Accounting and social movements: An exploration of critical accounting praxis

Abstract A central tenet of critical accounting research maintains the need to challenge and change existing social relations; moving towards a more emancipated and equitable social order. The question of how critical accounting research upholds this principle has been intermittently discussed. This paper aims to engage with, and further, this discussion by contributing to research linking accounting information to social movements.


Introduction
". . . accounting cannot be independent of its social conditions. Under capitalism, the moving force of accounting lies in political economy -in class contradictions. Accounting is made, in part, by adjustment to the economic needs of the ruling class." Catchpowle and Cooper (1999: p. 712) "Neither the most sophisticated political debates nor the best researched expose's of corruption or political failure will bring about change. Social movements, in their various forms, are required to do that." Cooper and Coulson (2014: p. 238) A central tenet of critical accounting research maintains the need to challenge and ultimately change existing social relations; moving towards a more emancipated and equitable social order. The question of how critical accounting researchers should uphold this principle has been intermittently discussed -for example Bryer (2014) ;Cooper, Taylor, Smith, and Catchpowle (2005); Cooper and Coulson (2014); Neu, Cooper, and Everett (2001); Sikka and Willmott (1997) or Spence (2009). Neu et al. (2001) explain how critical accounting academics have intervened in different arenas such as exposing corporate abuses in accounts, engaging in social audits or enhancing accounting education. Sikka and Willmott (1997) review the way early critical accountants worked in diverse fields and with various audiences. This paper follows in that tradition by exploring the class nature of accounting information and its use by social movements. The central aim is to address what we as critical accounting researchers should do to advance a better alignment between accounting information and emancipation. Our intention is to focus on the link between the praxis of critical accounting researchers and the class nature of accounting information. In this regard the historical analyses of accounting information being mobilised in the Match Girls Strike (1888) and radical newspaper Forward (Gallhofer & Haslam, 2003 provides a catalyst for initial deliberation. Gallhofer and Haslam's (2003) main concern is to better align accounting and emancipation; one aspect of which concerns praxis. Accounting information may not easily be conceived as a natural tool for social movements; however, as Gallhofer and Haslam show historically accounting information has been mobilised by socialists, campaigners and trade unionists to illustrate the inequality and exploitation that capitalism generates. As we illustrate below, in the early 21st century social movements continue to use accounting information in their struggles against neoliberal policies. Haslam's (2003, 2006) historical analyses also encourages contemplation of the emancipatory possibilities of accounting today. They conclude that accounting information can be used by social movements because it has no necessary class belongingness; arguing that when campaigners present accounting information primarily as fact such information goes beyond the complete capture by any particular group, class or social body. We concur with many of Gallhofer and Haslam's (2003) findings. However, our departure point is the proposition that accounting information is not neutral and does have a particular 'class belongingness'. This stands in contrast to Gallhofer and Haslam's (2003: p. 7) formulation that 'accounting has no necessary class belongingness'.
By way of extending this analysis to contemporary times, the paper examines an embryonic social movement which used accounting information. The example of social movement unionism (SMU) depicts an exchange between the Vice-Chancellor (VC) and a UCU (University and College Union) branch of an English university. The purpose is to discover firstly whether accounting information is still mobilised by modern day socialists and campaigning groups. Secondly, to explore the basis on which modern social movements use accounting information in the context of Gallhofer and Haslam's claim that this is achieved by presenting accounting information primarily as fact. Thirdly, whether the claim that accounting information has no necessary "class-belongingness" can still be upheld.
To analyse these issues the paper adopts an alternative theoretical framework to that of Gallhofer and Haslam's. It considers the work of Bakhtin (1981), whose theory of language (we argue) is grounded in historical materialism (Roberts, 2004); along with a classical Marxist analysis of society (Harvey, 2010(Harvey, , 2014Rees, 1998). This approach is utilised to show how accounting information is an expression of, and intimately interwoven in, a contradiction-ridden capitalist society. These contradictions create the space for campaigners to mobilise accounting information in support of their aims; and inform our knowledge of praxis.
The paper is structured as follows: subsequent to this introduction, Section 2 briefly reviews the literature on accounting and social movements, including a discussion on the nature of accounting information and justification for a direct focus on social movement unionism (SMU). Section 3 introduces the theoretical framework used to analyse the SMU example. This framework integrates classical Marxism and dialogics. Section 4 explores and analyzes the example of accounting information and social movements, covering a trade union pay dispute. The paper concludes with further analysis and discussion of the class nature of accounting information, including an exploration of the implications for accounting praxis and agency in the struggles for an emancipated world.

Review of accounting and social movements
Radical accounting researchers have developed critiques of the nature of accounting information and the role of the profession. In the US, Abraham Brilloff has a large body of work in the form of dissecting the accounts of corporations and showing how their accountants have sought to mislead and deceive (Tinker, 2005). In the UK, Prem Sikka has written of the exploitation and racism, money laundering and bribery, price fixing and tax evasion schemes involving professional accountancy bodies and their corporate clients (Mitchell, Sikka, & Willmott, 1998;Sikka, 2008;Sikka & Willmott, 1997). Other critical accountants have analysed the discourse in annual reports to illustrate the hidden struggle between capital and labour (Neimark, 1992); while others have sought to theorise the role of accounting and accountants (Tinker, 1991); and have exposed the limitations of research streams within the accounting discipline (Chua, 1986;Tinker, 2005;Tinker & Puxty, 1995). However, limited attention has been placed directly on the way accounting information has been used by social movements in their struggle for a better world. Neu et al. (2001) reflect on two high profile interventions of accounting academics into social conflicts: the Miners' strike in the UK in the mid-1980s and the 'Debt and Deficit' debates in Alberta, Canada in 1993. Neu et al.'s (2001 conclusion focuses on making effective interventions, in the main through the mass media. Alternatively, Spence (2009) critiques the social accounting project, where to unlock the emancipatory potential of the project the reliance on corporations to act as agents of that change is misplaced. Spence establishes the limits of current social accounting practice, by placing the project within a base and superstructure framework. In the process he identifies civil society social accounting examples in the form of adbusting and culture jamming actions, as alternatives. Cooper and Coulson (2014) have reflected on the experience of a group of academics implementing Bourdieu's concept of the "collective intellectual". The reflection in part explores the relationship between academics and a social movement; in their case the campaign calling for a public inquiry into an industrial disaster in Glasgow in 2004. In contrast, Bryer (2014 analyses the manner in which accounting information developed in the "empresas recuperadas" in Argentina following the economic crisis in 2001. Bryer is concerned with how to theorise accounting information in a situation of social change; rather than the relationship between accounting information and social movements. During the 1970s and 1980s there was a small discussion on the role that accounting information can play in industrial relations (Amernic, 1988;Ogden & Bougen, 1985;Owen & Lloyd, 1985). In the main this focused on managerial disclosure of accounting information to recognised trade unions during negotiations, a different focus from that in the example below. However, Amernic (1988) does highlight one example where existing publicly available accounting information was used by the players union in a dispute with Major League Baseball owners. Ogden and Bougen (1985) briefly discuss the contradictory position trade unions can find themselves in. On one hand disclosure provides a more level basis on which negotiations can take place. On the other they identify 'the latent function of accounting information as an ideological mechanism for propagating and reinforcing managerial values and purposes' (Ogden & Bougen, 1985: p. 220). This formulation is closest to that advanced below where we argue for locating the source of the contradiction in the nature of capitalist social relations (i.e. class society).
Whilst this is far from a comprehensive exposition of the activities of all those who would classify themselves as critical accountants in its broadest sense, 1 it does illustrate the point that such a range of highly illuminating investigative societal concerns about the nature of accounting information, would still benefit from a central focus on its effective use by social movements. In recent years, this has been addressed by Gallhofer and Haslam, in their work on historical social movements of late 19th and earlier 20th century in Britain.
2.1. The nature of accounting information Gallhofer and Haslam (2003) study Herbert Hyde Champion's use of accounting information in the socialist journal, Common Sense, in the lead up to, and during, the Match Girls strike 2 of 1888. The role of "facts and figures" were prominent in Common Sense from the beginning, as Champion sought to expose and publicise the exploitation prevalent among the women workers by using conventional accounting disclosures made at shareholders' meetings.
In analysing the Match Girls strike and other works, 3 Haslam (2003, 2006) establish three conclusions. Firstly, there is a history of accounting information being mobilised by trade unions and social movements in the struggle for emancipation. It was used by counter-hegemonic forces such as the radical publication Forward, and still has a role as an instrument for counter-hegemonic intervention of current social movements. Secondly, campaigners (despite, at times, voicing distrust of accounting practices) overwhelmingly present (and appear to accept) accounting information as fact: 'The above are FACTS' (emphasis in the original; from Forward June 1915, quoted in Gallhofer & Haslam, 2006: p. 239). Conventional media of the time, such as The Economist, may have utilised the accounts in a discourse supportive of capitalism but radical newspapers mobilised the same 'accounting facts consistent with a labour orientated, socialistic and counter-hegemonic intervention' (Gallhofer & Haslam, 2006: p. 237). Haslam's (2003, 2006) final conclusion (which flows from their second) is that accounting information has no necessary class-belongingness -although they in no way ignore the importance of class. They draw from Stuart Hall's work on the mass media which holds that meaning depends on how things are signified not their real existence; enabling similar events to be signified in different ways (Gallhofer & Haslam, 2006: p. 248). To support this conclusion they cite the work of Lehman and Tinker (1987) and Tinker and Neimark (1988). Accounting information in this respect has no 'necessary' class belongingness; if it can be supportive of hegemonic capitalistic forces, it can also be mobilised on behalf of the counter hegemonic forces. It can take on different meanings and significance (Gallhofer & Haslam, 2006: p. 248).
However, such analysis could be said to soften the material roots of the class nature of accounting information, instead seeking to emphasise its socially-constructed nature. In contrast, Catchpowle and Cooper (1999), while charting the journey of the African National Congress (ANC) in South Africa from outright opposition to embracing privatisation as a policy, rely on the work of Volosinov (1973). They show that the discourse around the ANC's change in attitudes had real material roots based in the class nature of South African capitalism. This paper draws from the same theoretical pool of ideas as Volosinov, to firmly root the nature of language (and by extension accounting information) as an expression of the class structures of capitalist society. This approach differs from Haslam (2003, 2006) and Lehman and Tinker (1987), who choose not to give precedence to the material roots central to the class structures of capitalism. As with Catchpowle and Cooper (1999), we do not intend to demean the importance of semiological or cultural aspects in the critical accounting literature, but would rather focus on the importance of the dialectical relationship between the economic base (capitalist) and the superstructure (culture, language, ideology and so on), and how the interrelationship between the two can to some extent be elucidated through critical accounting praxis. The paper now turns to an alternative theoretical framing that enables a different alignment of historic and contemporary examples.

Theoretical framework
The objective of this paper is to explore the theoretical implications of social movements using accounting information in their struggle for a more emancipated world, specifically addressing accounting research and praxis. This paper adopts a theoretical framework grounded in the classical Marxist analysis of society (Harvey, 2010(Harvey, , 2014Rees, 1998), whilst extending this understanding by integrating a dialogical approach (Bakhtin, 1981;Barker, 2006;Joseph & Roberts, 2004;Roberts, 2004;Sullivan, 2012;Vice, 1997). A similar framework was utilised by Smyth (2012) to analyse the competing claims of public accountability during the privatisation of public housing in England.

Marxist understanding of society
Essential to an understanding of classical Marxism is comprehension of its philosophical tool, historical materialism (Callinicos, 2006;Harvey, 2010;Rees, 1998); where, according to Bhaskar historical materialism, is the research programme founded by Marx (Bhaskar & Callinicos, 2003: p. 101). There are three key elements to historical materialism -a materialist (realist) ontology; the utilisation of the Marxist dialectic; the combination of the prior elements leading to an analysis of class society including an understanding of the base/superstructure dichotomy.
Central to historical materialism is the dialectic (Cooper et al., 2005;Rees, 1998). Dialectical thought in its modern form emerges from German philosophy following the Enlightenment. While Hegel had developed a dialectic that was based on idealism, it was Marx and Engels who took Hegel's ideas and applied them in a materialist manner, by turning the dialectic the right side up and making humanity the centre of the historical process. As part of this process, Marx and Engels re-worked Hegel's categories of the dialectic, as Rees (1998: p. 7) summarises: 'this then is the general form of the dialectic: it is an internally contradictory totality in a constant process of change . . . [where] the parts and the whole mutually condition or mediate each other'; further, in the classical Marxist tradition these categories are '. . . not simple intellectual tools but real material processes and so this is a materialist dialectic' (Rees, 1998: p. 8).
By applying historical materialism to an analysis of capitalism Marx was able to develop a critique of political economy, at the heart of which are the tensions and contradictions in the commodity form of production. This includes the development of the forces and relations of production, accompanied by the rise of class society and the ensuing dynamic of class struggle; these in turn lead to the struggle between capital and labour, and hence the campaigns and struggles of trade unionists and social movements.
Historical materialism starts from real individuals and groups, it analyzes their interaction based on the need to produce and reproduce human life. As the productive forces develop so does class society, the state, politics etc.; however, the form these take is linked to the economic base of society. In present society, this analysis is based on recognising that capitalism is a whole (totality) system, where all aspects of human action and behaviour are imbued with its priorities and prejudices. But it is a totality that is riven with contradictions -for example between exchange and use values (Harvey, 2014). These contradictions provide the basis ultimately for either the transcendence of capitalism to a new socialist society or the common ruination of all classes. To avoid the trap of 'crude' economic determinist theorising, it is important to stress that the economic base of society cannot be independent of its social conditions (or vice versa). In turn, we argue, along with Bakhtin (1981), that language is also an emergent expression of the social conditions in which it operates. The next section explores the various formulations of dialogical theory that have been applied to accounting information.

Bakhtinian dialogics
There has been intermittent use of dialogics as a basis to understand the nature of accounting information. The version of dialogics applied here is based on Bakhtin's (1981) original work in contrast to other approaches. For example, Bebbington, Brown, Frame, and Thomson (2007) draw on Freire's (1972) work to find a progressive engagement for social and environmental accounting, where'[O]nce both groups uncover the common ground between them, then dialogic engagements can begin between these previously antagonist groups ' (2007: p. 364). From a post-structuralist perspective, Macintosh, relies heavily on Bahktin's formulation of heteroglossia in his work (Macintosh, 2002).
More recently dialogics have been coupled with pluralism and agonistic democratic ideas (Brown, 2009) in an attempt to develop alternative accounting information systems (Blackburn, Brown, Dillard, & Hooper, 2014;Dillard & Yuthas, 2013). We agree with Brown (2009: p. 320) when she states that 'taking pluralism seriously requires discarding the idea of a Habermasian or Rawlsian rational consensus'. However, she appears to believe that democratising accounting technologies can take place without democratising the economic base on which social relations are built. Further, the focus of Brown's (2009), Dillard and Yuthas' (2013) work concerns the development of future accounting information system; whereas, we are concerned with the nature of existing accounting information and how it can be mobilised by social movements.
In contrast to the versions of dialogics outlined above our integrating of Bakhtin's dialogical approach with historical materialism holds a more antagonistic and dialectical stance, where language continues to be fought over (Barker, 2006). There is an unresolved debate about the authorship of the works produced collectively by the Bakhtin Circle 4 (Holborow, 1999;Sullivan, 2012;Vice, 1997). The debate concerns the disputed authorship of certain texts, 5 mainly between Bakhtin on the one side, and Volosinov and Medvedev on the other. There is some debate as to whether Bakhtin and Volosinov was the same person (Holborow, 1999). It is most likely that they were two individuals, both part of intellectual circles in Nevel, Vitebsk and Leningrad between 1924 and 1929. Volosinov "disappeared" sometime in 1934 during Stalin's purges and Bakhtin never officially denied or accepted that he wrote the texts ascribed to Volosinov (Holborow, 1999). However, in the introduction to Bakhtin's (1981) The Dialogic Imagination: Four Essays, the editor Michael Holquist states his opinion that over 90% of the three disputed texts were written by Bakhtin, which includes the avowedly Marxist oriented, Marxism and the Philosophy of Language.
Vice sets out the implications of this authorship debate: 'opinion on the authorship debate generally line up with particular views on Bakhtin's Marxist leanings' (Vice, 1997: p. 8). For Vice (1997) there are three positions; either an open-mind should be kept, or Bakhtin's Marxism is a case of window-dressing, or finally, it is a genuine element. While space limits a detailed discussion of each position, it should be noted Vice argues that Bakhtin's dialogics are compatible with Marxism; a point also endorsed by Sullivan (2012). In contrast to mono-logical attempts at resolving the authorship debate, Roberts (2004) applies dialogical ideas to the texts of Bakhtin,Volosinov and Medvedev. 6 He identifies elements of weak social constructionism, critical realism and Marxism in the texts, before stating that 'Marxist materialism, encompassing a form of realism, is a more satisfactory way of exploring the discourse theory developed by the Bakhtin Circle' (Roberts, 2004: p. 90). Following in the spirit of Roberts' work, rather than trying to definitively classify Bakhtin as a Marxist or resolve the authorship debate, we illustrate the ways in which Bakhtin's model of language can be understood using Marxist dialectical thinking; in particular the categories of totality, change, contradiction and mediation (Rees, 1998). Bakhtin's (1981) model has three key components. First, that language is stratified by real concrete social forces. For example, Bakhtin (1981) argues language is not neutral; rather it is tension-filled and stratified, interacting with the concrete socio-historical environment (heteroglossia 7 ). Here the emphasis on contradiction, opposing social forces and change over differing time periods comes to the fore. Second, an episode of language (an utterance 8 ) enters into a heteroglossia (totality) so that it does not relate to its object in a simple single manner but interacts (mediation) with a changing environment of alien words and themes about the same object.
Third, the foregoing illustrates the internal contradictions, the centrifugal (separating) and centripetal (unifying) forces, at work in any utterance. Centripetal forces seek to unify a single meaning for the utterance, to make it monologic; whereas, the centrifugal forces counter the common perceptions and multiple meanings based on dialogised interactions (Bakhtin, 1981: p. 272). Dialogic theory treats those involved in discourse not as passive receptors but active participants; 'they are beings 'full of words', with all their own experiences and 'apperceptive background' encoded in inner speech, critically appraising what they hear' or read (Barker, 2006: p. 46). Each utterance does not stand in isolation but is part of a chain of dialogised exchanges where replies are made in response to previous utterances and participants are constructing their own responses as each new utterance is generated (Barker, 2006).
Roberts concludes that the 'materialist accents in the Bakhtin Circle suggest that discursive method must isolate a "contradictory unit" of analysis that contains an interconnected contradictory "developed whole" within itself' (Roberts, 2004: p. 103). The point of the Bakhtin Circle's work is to provide the link between those competing conceptions (of accounting information) and real material (socio-economic and historical) circumstances (for example in a pay dispute) that the utterances are formed in and interact with.

Summary
As this section has introduced a number of complex ideas and concepts, not all of which are well known in the accounting literature, it is worth pausing to summarise the overall theoretical framework utilised in this paper. The framework identifies language as a stratified, contradictory social practice that has real material roots in the social world. Utilising a classical Marxist analysis, this social reality is characterised as capitalist with the commodity form of production and related exchange dominating, in turn giving rise to class struggle between capital and labour. Bringing these approaches together the next section sets out a contemporary example of how accounting information can be mobilised by social movements. The example provides the lead into discussing Haslam's (2003, 2006) work and the conclusions they draw.

Illustrative example -contemporary social movement
A burgeoning social movement's use of accounting information is examined below. The example is reminiscent of Cooper et al.'s (2005: p. 951) formulation that 'the production of something akin to early social audits aligned to contemporary social struggles and action groups (e.g. trade unions) would promote the potential to create a more equitable society'.

Background to social movements
Before proceeding directly to the illustrative example it is important to define what we mean by social movements. In this sub-section we outline the different generations of social movements, before looking at the recent development of social movement unionism. There are several definitions of social movements with the common characteristics of coalescing individuals and groups into a collective that seeks to challenge extant power and authority whether political or cultural (Diani & Bison, 2004;Snow, Soule, & Kriesi, 2004;Tarrow, 1994). This leads Tilly to define social movements as: a series of interactions between power holders and persons successfully claiming to speak on behalf of a constituency but lacking formal representation, in the course of which those persons make publicly visible demands for changes in the distribution or exercise of power, and back those demands with public demonstrations of support. Tilly (1984: p. 306) A theme that permeates all types of social movements is the centrality of political opportunity. Their existence promotes the emergence and development of such movements which constitute a potential rival to the political representation system and may restructure the relationship between the state and civil society (Jenkins & Klandermans, 1995). Therefore, a good way to understand social movements is an 'agent of change', that can become a force for new political identities, creating new ideas and allegiances, with the potential to transform the political system (Heberle, 1951).
However, we might better understand social movements in terms of their relationship to capital accumulation (Bebbington et al., 2008). Harvey identifies two distinct types of "accumulation by exploitation" and "accumulation by dispossession" (Harvey, 2003). Social movements addressing the former resemble the "prototypical" type that attracted the interest of Gallhofer and Haslam. This workplace-centred form of accumulation -perhaps best described as 'traditional'historically created labour movements, including trade unions and related political parties. However, Aminzade (1995) adds a further political dimension to Harvey's categorisation, when he argues these traditional social movements entailed using social protest as an alternative to electoral action; emerging as the historical results of nineteenth century struggles over political representation.
To summarise, the growth of early (traditional) social movements can be defined as being connected to both broad economic and political changes including parliamentarization, market capitalization and proletarianization (Tilly, 1984). The labour and socialist movements of the late nineteenth century are seen as prototypical social movements leading to the formation of communist and social democratic parties, and trade union organisations.
These traditional social movements are in direct contrast to the 'New Social Movements' (NSM) of the post-war periodfor example, women's, gay and civil rights movements; peace, anti-nuclear and environmental movements -which provided the basis for Green parties and new left organisations. These NSM would best be described as belonging to the second group of resistance to accumulation by dispossession (Harvey, 2003). These new social movements were usually formed as a result of protest around issues such as "minority rights" (Hickey & Bracking, 2005). Habermas (1987) argues that social movements materialise when people's "lifeworlds" -their spheres of common, meaningful habitation -are "colonized" by forces which threaten their worlds and people's ability to control them. When confronted by this colonisation he suggests that social movements emerge to defend and win back their lives and social organisations (Bebbington et al., 2008). Many of the NSM groups emerged as a result of dissatisfaction with traditional organised labour movements, whose inability to transform society was seen as a direct result of their increasing bureaucracy and remoteness from the mass of people (Frolich, 1972). Gallhofer and Haslam (2003) draw heavily on the work of Laclau and Mouffe (2001), who in turn base their analysis on this form of NSM. Touraine (2007) in particular, expresses disappointment in the inability of these post-1968 NSMs to transform into major social movements. He observed that many rapidly became exhausted and served only to extinguish the hopes of its supporters. More especially, a number of leading left-wing/Marxists and philosophers, such as Foucault, became disenchanted by their involvement in such activities.
However, it is a third type of social movements that leads to the contemporary example in this paper. In recent times, models of post-war NSMs have been superseded by the emergence of social movements of the 1990s, more commonly known as the anti-globalization or anti-capitalist movements. These social movements usually stand outside of conventional political channels but commonly enjoy a level of public support owing to their petitioning for change (Cohen & Rai, 2000). Whilst the 1990s social movements might share a broad concern about deprivation in a general sense, and could be described as a type of resistance to accumulation by dispossession, the strategies and policies they employ are typically very different, and can range from complete rejection of the capitalist system through to demands for greater participation in decisionmaking, or calls for a fairer division of the economic benefits derived from the exploitation. What is more, the diverse approaches (tactics of negotiation, confrontation and direct action) serve to produce very divergent forms (Bebbington et al., 2008). Hirst and collaborators (Cutler, Hindess, Hirst, & Hussein, 1977;Hindess & Hirst, 1975;Hirst, 1988Hirst, , 1993 appear to concur with such an analysis, by identifying the emergence of a new type of politics, one very different from the traditional left-right opposition. More crucially they rejected the premise that social movements should be based on the material interests of any one group, namely the working class (Cutler et al., 1977). In their critique of Marxism and its 'economistic' strand, 9 they turned towards the notion of 'pluralism' within modern capitalist societies with democratic political systems, gelling around a plurality of collective actors. Each of these actors has their own unique constitutive procedures, decision-making and execution. This leads to the argument that any effective future socialist politics would necessitate the tolerance of 'multiple economic and social loci of contemporary capitalist democracies', and provides a useful insight into the characteristics of modern 1990s social movements (Edwards and Knight, 2008). Touraine (2007: p. 26) also describes the 1990s social movements (whom he welcomes as a fresh alternative to the post-1968 NSM) as having a diverse range of demands gathered around the general theme of anti-globalization and converging around a desire to challenge the leaders of the global economy. Although often disparate these movements look to a very different type of world; one that refuses to tolerate the exploitation of local interests, minorities and the environment by the wealthy and powerful. Notably though he sees their demands as new and different from those previous movements; he argues that while their membership are diverse they all demand recognition of 'new' types of rights, that he describes as cultural rights.
Yet 1990s social movements are not entirely to be found outside the workplace. The 'new' anti-capitalist activists also include a resurgence of 'new' trade union organising models, known as social movement unionism (SMU). The organising strategy for SMU involves a number of innovative approaches that go beyond traditional trade union labour practices, commonly linking or developing partnerships with organisations beyond the workplace (Upchurch & Mathers, 2012). Essentially, the union looks to, and forms links with, the demands of other new movements -neighbourhood groups, women's organisation, environmental movements -often engaging in the political arena during periods of flux, and linked to resistance against capitalism (Moody, 1997). SMU often incorporates 'activist mobilization-based unionism', which contrasts to traditional unionism because of its emphasis on social, as well as economic change (Turner & Hurd, 2001;Waterman, 1993). It combines innovative organising, traditional collective bargaining and involvement in local politics.
A good example of SMU is the role played by the trade union movement in challenging the apartheid regime in South Africa (Moody, 1997). Although the strikes did not result in the total transformation of South African society they played a decisive role in raising the class consciousness of black workers and giving them the power and confidence to contribute to the abolition of apartheid (Moody, 1997). As a result, in post-apartheid South Africa the trade unions have acquired a central role in civil society (Baskin, 1996). Whilst this success has allowed the South African labour movement to be defined as the prototypical model for social movement unionism, many others exist on a smaller scale in the rest of the world; not least the unexpected resurgence of the labour movement in the US in the 1990s. These union campaigns covered recognition rights and living wage reforms (Hauptmeier & Turner, 2007;Luce, 2006). For example, several thousand janitors (mostly Latino) joined the union-led Justice for Janitors campaign, aimed at organising mass demonstrations in support of union recognition. These activities mobilised the Latino community and led to a coalition-based social unionism which achieved bargaining and public policy victories. Although ultimately the goal to increase union density was not achieved, the political and social influences were still significant (Hauptmeier & Turner, 2007). In the UK, union activists have often focused on national strategies to the detriment of local initiatives. Nonetheless, unions like the T&GWU, UNISON, UNITE, NUT, and UCU have still managed to build coalitions with other social actors and community groups around issues like the Living Wage, education and the NHS (Holgate & Wills, 2007).
However, there is a clear distinction between social movement unionism and social movements. Social movements are broader and occur at particular historical moments. In contrast SMU may adopt social movement strategies -coalition building, grassroots mobilisation, aggressive organising, demonstrations -but they are not social movements in the fullest sense (Turner & Hurd, 2001). More specifically, SMU becomes more difficult to organise when social movements are absent, their widespread presence means they can incorporate the unions in their campaigns (Turner, 2007). All of which may go some way to explaining why, in the absence of widespread social movements, the current influence of SMU has not been far reaching.
Our example below holds some of the characteristics of 1990s social movement unions (SMU), albeit in embryonic form. UCU has an active and thriving rank and file organisation which, in common with Turner's (2007) view, combines innovative organising with collective bargaining and an involvement in political and cultural rights campaigns. The role of a significant number of UCU Left supporters in the example is explored further below.
In addition, our example has been chosen not only as an example of SMU but, equally (and perhaps more significantly for this paper), because it is an example of a 'present day' working class organisation, albeit different but still continuing to engage in political and economic resistance. This distinction is important because, in contrast to the post-industrial writers definition of the working class as 'industrial workers' (Bell, 1999;Hardt & Negri, 2001), we see the definition of class as organisational rather than structural (Eagleton, 2011). Post-industrial theorists argue that the structural changes to the economy -the shift to services, the decline of traditional industries, and the rise of an information economy -has heralded the demise of the working-class as an agent of change. In contrast, we argue for a broadening of the narrow 'industrial worker' definition to include all groups of wage workers. One of the strengths of Marx and Engels' analysis of class is their understanding that its dynamic composition is linked to the constantly changing nature of capitalism (Eagleton, 2011;Harvey, 2010). 'Class changes its composition all the time. But this does not mean that it vanishes without trace' (Eagleton, 2011: p. 162).
The centrality of the working class does not rest on the type of work performed or whether workers initially see themselves as part of a united, self-conscious class. As Eagleton points out: 'Marxism does not define class in terms of style, status, income, accent, occupation or whether you have ducks or Degas on the wall' (2011: p. 160). Class is defined by the relationship to capital and the nature of exploitation. This means the modern working class changes to include many different groups, such as white collar and service sector jobs like health workers, teachers and university lecturers. Understanding the working class in this way places workers in a unique position in society and pushes them to resist -initially in defence of their own social conditions (e.g. education or wage levels) but ultimately towards a process that can lead to a more emancipated state. Nevertheless, it is not our intention to present this example as a unique example of this dynamic working class. The example remains small and in no way fully representative, but as a newly emerging group it does enjoy some relevant, common characteristics with anti-capitalist social movements and goes some way to illustrating the complex, yet central role of the changing working class in the 21st century. It is to this example we now turn.

Example -social movement unionism
This contemporary example highlights the actions of a University and College Union (UCU) branch at a university in the north of England -Northern University. The branch and national UCU hold some characteristics that resemble SMU (e.g. campaigning and mobilising around political issues and cultural rights). In this case the local UCU branch used accounting information to counter-act the arguments presented by their VC during a pay dispute.

Background to the dispute
The higher education sector in England, as much as any other public service, has been subject to new public management reforms over recent decades (Parker, 2011). Universities have become more commercialised and marketized, with pay levels for senior managers increasing dramatically. The ten highest paid vice-chancellor's in 2013 received benefits packages of over £360,000 per annum (Grove, 2013). The main justification for this level of salary is that universities operate in a competitive market and need to attract the best quality individuals to run complex, international, multi-million pound organisations (Gardner, 2014).
In contrast, in the autumn of 2008, as the UK economy descended into recession, university lecturers received the final instalment of a three-year pay deal. The pay award had been the result of long and protracted industrial action (including strike action) which had concluded in 2006. The unions (at that stage AUT and NATFHE 10 ) argued that lecturers' pay had fallen by 40% behind equivalent professions over the previous 20 years; whilst their productivity had increased by 150% as student-to-staff ratios increased dramatically. The union demanded a 23% rise over three years, which they claimed was affordable given the £3.5 billion coming into the sector from top-up fees. 11 The membership of both unions recorded 50% turnouts in their ballots for industrial action, with more than 80% backing an assessment boycott (Baty & Thomson, 2006). When the pay deal was finally agreed in 2006, the final year of the three-year pay award (2008) was the greater of 2.5% or RPI. Initially this settlement was viewed in an unfavourable light by many union members, with some local branches calling for a rejection (Carter, 2008). However, September 2008 saw the peak of the inflation cycle and resulted in a 5% pay rise, at a time when other public sector workers were suffering a pay limit of 2.5% or less. The increase meant that pay for higher education lecturers had risen by over 15% from July 2006 (UCU News, 2008c).

UCU and development of social movement unionism
Although the newly merged UCU was clearly a traditional union its actions went beyond many conventional union practices by adopting a variety of innovative methods more akin to the practices of established social movement unionism (Bronfenbrenner & Juravich, 1998). Upchurch and Mathers (2012) state that anti-corporate campaigning, union-community coalitions and alliances, and internationalism can all be used by unions trying to re-establish their societal efficacy, based on a social movement orientation. UCU demonstrates some of these characteristics. For example the discussions at the UCU conference in 2008 were wide-ranging with the General Secretary, Sally Hunt, urging delegates to help shape an alternative vision of education, whilst also addressing international responsibilities in debates on Palestine. 12 The following year, the 10 Natfhe and AUT amalgamated in June 2006 to form the University Colleges Union (UCU). Sally Hunt, the previous General Secretary of the AUT, was elected its General Secretary. 11 In 2004 the New Labour government introduced 'top-up' fees to attend university in England capped to a maximum of £3000 per annum per student. 12 For example Sally Hunt stated to conference in 2008: "The landscape in which we work is changing dramatically. Further major reforms of further and higher education to make them increasingly demand led, employer dominated and market driven are underway. The world will not wait for us to get our act together. . .. Our campaigning work around challenging the market is a defining moment in that defence. We have to put forward an alternate vision of education and to impress our values on the minds of the public. UCU has a unique responsibility to speak out against the oppression of educators. Freedom of thought and the freedom to learn are rights that are at the heart of democratic civil society. As I have noted before, it is no surprise then that those who wish to oppress their citizens often single educators out for the harshest treatment. . ." (UCU News, 2008a). annual conference was addressed by Amjad Barham, president of the Palestinian Federation of Unions of University Professors and Employees. Further, in early 2008 UCU organised a conference to coincide with the annual LGBT history month, at Northern University. The UCU branch at Northern University had been part of broader social movement unionism, when in 2008 they successfully campaigned, to prevent the privatisation of international student recruitment and education functions.
The Northern University UCU branch had a number of officers who were active members of UCU Left. This internal union organisation was set up in 2006, and grew out of members' dissatisfaction with the union leadership's settlement of the strike of that year. UCU Left consists of activists operating openly and widely within the union. Its publicly stated aim is to 'build a democratic, accountable campaigning union' which can mobilise and involve 'members in defending and improving pay and conditions whilst defending progressive principles of education' (UCU Left, 2014). UCU Left has a broad manifesto which involves not only fighting for better education funding but also the defense of academic freedom and civil liberties in a neo-liberal economy. This organisation stands for 'the principle of free universal education under democratic control of education workers, local communities and learners at all levels' (UCU Left, 2014). The actions of the UCU branch were informed by the perspectives of UCU Left including the local campaign against privatisation and the decision to challenge Northern University's VC over the final year of the pay deal.
The increasing commercialisation of the higher education sector, a general public sector pay limit and the "unexpected" RPI-based 5% pay-rise led to calls from a number of VCs not to honour the final year pay rise in 2008. Such a move threatened not only the final year of the pay award but also to seriously undermine national collective bargaining. This led to warnings from UCU nationally that if VCs failed to honour the third year of the national pay deal they would face the full force of the national union (UCU News, 2008b). As Sally Hunt (General Secretary) argued: members fought hard to secure their pay rises, they deserve those rises and they will get them. This is not a secret deal, universities have had two years to prepare for this, and I would be concerned about the financial acumen of any institution that has failed to budget for it.
UCU News (2008d) The national dispute was played out in microcosm at Northern University, where the VC argued that such a pay award was untenable. In October 2008 he used his column in the university magazine to make a series of points concerning staff costs, the 5% pay-rise and the overall university finances. He claimed: The sector also faces the combined effect of the increase in salary costs, the costs of maintaining a final salary pension scheme, the full impact of the framework agreement and the escalating costs of energy. For [Northern University] the budget moves from a surplus of £7.1 million in 2007/08 to a projected deficit of £3.5 million in 2008/09. This is largely driven by a one year increase of salary costs of £15 million. This level of increase, partly caused by the RPI element of the pay award, cannot be sustained.
In response the recognised trade unions at Northern University also sought to use conventional accounting information as a means of challenging the VC's arguments. They questioned the £15 million increase in salary costs at the next university negotiating committee (UNC) in November 2008, and asked for fuller analysis of the figure. The unions' suspicion was that much of the increase was due to a rapid and large expansion of senior management posts in central university functions.
For example, in 2002/03 the financial statements reported 12 members of staff earning over £70,000 p.a.; by 2005/06 this had increased to 30 members of staff.
Despite the request being made in accordance with the union recognition agreement, no was provided. Instead, the VC wrote to each member of staff at their home addresses and again restated the £15 million increase in staff costs, claiming had: placed a severe strain on our recurrent resources. Our recurrent budget moves from a strong surplus of £7 m in 2007/08 to a projected deficit of more than £3 m in 2008/09.

(Letter to all staff, November 2008)
In repeating the same accounting numbers through another communication channel the VC attempted to present a monological explanation of the university's finances. However, shortly afterwards the university published its annual financial statements for 2007/08 reporting a surplus of £25 million, which equated to 10% of total income. While it is to be expected that the annual surplus from the financial statements would not match exactly the "recurrent budget" figure, the enormous difference raised serious questions among the UCU branch leadership regarding the reality of the supposed deficit resulting from the pay rise. Historically, Northern University made an average annual surplus of £2.5 million. It is also relevant to note that in the five years after the 2008 pay rise the university continued to make significant surpluses 13 ranging from a low of £15 million (6.4% of income) to a high of £35 million (14.1% of income).
In these circumstances, the UCU branch committee attempted to dialogise the VC's claims by issuing a statement on the financial position of Northern University. In doing so the branch was utilising accounting information for its own purposes, replicating the actions of campaigners highlighted by Gallhofer and Haslam; and endorsing their proposition that the use of  1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

year -Staff C osts percen tage
Source: UCU leaflet (2009). accounting information was not just a historical phenomenon but a tool that could usefully be employed by contemporary social movements. Expressing the centrifugal forces in the accounting information, the UCU branch's statement largely reported the numbers direct from the annual accounts but challenged the veracity of the numbers quoted by the VC. In addition, one small calculation was made showing the staff costs as a percentage of total income, the results of which showed a decline in 2007/08 to 54%, as opposed to 58% in the previous two years.
Confronted by a dialogical account from UCU, the VC condemned the union's analysis as '. . . both misleading and confusing' 14 (from VC statement January, 2009). In a statement to all staff, the VC explained that the reported surplus needed to be seen in the context of capital funding and also that UCU's calculation of staff costs percentage needed to be challenged. Both these points were supported by a three-page explanation that made a series of technical adjustments to the reported numbers in the accounts. The outcome of which, the VC claimed, showed the "real" surplus was £4.7 m (or 2% of total income) and that the staff costs percentage has actually remained at the same level, 58%. Here the VC is expressing the centripetal forces arising from his social position with its antagonistic relationship with organised labour in the university.
The UCU branch committee decided to respond again to the VC's January statement, employing the tactic of presenting their analysis as neutral and factual within the limitations of existing accounting practice. Again clear comparisons can be made with Gallhofer and Haslam's research; the UCU leadership, in a manner akin to historical social movements, were presenting accounting information as factual evidence. Further the UCU branch relied upon the interaction of their utterance (i.e. the accounting information) and the heteroglossia surrounding the pay dispute and industrial relations in the sector. For example a leaflet was produced with the heading 'NO MANIPULATION FROM UCU'. The leaflet carefully developed a refutation of each accounting adjustment made in the VC's statement, but also presented the branch's own analysis of the state of industrial relations over pay and national bargaining; a process by which they were attempting to utilise the heteroglossia of the pay dispute to address and theorise the exchange between the VC and UCU branch, whilst relying on evidence from functional accounting information. Conscious that this exchange was part of a process of explaining the financial position of the university to members and wary of losing those members in a back and forth trading of technical terms and adjustments (Neu et al., 2001), this refutation also included a graphical representation of the long-term decline in the staff costs percentage of total income (Table 1).
Further, illustrating the contradictory, tension-filled nature of utterances, the UCU branch used the VC's own report to the university's governors' meeting in December 2008: We did not adjust, amend, manipulate or massage the figures. Indeed the auditors signed off the accounts with a £25 m surplus and From the Boardroom 15 exclaims, the "accounts showed a successful year with a surplus of £25.3 m . . . which represented 10% of income".
(UCU leaflet, February 2009) The UCU Branch also emphasised that it was the VC who was manipulating the numbers because he was using the same information to provide very different results to different audiences, dependent on what he wanted to portray: The VC can't have it both ways -a huge surplus reported to the Governors and a small one to staff.
(UCU leaflet, February 2009) 14 It is noteworthy that the intervention by critical accounting academics in Accountancy magazine during the Miners' strike in Britain was also charged as containing '. . .major misunderstandings and inaccuracies . . . ' (quoted in Neu et al., 2001: p. 752) by a member of the ICAEW who also sat on the board of the NCB. 15 From the Boardroom is the report of each governors' meeting.
However, such a statement potentially poses campaigners with the possibility of undermining their own arguments. Thus, it may be possible, as the journalists in Forward did (Gallhofer & Haslam, 2006) at times, to highlight questionable accounting techniques used to hide large profits, but in doing so campaigners run the risk that not only are questions raised about the monological interpretation but also any dialogical interpretation advanced. However, this example does not allow for an exploration of this aspect as by February, 2009 both parties (the VC and the UCU branch) moved on to other issues. From the branch's perspective this was appropriate as the final year of the pay deal had been paid and national bargaining had been defended.
4.2.2.1. A dialogical analysis. Accounting information, as with language, is often based on an idealised world without social context -or at least, that context is often uncontested (Holborow, 1999). All too often, as Macintosh and Baker (2002) contend (and illustrated above) such information is presented in monological forms allowing only one voice with a single interpretation to dominate. But as Bakhtin argues by simply concentrating on words (facts) outside of their context, the fundamental capacity of those facts to have meaning is lost. In this example we can observe an ideological battle between groups (management and unions) to assert their version of events. However, what is developing is a dialogic interpretation of the accounting information with many different voices (union officials, its members, and management), communicating, and challenging the single interpretation (Macintosh & Baker, 2002).
Moreover, as the paper has already stressed the aim is to focus on the importance of the dialectical relationship between the economic base (capitalism) and the superstructure (culture, language, ideology and so on), and how the relationship between the two can be made clear through language. Bakhtin's theorising provides the link between those competing conceptions (of accounting information) and real material (socio-economic and historical) circumstances. These interpretations are not just different, neutral communications they represent a reflection of class society where language is an arena of class struggle which can result in a more informed consciousness as a result of the dialogue and 'real' material experience of events (Catchpowle & Cooper, 1999).
Therefore, such language cannot be divorced from its material reality -namely, the UCU members should not expect future pay rises if the management's interpretation is accepted. It is this reality, alongside the potential to raise consciousness and possibly industrial action, which makes the union's language different. This is not to argue that economics determines everything, if that was the case there would be little point in mounting resistance, but it does mean recognising that language is stratified resulting in class-based meaning. Moreover, it is the shared experience of those union members that alters the sort of language they use and share -much of which is totally incomprehensible to vice-chancellors (hence the charge that UCU's analysis is "confused" and "misleading"). Therefore whilst (accounting) facts can be used, they are constrained, mediated and contingent on a class basis. Furthermore, such a perspective complements a Marxist analysis of society, an approach that views capitalism as a whole (totality) system, where all aspects of human action and behaviour are imbued with its priorities and prejudices, but riven with contradictions typical of the material relationship between, for example, capitalist and worker. To ignore such contradictory relations associated with the use of 'objective' facts serves only to misconstrue the workings of the current society. This is where the work of Bakhtin (1981) in seeing language -whether numbers or text -as a tension-filled, stratified, contradictory whole with a real material base, that is fought over by those engaged in a dialogue, is relevant. For example, the clash between the VC and UCU branch, over the actual accounting surplus made by Northern University is the product of deeper generative mechanisms such as the commercialisation and marketisation of higher education with senior managers increasingly acting like their private sector equivalents.
Further, the case highlights the dangers of accepting accounting information as neutral. The fundamental point is that accounting information is never impartial and in so arguing campaigners restrict the opportunity to use political, rather than solely economic arguments to challenge the effects of supposedly uncontested numbers. It is more appropriate to accept the premise that such information is the product of a contradictory and contested arena which allows campaigners and activists to mobilise such information. Therefore, it is acknowledged that social movements can utilise accounting information in specific contexts to advance their overall aims. But they must do so in the full knowledge that this information is not neutral because it has been produced within the present system to further the aims of the capitalist (ruling) classes. The results of which has implications for the praxis of accounting academics and the better aligning of accounting with emancipation, as we discuss next.

Discussion
The inspiration for this paper has come from the work of Haslam (2003, 2006) which opens up a field of study in critical accounting research that has the potential to influence the struggle for an emancipated world. In this respect, their work deserves not just to be noted but seriously engaged with, as is attempted in the following.

The "facts" of struggle
Among the conclusions made by Haslam (2003, 2006) three points stand out; that campaigners can utilise accounting information; this is primarily done by presenting accounting information as factual; and that accounting information has no necessary class belongingness. Taking each point in turn, the example in this paper confirm Gallhofer and Haslam's first conclusion and provides evidence that it was not just a historical phenomenon but continues among today's activists or as Tinker demands it is 'present in the present' (Tinker, 2004: p. 195). On the second conclusion, the example confirms the dominant approach of campaigners to present accounting information as facts in a manner not unlike the functionalist approach of Mainstream Accounting Research (MAR) (Chua, 1986;Hopper & Powell, 1985). This potentially poses the campaigner/new social movement with the possibility of being constrained by the priorities of managerialism embedded in the accounting information (Ogden & Bougen, 1985) or undermining their own position. The latter runs the risk that not only are questions raised about the interpretation of accounting information by the powerful but also the interpretation by the campaigners themselves. This is the position Gallhofer and Haslam reach when they cite Stuart Hall to support their proposition that accounting can take on different meanings and significance: 'meaning does not depend upon how things are but on how they are signified, so similar events can be signified in different ways' (Gallhofer & Haslam, 2006: p. 248). The problem is such theorising can be lost on campaign supporters and thus undermine their support.
To save us from this outcome we need to identify the contradictory nature of accounting information in a material understanding of capitalist society, whose form and content is imbued with its priorities (Ogden & Bougen, 1985). Considering alternative accounting systems may be one method of overcoming this impasse (Blackburn et al., 2014;Dillard & Yuthas, 2013). In contrast the cases that Bryer (2014) examines, where the formation of a co-operative provides for a degree of independence whilst still operating as part of the capitalist system. Nonetheless, the workers in Bryer's examples still find themselves subject to the logic of capital accumulation as illustrated by one worker's comments that it '. . . is right that we need to keep aware of our profit through our accounts, but a lot of the time I feel like profit controls us' (Bryer, 2014: p. 5). The trade union in this paper did not (nor cannot) generate their own alternative accounting information under the present system. Instead, they utilise accounting information generated by their employer, in a dialogical manner, with the aim of defending their pay and national bargaining rights.

The class basis of accounting information
In contrast to Gallhofer and Haslam's third conclusion -the claim concerning the lack of a necessary class belongingness of accounting information -the theoretical framing of this paper locates accounting information as part of the fabric of the capitalist class system. This is not to argue that Gallhofer and Haslam ignore the issue of class. To clarify, Gallhofer and Haslam respond to Tinker's (2004) critical review by stating: A contention here is that accounting is a practice, actual as well as potential, with no class belongingness or political affiliation in an absolute sense. Rather, it bears, in different ways in the trajectory of its dynamic, the signs of conflict and tension -including class conflict and tension -and is struggled over. Gallhofer and Haslam (2004: p. 201) We wholeheartedly agree with the second sentence in the above quote. However, it seems that the formulation of 'no class belongingness or political affiliation in an absolute sense' leaves open interpretations of accounting information, which could undermine the radical emancipatory possibilities of critical accounting praxis that Gallhofer and Haslam are aiming for. Hence, if accounting information has no necessary class belongingness, we assume that it is in some sense beyond or separate to class society. The concern here is that Gallhofer and Haslam have ontologised accounting information as a separate realm 'understood as an element of the communicative structure of society' (Gallhofer & Haslam, 2004: p. 200). Further, whereas it may well not be their intention, this opens up a scenario where the wider economic structures are not addressed or at least, not seen as a central concern for critical accounting praxis.
In contrast, the theoretical framework developed in this paper seeks to firmly root accounting information as emergent from the very nature of capitalist relations; it is in no way separate from capitalism but is interwoven into the fabric of the system. As we have argued, this is a system based on a series of contradictions (Harvey, 2014); the most fundamental of which is the struggle between capital and labour. Hence, the contradictory nature of accounting information is generated by the contradictory nature of capitalism.
This point is further highlighted when Gallhofer and Haslam address the question of agency. While seeking to work with 'the lens of a Marxist oriented praxis' (Gallhofer & Haslam, 2004: p. 202), they accept the critique of Marx's emphasis on the centrality of class in agency: It is the case that we would not embrace a working with Marx that narrowly conceives class to be the agent of radical universal change and that takes history to be inevitably progressing towards the overcoming of basically capitalist structures. Gallhofer and Haslam (2004: p. 206; emphasis in the original) While again we agree the inevitability of working class transcendence of capitalism is a distortion of Marxism; however, removing class as the central agent of universal change raises the question of what replaces it. Clearly Haslam (2003, 2006) see that labour has an important role to play with the emphasis in their historical writings on Champion, The Match Girls strike and the socialist newspaper Forward. However, they appear to inadvertently accept the premise of the post-industrial and post-Marxist writers that in the modern era labour is either no longer relevant or can at best play a severely restricted role in changing society.
As an alternative, the theoretical framework in this paper seeks to hold onto the centrality of the working class as the universal agent of change, whilst continuing to stress the role accounting plays in relation to the very fabric of the capitalist class system. Cooper (2015) emphasises the importance of labour when she argues only human labour can create value. She links this argument to nature of accounting and explains how it is vital to the current economic system, pointing out that: Marxist theory would argue that accounting technologies play a key role in the control of labour and the extraction of surplus value from labour by increasing productivity and reducing wages. Cooper (2015: p. 75) Furthermore, in support of her argument she highlights  claim that accounting is 'capitalism's premier control system because it provides senior managers with objective measure of the generation of and realisation of surplus value' (Cooper, 2015: p. 75). Moreover, it is the working classes position within capitalism that ensures the need for accounting to control and direct their activities to safeguard a greater surplus -hence accountancy and its information does have an essential class belongingness.

A question of praxis and agency
As previously stated the main aim of this paper has been to identify the most effective ways critical accounting researchers can challenge and change society; moving towards a more emancipated and equitable social order. In this regard, the question of praxis is a recurrent theme within critical accounting literature. For example, Neu et al. (2001: p. 741) ask: how can we intervene in social struggles in a manner that takes advantage of our expertise and offers the greatest potential for transforming and improving social practices?
In contrast to Neu et al.'s (2001) emphasis on intervening (one implication being academics simply pass on their expert opinion to the masses), Cooper et al. (2005) argue that social accounts which do not make connections with the agency of social movements will be flawed in terms of their usefulness and effectiveness.
An alternative formulation of praxis comes from Spence's call for civil society to use social accounting that seeks to move beyond the direct influence of the economic base; with a social accounting not undertaken by corporations but civil society organisations that expose the contradictions of the present economic modes of organisation (Spence, 2009: p. 206). Such activities and tactics are to be welcomed and civil society organisations (including social movements) can become enlightened and changed by such activities. However, it is important to guard against the problems associated with Bryer's (2014) examples where groups -however well intentioned -can become co-opted into what Gramsci describes as the 'hegemonic coalition of business, government, professional and intellectual elites' (Spence, 2009: p. 210). Indeed as Spence recognises green campaigns, now only attempt to provide ecological solutions within capitalism; frustration has forced them to abandon outright criticism in place of making the 'business case for sustainability'. Such problems raise serious questions about agency and the relationship between accounting academics and social agents. The question of agency is crucial and critical accountants should guard against looking to enlightened managers or businesses, as is common in much of the Corporate Social Reporting (CSR) and environmental accounting literature.
Awareness of the limitations of such an emphasis has led us to develop an alternative perspective, akin to Cooper and Coulson's (2014) formulation that seeks a fusing of the academic and the activist. The disclosure of accounting information in isolation has little or no impact on the struggle for emancipation. Even the introduction of completely new reports does not change the social world. However, linking accounting information to a social movement can have a greater impact. We recognise and support Cooper et al.'s (2005) argument that accounting information in and of itself is not enough to affect social change; it needs to be coupled with a social movement that can impact on the existing social structures.

Conclusion
This paper has been inspired by the struggles of campaigners and trade unionists, as well as the work of Haslam (2003, 2006). We have argued that it is the combination of accounting and social movements that, to borrow Gallhofer and Haslam's terms, better aligns accounting with emancipation. In large part, this paper has sought to illustrate their work in the context of a contemporary trade union dispute. This culminates in seeing agency beyond the ranks of the academics (including critical accountants) and in the social movements including mobilising accounting information. This utilisation of accounting information was not pursue on the basis of trying to establish a 'real' dialogue with the VC; we think such an outcome is not possible under the current social relations. In part this is because accounting information is imbued with capitalist/managerial priorities, as Ogden and Bougen (1985) and Owen and Lloyd (1985) have argued. But also because social movements (including organised labour) do not have, nor ever will have, enough counter-hegemonic force to develop their own alternative accounts that will gain widespread support within capitalist social relations. This argument brings us to the position where to change accounting we have to change capitalism, not the other way round.
To reach such an unambiguous conclusion it is necessary to reject the formulation of accounting information as having no necessary class belongingness. Instead accounting information is seen as being integral to capitalism, information produced by capitalists for capitalists. However, capitalism is a complex and contradictory system and so the information it relies upon can be used by social movements to challenge existing exploitation and oppressions. It has been argued that a more appropriate theoretical framework within which to see this is provided by a Bakhtinian analysis of language and a classical Marxist understanding of society. The theoretical framework adopted in this paper stresses the importance of class, not in an unchanging and deterministic manner but instead emerging from the dynamic capitalist relations of production. In turn it emphasises the interaction of structure and agency that sees accounting information, like language, as a contradictory expression of capitalist society.
We conclude with a reminder of Marx's own attitude towards praxis. In the Theses on Feuerbach, Marx applies his grasp of the dialectic and materialist philosophy to illuminate the weaknesses in Feuerbach's "contemplative materialism" famously concluding: 'the philosophers have only interpreted the world in various ways; the point, however, is to change it' (Marx, 1988: p. 574; emphasis in the original). Any discussion of changing society will quickly come to the point of asking how this change can be achieved. Haslam (2003, 2006) highlight the role of trade unions and campaigners as agents for change in the struggle for emancipation. This paper has followed in those footsteps and in doing so sought to extend the discussion of agency and praxis in the critical accounting literature.