grant recipient body has been positive for those involved at partnership level.The conflict with the local authority ceased when the partnership became able to claim money for projects and events in a reflexive way more suited to small scale community development work.In addition after 2004 the partnership had its own trading arm in the form of a café which builds financial capacity to fund two members of staff.This has meant an increasing sense of financial 'sustainability', valuable for the partnership in light of potential funding cuts in 2013.This would not have been possible without the degree of control over funding coming from grant recipient body status gained through the partnership version of 'independence'.
...we fund four staff members of our own without Communities First money all from trading and we want that to grow and become stronger...we need sustainability because the type of work we're doing in these communities is going to take...a lot longer than the length of the programme.(Communities First Coordinator, West Wales, 2008) All involved in the research agreed that becoming a grant recipient body it is a positive step towards 'sustainability': The partnerships most likely to disappear in 2012 are the small, disparate partnerships and the ones that are council projects.(Communities First Coordinator, North Wales, 2008) Findings also highlight the symbolic value of independence for partnerships.In some cases it has allowed confidence to grow and subsequently for the partnership to feel able to challenge the local authority on issues of contention.In this way symbolism can lead to tangible benefits for partnerships.
However, partnerships' 'independence' in decision-making outside their local authority remit was questioned by one local councillor who felt that giving a lead to community members in partnership had led to unsuitable community leaders and bad decisions.He stated that more mechanisms of accountability should be put in place: The emergence of that sort of individual with that sort of authority over the partnership is a clear example of a weakness in the [CF] system.(Local Councillor, South Wales, 2008) The research findings also show that the Welsh Government did not encourage partnerships to become their own grant recipient bodies.
...partnerships becoming their own GRB is...an illusion, a mirage in the desert.(Welsh Government Civil Servant, 2009) In the context of Welsh Government rhetoric on Communities First 'independence' relates to the end of core Welsh Government funding and the potential move from state-funded to 'independent' partnerships.This means a reaction to the end of revenue funding by increasing grant funding and trading arms.Cuts in revenue funding from April 2013 posed a 'sink or swim' situation for most partnerships and some degree of financial independence could mean better chances of survival as a partnership, as could closer absorption into the local authority.
Since the Brundtland Report Our Common Future (WCED, 1987) for the World Commission on Environment and Development, use of the term sustainability has been common in mainstream policy rhetoric.However, the meaning has been less consistent.The initial Brundtland Report definition was: Meeting the needs of the present without compromising the ability of future generations to meet their own needs.(WCED, 1987: online) The term is currently most comonly used to mean either longevity, economic or otherwise, or environmental continuity.Tensions arise because the flexibility of the term and the way it has been adapted to so many different areas of policy leave uncertainty around its application.It brings with it a degree of maneuverability, which can obstruct accountability of policy makers.
The term 'sustainability' in the context of Communities First has changed its meaning from longevity and community capacity (Welsh Government, 2001) to mainstreaming, programme bending and local authority involvement (Welsh Government, 2008).
The Welsh Government most recently states that sustainability is part of a One Wales (2007a) strategy which supports the creation of: Safe, sustainable, attractive communities in which people live and work, have access to services, and enjoy good health and can play their full roles as citizens.(Welsh Government, 2007: 1) In 2007 the Welsh Government consulted on Communities First following six years of the programme.The language used in the consultation document was consistent with previous narrative surrounding Communities First from 2001; as a communityfocused programme: ...what will sustain the momentum of community regeneration are the skills, ability and confidence that local people have learned and developed through their involvement in the process hitherto.(Welsh Government, 2007: 66) In the context of this research it has been used by Welsh Government actors to describe the integration of Communities First partnerships into local authority structures.However, it is possible to argue that the amalgamation of community partnerships into local governing structures to ensure 'sustainability' is, paradoxically, a move away from sustaining local community partnership independence.

Critical Discourse Analysis
CDA is used here to constructively critique Welsh Government use of the terms 'independence' and 'sustainability' in policy rhetoric.CDA is the analytical tool used here because of its practical and engaged approach to analysis and its situation within the field of social policy.However, while it has been applied to research on neighbourhood governance structures in England (Farrelly, 2010), it has not been used to analyse constitutional change.The fast-evolving process of devolution in Wales presents an excellent opportunity to contribute to CDA debate in this way.
Termed a 'problem orientated interdisciplinary research movement' (Fairclough, Mulderrig and Wodak, 2009: 357), CDA views discourse as a form of social practice within a dialectical relationship, in this case a two-way relationship between institutions and practices.Jessop (2008) has expanded on this idea in his work on state power and uses the term 'dialectical duality' to explain the on-going interactions between institutions or structures and actors in government.
In response to a call for 'more systematic methods of 'interpretation' in political studies ', Farrelly (2010: 2) argues that the CDA approach: …should be pitched at where political theory (or a hunch which is later theorised) suggests a problem lies.(Farrelly, 2010: 4) Discourse, in this case, is the use of the terms independence and sustainability in policy documents.These can be enacted in genres through actions such as the publication of a policy document, a decision made in National Assembly Plenary or a political speech.In this sense policy discourse reflects a genre represented through the use of certain terminology.It is possible to apply the same argument to Welsh Government narrative on Communities First.Fairclough (2003) applies CDA to understand contemporary processes of social transformation such as globalisation and neoliberalism.Taking a transdisciplinary approach to unpick the extent to which certain elements of political change are discursive and others are separate from discourse but represented within it.It has been chosen as a methodological tool because it can be used to further investigate an identified narrative of interest.While tools used to analyse discourse are often applied to identify patterns of narrative, electoral discourse analysis for example.Chaney (2013) uses electoral discourse analysis to systematically research the formative function of discourse in policy surrounding Scottish and Welsh devolution since the Second World War.He does this by quantifying and categorising use of language in rhetoric under the headings: values, state structures and, identity and autonomy.He explores the salience of certain issues as they rise and fall pre and post elections in Scotland and Wales over seven decades.He concludes that the agenda-setting stage of constitutional reform needs to be incorporated into future studies on state restructuring.While this is a fruitful approach for detecting patterns in discourse, the area of investigation in this paper has already been identified using primary qualitative research.For this reason CDA has been chosen because its problem orientated approach allows a focused analysis of a pre-identified hypothesis formed through primary research.

Discussion
While the term 'sustainability' in the context of Communities First has consistently meant 'longevity' using CDA it is possible to say that the genre through which this discourse is enacted differs alongside disparities between understandings of the object of application.It is therefore possible to argue that this discursive disparity represents contradiction around the means through which a Communities First partnership achieves longevity and the instructions on doing so issued by the Welsh Government.
In 2001 the Welsh Government discourse surrounding sustainability included the terms: 'build confidence', 'raise self-esteem', 'increase incomes', 'improve wellbeing', 'encourage active citizenship' and 'ensure public services are delivered in ways which are more responsive and more locally accountable.' (2001: 1) These are all terms that can be difficult to measure and subsequently hard to deem a 'success' or 'failure'; this could be seen as pre-empting the potential difficulties surrounding a rapidly shifting political landscape.In addition, it raises questions around the reasons for a potential absence of strategy.Knowing that in 2001 the Welsh Government was an executive body implementing legislation made in Westminster, one possible explanation could be that the Welsh Government discourse reflects a lack of institutional confidence backed-up by policy making powers and coupled with the absence of majority public and political support for devolution at the time.In addition, the focus of the discourse is on small-scale and individualised goals, self-esteem, for example, is more easily measured individually than collectively.
In 2007 the Welsh Government consulted on the first six years of Communities First.The language used in the consultation document was consistent with previous narrative surrounding Communities First from 2001; describing it as a communitydriven rather than a community-focused programme: ... what will sustain the momentum of community regeneration are the skills, ability and confidence that local people have learned and developed through their involvement in the process hitherto.(Welsh Government, 2007: 66) And: The Communities First Programme is about a process of supporting communities to determine their own needs and play an active part in shaping the future of their community.It is about enabling them to develop the confidence, knowledge, skills and experience to take independent action.(Welsh Government, 2007: 65) Again there is a lack of strategy visible in this discourse, something that was highlighted in the Richard Commission Report ( 2004) three years earlier.The focus is on 'local people' taking the lead in tackling what is essentially a national issue of deprivation.There is also a lack of clarity in phrases such as 'sustain the momentum of community regeneration' which is broad and descriptive without being strategic.In the second statement the onus is placed again on 'local people' and a community-driven process led by 'them'.This discourse fits well with the idea of partnerships as 'independent' from local authorities, as at no point in the discourse before 2011 is the local authority discursively placed in a leadership position.Despite this the majority of grant recipient bodies were local authorities from 2001.
In 2011 a Communities First consultation showed a shift in use of language on partnership structures and, more specifically, the aim to incorporate partnership work into wider regeneration activity: We will ensure that Communities First complements wider regeneration activity as part of an integrated programme of investment.We will also look to strengthen links with Housing Associations who play a key role in this area.(Welsh Government, 2011: 1) This is a sharp move away from partnerships as independently functioning 'community-driven' entities.In addition the tone of this consultation is a great deal more strategic and focused, seen here by use of the phrases such as 'integrated programme' and 'strengthen links'.There is a stronger steer here from the Welsh Government coupled with a move away from micro-scale ambitions; towards discourse that can be more closely associated with a government, rather than a local authority.
Within Communities First partnerships 'independence' from local authorities may mean financial instability as funding is reduced.This could lead to a lack of 'sustainability' for partnerships.However, integration into larger institutional structures could mean a lack of partnership 'independence' and a narrower forum for nonpolitical actor discourse.It is therefore possible to argue, given the complex nature of partnerships and the changing nature of community development, that 'sustainability' is a difficult state to achieve without losing a degree of 'independence' through integration into large, more powerful and resourceful organisations.
Membership within a Communities First partnership is likely to be transient over the long term.This means that power based on individuals pursuing strategically selective objectives is lost from the partnership if the individual(s) departs or ceases to pursue the issue in question.While influence and resources from within the local authority can contribute positively to Communities First partnerships in terms of power to change things within the community, the balance between support and integration arises again in the issue of independence.It is difficult to see how partnerships can achieve independence and long term sustainability simultaneously when their decision-making is framed by local authority strategically oriented structures and strategic actors.
The anomaly in this scenario then is the Welsh Government discourse from 2001 advocating individualised and locally driven regeneration activities.There are at least two possible reasons for this gauged though applying CDA to the policy documents highlighted earlier.
Firstly, using CDA to critique discursive narrative and practice, it is possible to argue that actors driving the poverty agenda from within the Welsh Government executive body required another geographical 'layer' of governance through which to 'devolve' power and responsibility.This is reflected in Welsh Government discourse and can be explained by organisational infancy coupled with pressure to tackle what is widely recognised as Wales' long-standing, deep-rooted and serious problem with deprivation (Beatty, Fothergill and Powell, 2006).In addition the successful 'Yes' campaign leading up to devolution in 1997 relied heavily on support from existing political powers, many at local government level (Andrews, 1999) and delivery post devolution subsequently relied on local authority governing structures.The primary data shows local authorities and the Welsh Local Government Association playing a key role in assembling new structures under the Welsh Government, resulting in delivery at local level and microscales, shown in the establishment of Communities First in over 140 areas in 2001 (Adamson and Bromilley, 2008).In CDA terms the three tiered analysis shows discursive practices in the delivery of the anti-poverty agenda, enacted in the form of Community Partnerships represented in changing and contradictory policy discourse.
Secondly, and linked to the first point, the temporal aspect of the Welsh Government discourse on 'independence' and 'sustainability' reflects change in practice.As a newly established institution in 2001 Welsh Government policy discourse emphasised the 'closer interface' between state and society in Wales.The primary data shows a struggle for power among local authorities under an executive and newly established Welsh Government in 2001: ...local government members [were] saying, this is the so and so partnership and these are the people we're inviting to it and obviously... (Local Authority Officer, South Wales, 2008) Using CDA as a theory of discourse reflecting social practices, it is possible to link the form and establishment of the Welsh Government in 2001 and the primary data, referring to events in 2001, to argue that the change in meaning of 'independence' and 'sustainability' reflects the way the Welsh Government changed its strategic approach to community partnerships in Wales over time.
The most likely scenario is a mixture of both combined with other interlinking political factors; intra-party politics, economic restraints and the influence of the media, for example.

Conclusions
CDA has been used here as an analytical tool to better understand how enacted discourse and key events are reflected in policy discourse.The paper is a response to findings from a research project and gives a short example of the way meaning and better understanding of an identified problem can be gained through focus on discourse in its wider context.Discourse shaping the anti-poverty agenda in Wales reflects wider constitutional change over fourteen years in Wales from creation and establishment of a Welsh Government to the ability of the National Assembly to legislate.What is presented here is a snapshot of enacted genres of discourse in this process interrogated and questioned through CDA.In practice, 'independence' and 'sustainability', as they are defined in Welsh Government anti-poverty discourse, are not mutually compatible within the initial programme structures of the Welsh anti-poverty programme, Communities First, to which their CDA practice refers (Fairclough, 2003).
Using CDA it is possible to say that a lack of strategic direction within Communities First from the outset has led to confusion around key terms and the repercussion of