STUDYING MANAGERIAL WORK: A CRITIQUE AND A PROPOSAL [1]

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/228048428 Studying Managerial Work A Critique and a Proposal ARTICLE in JOURNAL OF MANAGEMENT STUDIES · MAY 2007 Impact Factor: 4.26 · DOI: 10.1111/j.1467-6486.1987.tb00702.x CITATIONS READS 102 340 1 AUTHOR: Hugh Christopher Willmott City University 238 PUBLICATIONS 9,692 CITATIONS SEE PROFILE Available from: Hugh Christopher Willmott Retrieved on: 03 February 2016 Journal of Management Studies 24:3 May 1987 0022-2380 $3.50 STUDYING MANAGERIAL WORK: A CRITIQUE AND A PROPOSAL''' HUGH WILLMOTT Organization, Personnel and Employment Division, Management Centre, Aston University INTRODUCTION A striking point about empirical studies of managerial work in capitalist enterprise is their disregard for its institutional formation and significance.' ' In general, the institutional conditioning and consequences of managerial work is either bracketed, taken for granted, or treated as an independent variable.'^' The purpose of this article is to highlight the hiatus between 'behavioural' and 'institutional' accounts of managerisJ work and to suggest a means of overcoming this dualism between'action'and'system'(Giddens, 1976; 1979; 1984). More specifically, the paper focuses upon the neglect of institutional analysis in behavioural studies of management practice. Studies of managerial work are found wanting in four inter-related respects. First, as already noted, they abstract the activities of individual managers from the institutional arrangements in and through which they act. In doing so, such studies generally disregard how the work of the manager is accomplished by enacting, and thereby reconstituting, institutionally produced rules and resources. Second, and relatedly, their focus is upon differences of individual behaviour or group allegiance, and not upon managerial work as expressive of the (developing) institutional arrangements that are at once a condition as well as a consequence of managers' actions. Third, these studies uncriticzJly trade upon a distinction between the technical/formal and the political/formal elements of managerial work. TechnicjJ/formal elements are assumed to be legitimate or politically neutral because they are ofTicially sanctioned. What is 'political' is thereby restricted to 'informal' departures or deviations from an officially sanctioned order, and even then such departures are generally understood to be of only local, organizational significance. The final limitation of these studies concerns the lack of revealed awareness about the problematical relationship between the observer (researcher) and the observed (manager). They do not appreciate the value-relevant (Weber, 1949) and cognitive interest-dependent (Habermas, 1972) nature of accounts of managerial work.'*' This failure to reflect upon, and stimulate an appreciation of, the intrinsically open and problematical nature of the relationship between Addressfor reprints: Hugh Willmott, Organization, Personnel and Employment Division, Management Centre, Aston University, Gosta Green, Birmingham B4 7ET. 250 HUGH WILLMOTT observer and observed allows existing studies to appear to provide a mirror image of the reality of managerial work. This is especially the case when they do little .more than reflect commonsense accounts of what managers do, and when so few alternative accounts, informed by a critical perspective, have been undertaken.'^' In this respect, it is relevant to acknowledge that the proposed framework does not escape dependence upon particular values and interests.'*' Specifically, it is guided by a concern to highlight the unavoidably social or relation^ - and, therefore, moral and political-character of managerid action, as well as by an interest in revealing how this action may both preserve and conceal what is problematical within organizations and society. The article falls into three main sections. In the first section, jdternative conceptualizations of management are briefly outlined before critically reviewing three major empirical studies of manageriad work by Dalton, (1959) Kotter (1982), and Mintzberg (1973). The second section extends the critical examination of these studies by considering alternative frames of reference for making sense of capitalist work organization. Finally, in the third section, Giddens' theory of structuration is advanced as providing an alternative methodological framework for the study of managerial work. Its relevance is then illustrated by reference to empirical materials presented by Nichols and Beynon (1977) in Living with Capitalism. ACCOUNTING FOR MANAGERIAL WORK In his textbook treatment of management, Child (1977) distinguishes between three different conceptualizations, or images of management. The first image of management is of the 'economic resource' that performs technicad, administrative functions. A second, related image is that of the 'professional corps' which is identifiable less by its function than by its expertise and credentials. Underdeveloped in comparison. Child suggests, is an image of the 'political aspect' of management {ibid. 113). When examined from this aspect, management is conceptualized as 'a system of power and authority within which different personal and group strategies are pursued' {ibid.). Within the 'political' literature on mjinagement. Child identifies three strands: studies that locate managers in the class structure {e.g. Nichols, 1969; Stanworth and Giddens, 1974), studies that attend to managers' orientations within and between different organizational and cultural settings {e.g. Sofer, 1970; Turner, 1971) and, finally, studies that focus upon the social and political processes embodied in managerial work (e.^. Dalton, 1959; Mintzberg, 1973). However, although these strands are exposed in Child's overview, the rifts between them are not. 'Institutional' writers have studied managers' positions in the 'system of power and authority' independendy of the examination of managers' personal and group strategies. And, on the other hand, 'behavioural' writers have researched managers' strategies without understanding these to be STUDYING MANAGERIAL WORK 251 a medium and outcome of a wider 'system of power and authority'. The contention that research has been premised upon a dualistic separation of institutional and behavioural dimensions will now be explored and illustrated by reviewing briefly three major contributions to the study of managerial work.''' Mintzberg In The Nature of Managerial Work (1973), Mintzberg presents an analysis of the work of five chief executives. To interpret his findings, he builds upon role theory to advance a contingency view of managerial work in which variations between managerial role sets are attributed to the deterministic influence of four nested sets of variables: 'environmental', 'job', 'person' and 'situational'. Or, as he puts it, 'the work any manager does at a certain point in time can be described as a function of the four "nested" sets of variables'. Mintzberg's approach represents the rejility of managerizil work as a set of discrete, 'observable' activities. The mix of these activities is then associated with the occupancy of a formal office or position which is seen to be shaped and modified by the four nested sets of variables. By conceptualizing the content of managerid work in terms of discrete activities, and by drawing upon contingency theory to account for variations in managerial work, Mintzberg effectively disregards the social or relational nature of managerial work. Not that he fails to note the significance of'environmental' and 'situational' variables. But this recognition that managerial work is shaped by institutional 'variables' does not lead him to develop a relational understanding of its reality. Instead, Mintzberg seems to be saying this: when we make sense of managerial work, we must not make the mistake of explaining it solely in terms of the personality characteristics ('person' variable) of the individual manager. In addition, we must recognize the impact of institutionjJ and interpersonal forces: the 'environmental', 'job' and 'situational' variables. In this way, Mintzberg appears to take account of the social nature of managerial work. But he does so without appreciating the relational and contested production of this reality. Yes, he does conceptujJize managerial work as the outcome of an interaction between 'social' variables. However, in his structured observational method, where work content is equated with commonsense description, and also in his theory, which involves the unproblematic identification of roles and variables, all appreciation of the socieJ formulation and maintenance of the 'content' of managerial work is excluded.'*' And so, despite an espoused concern to examine managerial work in terms of the institutional roles performed by managers, Mintzberg's methodology and conceptual framework effectively deny him the possibility of studying the historical and politiceJ processes that underpin, channel and provide rationales for the work that managers do. Dalton In turning to consider Dalton's Men Who Manage (1959), it might be anticipated that a more penetrating account of the 'political aspect(s)' of managerial work 252 HUGH WILLMOTT would be found. Certainly, the method of participant observation enabled Dalton to get beneath the surface of the formal and/or officially sanctioned features of managerial work. In doing so, he exposes much of the underlife of organizations {cf Goffman, 1959) in which 'informal' arrangements replace, impede and are 'mixed' with formal, ofTicizd procedures. However, on closer inspection, it is apparent that Dalton's appreciation of the social and political processes of managerial work is quite limited. There is much reporting of the shifting individual and clique-based strategems developed by managers to 'loosen controls on themselves and tighten them on others' (Dalton, 1959, p. 19). But the focus of Dalton's research is almost exclusively upon the psychological struggles of the individual manager who is seen to strive to reconcile (unreconcilable) 'rational, emotional, social and ethical claims' in the context of large, impersonal corporations {ibid. p. 258). In elaborating and illustrating the existence of tensions between roles and their players, Dalton's analysis conceptualizes these tensions primarily as the result of personal sentiments rather than, say, the contradictory, oppressive structure of social relations. There is minimal detailed description or theoretical penetration either of role tensions or of the structural source of the rules and resources drawn upon by managers in the pursuit of their 'personal' values and interests. Kotter The fmal study to be considered is Kotter's The General Managers (1982). Like Mintzberg, Kotter focuses his attention upon the work of a number of senior managers. In doing so, he concurs with Dalton in the view that 'reed progress' in the study of managerial work depends upon the use of unstructured observation methods {ibid. p. 153). These, Kotter indicates, are essential if the researcher is to get 'inside' descriptions of activities undertaken or roles played in order to reveal how managers construct and maintain their relationships with others. Kotter's study begins to disclose how 'effective' managers skilfully use their access to institutional resources to build political alliances and influence people. He reve2ils, for example, how managers contrive to glean information and support from supervisors without appearing to be over-demanding or inadequate; how they seek to motivate and supervise subordinates, and how they elicit co-operation from corporate and external groups despite resistance, etc. {cf. Blau, 1965). As Kotter observes, the managers in his study tried to make others feel legitimately obliged to them by doing favours or by stressing their formal relationships. They acted in ways to encourage others to identify with them. They carefully nurtured their professional reputations in the eyes of others. They even manoeuvred to make others feel that they were particularly dependent on the general managers for resources, or career advancement or support (Kotter, 1982, pp. 69-70). STUDYING MANAGERIAL WORK 253 It might be said that Kotter's observations are merely statements of the obvious. Don't we already know that managers, like everyone else, skilfully construct and employ symbols and other 'stage props' to secure the sense and solidity of their position and prerogative? Perhaps we do, but most studies of managerial work serve to dim rather than heighten our awareness of this, as Kotter (1979; 1982) forcefully points out. His major contribution, then, has been to highlight the building and maintaining of networks of 'relationships' in a sine qua non of general meinagerial work. In doing so, he reveals that the power that is exercised and replenished through the successful management of inter-personal networks should not be regarded as peripheral to managerial activity nor be treated as a deviation from formally or officially defined roles. In these aspects, Kotter's study goes further than most of its predecessors in acknowledging and stressing the central importance of social relationships and power in the routine accomplishment of managerisd work {cf. Sayles, 1979). However, Kotter's interest in relationships and power extends only to the identification of their significance and potential for achieving a more effective use of the individual manager's talents. The 'position-power' of managers is perceived to be unproblematical. According to Kotter, 'politicking' by managers is necessary only because of the complexity of large organizations that makes it difficult to obtain reliable current information about diverse operations. No mention whatever is made of the structures of social and economic relations that support the legitimacy of the rules and the accessibility of the resources drawn upon by managers in deflning, refining and defending the content of their work. A unitarist or perhaps pragmaticjilly pluralist view of the organization is thus taken for granted (Kelly, 1982) as the pursuit of existing priorities by more effective managers is assumed to be in the interests of everyone. To summarize this section, it has been argued that prominent jmd influential studies of managerial work have been guided by frameworks of interpretation that have disregarded or trivialized its institutional reality and signiflcance. By separating work from its social context, these 'behavioural' studies have largely disregarded the 'political aspect' of managerijil work; or they have identifled it exclusively with the skills and strategies devised and applied by individujil managers to perform their formally deflned roles and/or to advance their career interests. Overlooked or obscured are the institutional grounds of mcuiagerial work as an expression of politico-economic relations of power. This critical assessment of the capacity of such studies to illuminate the political aspect of management wiU now be expainded. To do this the 'unitar/ and 'pragmatically pluralist' assumptions common to these studies will be compared and contrasted with Em edternative, 'radicsd' view of the structure of power relations in organizations and society. 254 HUGH WILLMOTT TOWARDS AN ALTERNATIVE AGCOUNT OF MANAGERIAL WORK In the previous section, the political aspect of management was equated with the view of management as a 'system of power and authority within which different personal and group strategies are pursued' (Child, 1977, p. 113). The particular virtue (and vice) of this formulation is its capacity to include (and its failure to differentiate between) the structural principles of the system and the associated structuring of individucd's strategic conduct. To characterize the variety of ways in which 'system' and 'strategic conduct' can be conceptualized, it is relevant to refer briefly to Fox's (1973; 1974) distinction between unitary, pluralist and radical frames of reference. By drawing selectively and critically upon Fox's analysis, differences in the formulation of the relationship between observer (researcher) and observed (managers) can be highlighted. Unitary, Pluralist and Radical Views of Management From a unitary standpoint, the structure of social relations within organizations is seen to embody rational efforts to develop the most efficient and effective means of achieving common interests and objectives. When studied in this light, managerial work is regarded primarily as an expression of the technical division of labour required to realize organizational goals. In other words, the work of managers is accounted for as a functional element of organization whose responsibility is for the rational design of all aspects of organization so that the shared objectives of its members can be secured. This view of the structure of socid relations is most clearly evident in classical and human relations accounts of the role of management. However, as noted earlier, strong traces of this perspective can also be found in recent studies of managerial work (e.g. Mintzberg, 1973). From a plur2ilist standpoint, the unitary perspective is unconvincing because, in large modern organizations, the complexity of the technical division of labour is such that individuals are conditioned by specialized training and motivated by self-interest to form coalitions for the pursuit of sectional objectives. Conflicts between a plurality of power-holders within organizations is thus seen to be endemic, and managerial work is understood to involve a continuous effort to gain consent and/or contain conflicts of interest in ways which, in the long run, allow at least minimal fulfilment of all members' sectioned objectives. This perspective has been advanced by theorists who have regarded the classic, unitary view of management as an unobtainable ideal. Desirable as this ideal might be, to confuse it with recdity or to strive to make recility conform to it, is perceived as self-defeating. Again, evidence of elements of this pragmatic pluralist framework is present within studies of managerial work reviewed earlier (e.g. Kotter, 1982). Finally, the radical standpoint challenges assumptions made in both pluredist and unitary accounts of the structure of social relations in organizations (and society). The basic charge levelled against them concerns their failure to STUDYING MANAGERIAL WORK 255 recognize the politico-economic nature of the technicEil division of labour. From a radical perspective, the horizontal and verticEil differentiation of tasks between individuals and groups cannot adequately be explained by references to functional imperatives. Instead, it is understood to reflect and sustain the structure of power relations within society. Managers are seen to play speciedist, technical roles in the division of labour. But these roles are perceived to be primarily conditioned by the political economy of labour processes in capitalist society. Capitalist Work Organization Central to Fox's conceptualization of the radical frame of reference is a distinction between spontaneous and manipulated forms of'consensus'.''' Whereas spontaneous forms of consensus are understood to arise when all involved are 'moved by a common cause', a manipulated consensus describes the position when one group has the physical or institutional power to exploit others' dependence in the pursuit of its own sectional interest. As Fox (1974, p. 284) puts it: People do not come together freely and spontaneously to set up work organizations, the propertyless many are forced by their rued for a livelihood to seek access to resources, owned or controlled by the few (emphasis added). Fox highlights the structurd dependence of the many upon the few. He also emphasises that the latter's influence in shaping expectations and aspirations tends to create a situation in which socially arbitrary conventions cind principles are in effect naturedized. However, conspicuous by its absence from his anadysis is any sustained appreciation of the existence Jind significance of structursd contradictions and forms of resistance within socicd systems. Certainly, he remarks upon the asymmetrical distribution of power and the conflicting objectives represented in the existence of low discretion roles. But little or nothing is said of the structural cleavages of interest - the contradictory forms of unity - that can be said to be inherent within capitcdist relations of production. Indeed, it would seem that, in Fox's allegedly radicjd frame of reference, the powerful have the unassailable power to suppress or over-rule such contradictions by using their power to 'determine the power of the many' {ibid. p. 284). In contrast, the position taken here is that relations of production are premised upon and structured around a basic contradiction between the 'principles' of socialized production and private appropriation (Giddens, 1979, pp. 136-7). Following Marx, Hindess and Hirst (1977) have outlined the central structural dynamic of those relations sdthough, as they observe in a later publication, it is 'capitals', not capitalists, which exist in capitalism in its socialised form (Cutler et al. 1977, p. 312). Capitalists buy mecins of production and items of personsd consumption from each other. They buy labour power from labourers in exchange for wages. With these wages the labourers buy items of personal consumption from 256 HUGH WILLMOTT capitalists and must then sell their labour power for a further period in order to be able to buy further means of personal consumption. Appropriation of surplus labour here depends on a difference between the value of labour power and the value that may be created by the means ofthe labour power. Surplus labour takes the form of surplus value Thus capitalist relations of production defme a mode of appropriation of surplus labour in the form of surplus value, and a social distribution of the means of production so that these are the property of non-labourers (capitalists), while the labour power takes the form of a commodity which members ofthe class of labourers are forced to sell to the class of non-labourers (Hindess and Hirst, 1977, p. 10). This contradiction between private appropriation (possession) and socialized production (non-possession) underpins and succours (but does not determine) a class division of labour. This is reflected in tensions within structures and strategies of control as capital depends on labour for the cycle of production and valorization to be sustained (Cutler et al. 1977; Storey, 1985a). As the servants of socialized capital, managers are formally required to organize the resources at their command to ensure the extraction of surplus value. In performing this function, management is required to forestall or check resistance to control that arises out of the contradiction between socialized production and private appropriation (Storey, 1983). As Winkler (1974) and Zeitlin (1974) have found, top managers, at least, tend to espouse values and priorities that are very similar to those of shareholders (and creditors). More generally, the (albeit ambiguous) class position and institutional role of managers at all levels inclines them towards ideas and actions that are not markedly opposed to the reproduction of a social order in which they enjoy a position of comparative advantage (Goldthorpe, 1982). However, formal requirements and concrete practices do not necessarily coincide. Managers are not omniscient. Even when intending to safeguard shareholders' interest, conflicting interpretation and judgement inevitably enter which, with hindsight, can be questioned. To put this another way, the interests of capital like those of labour, are not given. They have to be organized. Moreover, they are organized by managers who themselves do not occupy an unambiguous position within the structure of capitalist relations of production. The variable measure of autonomy enjoyed by managers can be used to organize and defend their own (specialist and hierarchically positioned) vzdues and interests. Moreover, despite being employed to preserve the interests of capital, most managers share with workers an oppressive requirement to sell their labour power to provide for the items of personal consumption. For these reasons it should not be assumed that managerial action is unmediated by 'opposition and resistance, opportunism and sheer irrationality and incompetence' (Storey, 1983, p. 7). What managers actually do cannot be convincingly analysed without allowing for the influence of'sectional conflicts, professional strategies, internal bargaining, occupational closure and so forth' STUDYING MANAGERIAL WORK 257 (Reed, 1984; Salaman, 1982). However, in taking this into account, it is equally imjjortant not to be distracted or become preoccupied with conflicts within management («.5. Dalton, 1959; Pettigrew, 1985). Otherwise, as Tinker (1984, p. 70) has cautioned, 'the "moderators" and "intervening factors" take on greater significance than that which is being moderated - structural conflict itself. To classify managers simply as 'part of the capitalist class' and 'members ofthe bourgeoisie', as for example Carchedi (1975, p. 48) does, is to overlook how their social location and work situation is constructed in the context of competition and struggle within classes as well as between classes. Crucially, the reproduction of class relations of production founded upon the contradiction between socialized production and private appropriation does not exist independently of non-economic forces (Coward and Ellis, 1977; Pecheux, 1982). Indeed, political and ideological elements may not only condition but also contain, conceal and thereby postpone indefinitely the resolution of this contradiction. In a similar vein, Burawoy (1985) has noted how the process of ideological and political struggle between classes can mollify the opposition of interests. The 'crucial issue', he argues is that the interests that organize the daily life of workers are not given irrevocably; they cannot be imputed. . . To assume, without further specification, that the interests of capital and labour are opposed leads to serious misunderstandings of capitalist control (pp. 28-9). 'Serious misunderstandings' arise where it is assumed that the interests of workers/managers can be 'read off from the primary economic contradiction of capitalist relations of production. In concrete social practices, political and ideological elements, whose plausibility and legitimacy is conditioned by the distinctive features ofthe capitalist mode of production, C£m obscure or minimize the class nature of the structure of production relations. In the factory or at the office, workers do not directly confront 'the class of possessors'. Rather, they encounter managers who, like themselves, exchange their labour for wages,'' ' and who may appear merely to be engaged in the universal, technical task of co-ordinating a complex labour process. From this perspective, the soci£il function of management, which involves preserving a (profitable) difference between the price paid for labour power and the value it creates, may be invisible to workers and managers jdike. The extraction of surplus value is obscured from view as wages appear to cover all the hours at work, not just the hours necesary for the reproduction of labour power; and the individual or collective wage bill appears to be determined, in a quasi-fatalistic manner, by the impersonal mechanism of the market, without awareness of how the allowance for unpaid labour is already incorporated into the structural operation of this mechanism.'"' Moreover, even those who recognize the class structure of capitalist relations of production may pledge or resign themselves to getting what they, individually 258 HUGH WILLMOTT or collectively, can out of'the system'. For, as Storey (1985b, p. 282) has stressed, labour is not simply in the position of being subject to capited, it also has a crucial stake in i t . . . many workers see their fortunes enmeshed in the prevailing order. People deeply criticcd of this order also act to make it work'. 'Co-operation' may thus be as much the product of informed cdculation as it is an expression of ignorance, manipulation or fedse-consciousness. Reflecting such ideas, the notions of a 'fair' wage is regularly formulated in relation to differentials and comparabilities rather in terms of the difference between necessary and surplus labour time. Nonetheless, from the perspective advanced here, such ideas, which serve to reproduce exploitative and oppressive relations of production are regarded as an expression of these class relations. For, as Giddens (1982, p. 40) has observed: it is quite possible to conceive of circumstances in which individuals are not only not cognisant of being in a common class situation, but where they may actively deny the existence of classes - and where their attitudes and ideas can nevertheless be explained in terms of class relationships (emphasis added). In sum, managerial work cannot adequately be studied and accounted for simply by identifying its economic function. Nor can it satisfactorily be explained purely in terms of its checking of resistance from below. Nor, finally, can it be understood only in terms of the autonomous, if ambiguous, values and interests of managers. Instead, the argument of this article is that managerial work is theorized better as reflecting and sustaining a fundamentally contested structure of social relations in which an institutionalized organization of the interests of capital is tempered and compromised: first by systematic contradictions; second, and relatedly, by individual and collective resistance from below; and, third, by managers themselves who, at the very least, interpret and act out their 'functional roles' in the light of their own (minimally) autonomous cultural and ideological values. This understanding will now be explored and illustrated by drawing upon Giddens' theory of structuration and Nichols and Beynon's Living with Capitalism. THE THEORY OF STRUCTURATION AND THE STUDY OF MANAGERIAL WORK It has been argued that if the study of managerial work is to take account of its institutional contexts as well as the strategic conduct of individual managers, then manageried work must be analysed sis an expression of the strategic conduct of individuEils and as a product of the institutionad order. Giddens (1976; 1979; 1984) has recently developed a conceptual framework that provides a means of doing precisely this: to overcome the dualism in social and organizational theory between 'action' and 'system' (Ranson et al. 1979; Willmott, 1979). His STUDYING MANAGERIAL WORK 259 theory of structuration will first be outlined before illustrating how it provides an alternative framework for the study of managerial work. The Duality of Structure in Interaction Central to the theory of structuration is the idea of structure as a duality. Instead of employing the concept of structure to describe a context in which social practices are situated or played out, it is conceived as the medium of strategic conduct. Structure is thus understood to reside within social practices and not to exist as an external context of constraint upon them {cf Bhaskar, 1978). The particular value of the theory of structuration is that it does not reduce 'action' to a function of 'the system'; nor does it neglect or take for granted the structural conditions and consequences of'action'. Instead, by conceptualizing structure as a duality - that is, as a medium as well as an outcome of (managerial) action - it is possible to better grasp the inter-relatedness of the 'institutional' and the 'strategic' dimensions of social practice. In the process of acting, agents are conceived by Giddens as mobilizing interpretative schemes, norms and other facilities which, collectively, are described as the modalities of structuration (see figure 1). Analytically, the modalities provide the linkage between the process of interaction and the structural components of social systems. Modalities are understood to be drawn upon by actors in the production of interaction. And, at the same time, they are the media of the reproduction of the structural components of systems of interaction. By placing an epoche upon the institutional dimension, the modalities appear in the form of the communication of meaning, the use of power and the application of norms. Conversely, when bracketing the strategic dimension, the modalities appear as the structured properties of social systems - as expressions of signification, domination and legitimation. Interaction Communication (Modality) Structure Interpretative Schemes Power Facilities Sanctions Norms Signification Domination Legitimation Figure 1. The duality of structure in interaction (Giddens, 1979, p. 82) To elaborate, from the standpoint of strategic conduct, the three analytically separable modalities are examined in relation to individuals' efforts to achieve their purpose and/or pursue their interests. From this perspective, the modalities appear as rules and resources strategically drawn upon by agents in the accomplishment of interaction. From an insitutional standpoint, in contrast, these rules and resources are studied as features of systems of social interaction. To repeat, the analytical significance of the modalities is that they provide 'the coupling elements' whereby the analysis of the 'dimension of interaction centred upon the communication of meaning, the operation of relations of power and the application of normative sanctions - is linked to the analysis of the 260 HUGH WILLMOTT Structural components of social systems - where the analytical focus is upon signification, domination and legitimation. In this way, whether the analytical focus is upon actors' strategic conduct or the structural components of social systems, Giddens' conceptual framework incorporates the recognition that The communication of meaning in interaction does not take place separately from the operation of relations of power, or outside the context of normative sanctions. . .no social practice expresses, or can be exploited in terms of, a single rule or type of resources. Rather, practices are situated within intersecting sets of rules and resources (Giddens, 1979: ibid. pp. 81-2). When the institutional dimension of social practices is methodologically bracketed, actors are seen to draw upon the structural properties of socijil systems so as to 'bring off everyday socid interaction. Conversely, when the dimension of strategic conduct is bracketed, it is not the involvement of these properties in the accomplishment of social practices that is of interest, but, rather, their particular composition within a given social system. As Giddens (ibid. p. 80) puts it: To examine the constitution of social systems as strategic conduct is to study the mode in which actors draw upon structural elements - rules and resources - in their social relations. 'Structure' here appears as actors' mobilization of discursive and practical consciousness in social encounters. Institutional analysis, on the other hand, places an epoche upon strategic conduct, treating rules and resources as chronically reproduced features of social systems. To this, Giddens adds: It is quite essential to see that this is only a methodological bracketing: these are not two sides of a dualism, they express a duality, the duality of structure (ibid. emphasis added). Giddens' theory of structuration offers a conceptual framework for connecting the 'strategic' and 'institutional' aspects of managerial work. By employing this theory, the social practices that constitute managerial work can be studied as the skilled accomplishment of agents and as an expression of the structural properties of systems of interaction. Thus, on the one hand, they are seen to be accomplished by managers who strategically develop and enforce rules and deploy resources. From this 'strategic' standpoint, it is evident that they are actively engaged in accomplishing and restructuring regularized relations of interdependence both amongst themselves and with other groups. On the other hand, such conduct is understood to be possible only because of the institution£il rules and resources that are (presently) at their disposed. The value of this framework will now be illustrated by drawing upon empirical material presented STUDYING MANAGERIAL WORK 261 by Nichols and Beynon in Living with Capitalism. This study is of particular relevance because of its authors' explicit concern to connect action and structure. For, as Nichols and Beynon lucidly observe in the preface So much of what passes for 'theory' (even Marxist theory) fails to connect with the lives that people lead, whereas most descriptive social surveys too often fail to grasp the structure of sociEil relations and the sense which people make of ''^' Living with Capitalism The speciflc focus of Living with Capitalism is the relations of production within a modern factory named Chem Co. Included in this is an examination ofthe labour of superintendence' (i.e. managerial work) in the context ofthe management strategy for the organization and control of the labour process. The following extract is illustrative of the practice of the 'new industrial relations' at Chem Co. in which managerial work involves the artful application of socied psychological theories and skills. In common with most qualitative research, it is based upon one informant's account of a set of events. Clearly, it would be of interest, for other purposes, to explore other accounts ofthe same events. However, for the present purposes of illustrating Giddens' framework, it is not necessary to be detained by considerations of the 'accuracy' or 'impartiality' of this particular account. In the following extract from Living with Capitalism, Colin Brown, a comparatively young and inexperienced manager describes how he uses a case of poor time-keeping to manage his relationship with a shop steward. Brown reflects Every man is born to do something and my function in life is to manage. I think this is a problem that most managers have failed to get to grips with. Now take an example. As far as I can see, any man who takes on the job of shop steward wants his ego boosting. But you've got to boost his ego in the proper manner. Now, if I get a bit of trouble - now take an example, perhaps of a serious case of a man who has been perpetually late. Now, I'm the manager, and it's my function to manage. It's my function to discipline this particular man. But I have to deal with the steward. So, what do I do? I take the shop steward aside and tell him that in half an hour's time this man Smith is going to walk into this room. That I'm going to stamp and bang the table and tell him that I'm going to put him out on the road. Then I'll say to the shop steward, 'and v/hatyou can do will be to intervene at this time. Make a case for the man. And we'll agree to let the man off with a caution'. Now the man comes in and I bang the table and the steward says 'Come on, Mr. Brown. Couldn't you give him one more chance?' I relent. The 262 HUGH WILLMOTT shop steward gets out of the meeting and says to him 'I've got you off this bloody time but don't expect me to do it again'. You see the shop steward gets his ego boosted. He gets what he wants and I get what I want. That's what good management is about (Nichols and Beynon, 1977, p. 122). Applying Giddens' framework, Golin Brown can be seen to draw upon a number of interpretative schemes to communicate the reality of his managerial work. He begins by employing the notion that everyone is born to do something, and that his predetermined mission, for which he is naturally fitted, is to manage. This scheme includes the understanding that in any organization there will be a separation between managers and managed. It is this inevitable fact. Brown suggests, that most managers have 'failed to get to grips with'. Brown's belief in the division between managers and managed within the natural order of things is also reflected in his view that shop stewards aspire to quasimanagerial positions because they have a need to get their ego boosted. This, Brown observes, presents the manager with a challenge, to boost the steward's ego in a way that is 'proper' for the effective execution of the management function. The notion of'proper' reflects a second interpretative scheme. Namely, that good inanagement is about the calculated contribution and negotiation of situations in ways that produce the maximum benefit for the minimum cost. These interpretative schemes allow Brown to make sense of and to organize his work. They underpin and are supported by his skilful use of facilities as he manages his interaction with the steward and Smith, the poor time-keeper. His use of facilities - rules and resources - enables him to 'take the steward aside', to rehearse his performance and generally stage-manage the disciplinary scene. Brown makes strategic use of his facilities, {e.g. his power to put Smith 'out on the road') by creatively identifying a low-cost opportunity to reinforce the relationship of domination over the steward. More specifically. Brown stagemanages the situation so that, in the process of getting his ego boosted, the steward becomes both incorporated into the management process and indebted to Brown for enabling him to play out his (managerially defined) role as shop steward. Norms are drawn upon by the various parties to the interaction. On this occasion at least, the steward seemingly accepts the right of the manager to discipline Smith. Conversely, the manager 'recognizes',and indeed exploits, the right of the steward to make out a case for his union member. The positive sanction of the ego boost is used by Brown to ensure that the steward follows his script to the letter, and with gratitude. As long as it works - and this depends appreciably upon the schemes of interpretation and facilities at the command of the steward - the asymmetrical relationship of power between the manager and the steward is hegemonically concealed in the very process of its reproduction. It should be clear from the discussion of this example that the distinction between the three modalities is purely analytical. In practice, interaction STUDYING MANAGERIAL WORK 263 involves a simultaneous and interdependent employment of interpretative schemes, facilities and norms. But, so far, we have considered only how Brown draws strategically upon the modsdities to manage his relationship with the steward.''^' The particular value of Giddens' notion of the duality of structure, however, is that is apprehends how the strategic use of modalities involves a mobilization of structurzd properties. For example, in the interaction between Brown and the steward, their communication can be seen to depend upon the sign or signification systems employed by both parties to discern or 'read' the reality of the situation Euid, in particular, the ontology of self and the other. Brown 'reads' the situation as an interaction between a self who has 'just got to manage' and emother, the stew2u-d, who 'wants his ego boosting'. The interaction is accomplished as Brown identifies self and other within this system of signification. Of course. Brown's understanding may not have been fully shared by the steward. But, for practic2d purposes, the steward's possible disbelief is suspended as he complies with Brown's script and acts out his role (albeit that this may have been played at a distance). Similarly, Brown's exercise of power over the steward depends upon his commzmd over resources. More specifically. Brown draws upon and reproduces a system oi dorhination which gives him authority over others as well as the power to allocate material resources, such as Smith's job. It is this system that shapes and naturzdizes the disciplinary function of the manager. Brown can perform this function only because he acts as the agent of the owners of the material resources and has the power to command others legally invested in him (in the employment contract and in company law). The interaction between Brown and the steward also illustrates how a system of legitimation supports the normative regulation of interaction. What empowers Brown to act is the symbolic value of the role he performs. It is the taken for granted legitimacy of his social position that enables him to take the steward aside. The normative regulation of such interaction is secured through the idea that 'good management' is achieved when the interests of seemingly opposing factions {e.g. capital/labour; Brown/Smith) are seen to be concretely co-ordinated, thereby preserving the basic structure of capitalist relations of production. In uncriticzilly acting out the role of advocate for Smith, the steward colludes with Brown in legitimating the form of his power-play. It is, of course, possible that the steward saw straight through Brown's 'spoofing' and played his role at a distance - perhaps because he preferred a quiet life, perceived himself to be powerless or wanted to lull Brown into a false sense of power or even because he wzmted the (ego-boosting) satisfaction of outsmarting Brown at his own game. However, whatever the steward's intent, the effect of his behaviour is to confirm, for the moment, the 'objective' reality and legitimacy of the normative order.''*' This suieJysis of the interaction between Brown Jind the steward, as reported by Brown, can be further developed by drawing upon the concept of contradiction, discussed earlier, which is also central to Giddens' theory of structuration. 264 HUGH WILLMOTT In Giddens' terminology, contradiction refers to disjunctions or oppositions between the structural principles of social systems. Its major analytical contribution lies in their capacity to illuminate the dynamics of social life. The existence of mutually incompatible or contravening principles of system organization helps to explain why stability and continuity in social systems is impermanent and problematic and, relatedly, why conflicts of interest arise and are worked out but reirely resolved. In the exeimple of Brown's managerial work. Brown requires 'his' men's labour to be reliable: to get to work when it is useful or valuable to him. Moreover, in contrast to other, less visible, breaches of discipline, (e.g. skiving), poor time-keeping poses a direct challenge to management's authority to police the labour process. On the other side. Smith, the man whom he is to discipline, has an interest in minimizing the time that he is subjected to manageried and organizational discipline.''^' This conflict between the msmager and the worker is expressive of the primziry contradiction of capitalist society discussed earlier; that is, the contradiction between the principle of the private appropriation of wealth which presumes, yet is conditionsd upon and negated by, the principle of socialized production. Brown, acting on behalf of the owners of the means of production, requires Smith, as a unit of labour cost, to arrive at work on time so that goods can be produced that will reeJize a surplus. Needless to say, as an individual. Smith is unimportant and dispensable; Brown repeatedly refers to Smith as 'the man'. As Brown threatens, this man can be 'put out on the road' and replaced by a more 'responsible' worker. What is crucisd is the issue at stake, the issue of manageri2d discipline. For if Smith is seen to 'get away' with poor time-keeping, the danger for Brown is that others will follow Smith's example. Thus, Brown's interview with Smith also indicates his own vulnerability in respect of labour on whom he, as an agent of capital, depends. More specifically, it can be seen how the class nature of the division between manager and managed acts as an obstruction to productive activity. Ironically, Smith's labour is necessarily idle during the period ofthe interview. Moreover, Brown is obliged to discipline Smith by making threats that can only deepen the divide between them, an effect that is in direct conflict with the conciliatory design of Brown's 'spoof. In short, the contradictions of capit2ilist relations of production both facilitate the occasion and constrain the effectiveness of Brown's disciplining of labour.''*' CONCLUSION The purpose of this article has been, firstly, to develop a critique of prominent and influential empiriced studies of managerial work and, secondly, to outline and illustrate an alternative approach based upon Giddens' theory of structuration. Through an examination ofthe research of Dalton, Kotter jind Mintzberg a number of major limitations were identified and explored. In general, these STUDYING MANAGERIAL WORK 265 studies were found to abstract the behaviour of managers from the institutional settings and media of managerijd work. Drawing upon Giddens' theory of structuration, the proposal is to advance the study of managerial work by appreciating how, within capitalist relations of production, the work of managers is both a medium and outcome of the structural properties of a social system founded upon the contradiction between socialized production and private appropriation. As Storey (1983; 1985a) has argued, such an approach appreciates how inter-subjective reality requires perpetual reconstruction and, at the same time, recognizes that managerial work, as a labour process, is accompjuiied within a totality of social relations containing contradictory forms of unity. The claim of this article has been that Giddens' concept of structure as a duality is consistent with such an approach, and that it provides a valuable alternative conceptual framework for advancing critical empirical research into managerial work. NOTES [1] In making revisions to this article I have been stimulated and guided by comments and suggestions from Stewart Clegg, David Knights, John Storey, Paul Thompson and Richard Whitley as well as from two referees. I would like to thank them for their assistance, while taking responsibility for any remaining inadequacies of the paper. [2] The paper is restricted to an examination of managerial work within capitalist enterprises. It does not aspire to contribute to the comparative analysis of management within the public sector or within non-capitalist economies. [3] For example, empirical studies of managerial work rarely relate the behaviour of managers to the so-called 'managerial revolution' in which power is alleged to pass from the owners to mainagerial elites within large corporations {e.g. Berle and Means, 1932; Bumham, 1941; Galbraith, 1967). Nor, equally, do these studies connect with the writings of critics of the managerial revolution thesis (e.g. Baran and Sweezy, 1968; Herman, 1982; Scott, 1979). [4] The lack of reflexivity in these studies does not, of course, invalidate their utility. Clearly, in terms of their own (generally implicit) purposes, they have been found to be of (limited) value. [5] Here it is relevant to note the existence of'critical' studies that are broadly comparable with Giddens' framework. Clegg (1975) and Silverman and Jones (1976) both provide theoretically sophisticated analyses of managerial work. In Glegg's case, however, his proposed framework (which is similar to that of Giddens, see Clegg, 1979) is somewhat detached from the brief ansJysis of his empirical materials. Silverman and Jones (1976) provide a much more detailed analysis of their materials, but present only a sketchy outline of the theoretical framework that informs their study. Two papers by Golding (1979; 1980) also make an important contribution to the critical study of managerial work by revealing how symbols play a central role in the depoliticizing and legitimation of managerial power. Roberts and Scapens (1985) present a lucid account of the potential of Giddens' framework for analysing managers' use of accounting information. However, this paper is disappointing in its failure to provide any systematic analysis of empirical data. [6] Nor is it invulnerable to criticism. See, for example. Archer, 1981; Dallmayr, 1982; Gane, 1983; Layder, 1981; Thompson, 1984; WUlmott, 1986. 266 HUGH WILLMOTT [7] It is clearly impractical to review all recent studies of managerial work within this paper. Had space permitted, more attention could have been paid to other relevant studies («.^. Pettigrew, 1973; Sayles, 1979; 1972). For more comprehensive reviews of the literature on managerial work, see Campbell et al. 1970; Glover, 1977; Glover and Martin, 1986; Mintzberg, 1973; Willmott, 1984. [8] Mintzberg recognizes that managers do have some discretion in shaping their commitments. But he does not explore the issue of how roles are negotiated or how collectively managers are involved in enacting/managing the 'external' and organizational contexts which, in Mintzberg's model, are reduced to 'environmental' and 'job' variables. Instead, he regards the exercise of discretion as a defining characteristic of successful management. To quote: 'all managers appear to be puppets. Some decide who will pull the strings and how, and they then take advantage of each move that they are forced to take. Others, unable to exploit this high-tension environment, are swallowed up by this most demanding of jobs' (Mintzberg, 1973, p. 51). [9] It is significant that Fox employs the term consensus rather than, say, compliance. Certainly, he acknowledges that those in control 'do not enjoy a monopoly of ideological communication' (1974, p. 92). He also recognizes that occupants of low discretion roles tend to regard them 'as expressive of management's intent to use them as instruments towards ends they do not share' {ibid. p. 93). But these sources of resistance and disaffection are, in Fox's assessment, marginal because 'the few can use their power to determine the power of the many not only directly but also indirectly through the many agencies of socialization, communication and attitude formation' {ibid. p. 284). Fox's formulation of the radical frame of reference thus conveys a strong impression that only incompetence or misjudgement on the part of the owners and controllers of resources is likely to destabilize the working consensus {cf Wrong, 1961). This impression is reinforced by his contention that 'society is already in the shape which serves their essential interests' {ibid. p. 278); and, further, that 'all the social institutions, mechanisms and principles which it is crucially important for them to have accepted and legitimized are accepted and legitimized already' {ibid. p. 277). [10] However, in contrast to workers, a greater proportion ofmanagers'labour is employed 'unproductively' to ensure that surplus value is appropriated and realised. The cost of this unproductive labour is, of course, met from the surplus value produced by workers and (to a lesser extent) managers when performing a co-ordinating role. This burden upon labour is also routinely obscured, for reasons discussed below. [11] In this respect, the obscuring of class exploitation is made easier to the extent that employed labour may be buffered from the harsher effects of continuing rationalization of productive processes by the establishment of a strong competitive or monopolistic advantage in product markets and/or the profitable investment of income in capital markets, etc. In this light, the strategic control oflabour processes and the concrete organization of workers' interests must be examined not only in relation to workers' capacity to resist the displacing devaluing, de-skilling and intensifying of their labour, but also in relation to the capacity of corporate managers to pursue and exploit alternative avenues for achieving a favourable return on capital. (Knights and Willmott, 1986; Knights, Willmott and CoUinson, 1985; Littler and Salaman, 1982). [12] Although Nichols and Beynon set out with the intention of connecting social action and social structure, they are only partieJly successful. Certainly, they highlight the fundamentatl, systemic contradictions and conflicts of interest that are characteristic of capitalist relations of production. They also acknowledge that 'to talk of capital is to talk of a social relationship' and that 'if social class means anything STUDYING MANAGERIAL WORK 267 at all it is to be found in the real lives of real people' (Nichols and Beynon, 1977, p. 76). However, very little is presented that reports and analyses actual interaction between classes and members {e.g. managers and managed) at Chem Co. Nor, indeed, is it clear how Nichols and Beynon theorize managerial work. For, on the one hand, they assert that it 'can increasingly be seen as labour {ibid. XV), while, on the other hand, managers appear not to act in any way as labour. That is, they are portrayed and theorized as pliant and passive agents of capital who are 'driven by the impersonal force of capital' {ibid.) and, again, whose actions are subject to the dictates of an impersonal force-capital' {ibid. p. 42). While labourers are seen to engage in various forms of resistance-albeit marginal, individuEilistic and often futile - managers are presented as mere functionaries. Moreover, when managers are perceived to deviate from their functional roles, this is explained in psychologistic terms. For example, the deviations of Edwards Blunsen are related to his personal ambition, those of George Smith are associated with his personal values and those of John Baird are anedysed in terms of his youthful innocence. No doubt an appreciation of these personality traits and biographical details is important for understanding each man's distinctive approach to his work. But if the intention is to study the work of these managers as an expression of class relations, then it is necessary to develop a dialectical analysis of the connection between individual character and social structure. The basic shortcoming of Nichols and Beynon's account of managerial work is that it embodies a dualism between individual and relational action. Instead of studying 'the labour of superintendence' as they call it, as a relational activity, the actions of individual managers are theorized as a deterministic product of the social structure of capitalism. Despite their stated intention to reveal the interdependence of action and structure, Nichols and Beynon consistently treat the latter as a container and/or determinant of the former. Or, as they themselves summarize 'we have attempted . . . to see the way class relations bear on the individual and to locate individuals always with class relations' {ibid. pp. 203-4, emphasis added). What is lacking theoretically is a conceptual framework capable of^ penetrating and overcoming the analytical dualism between action and structure. Empirically, what is absent is observational data on the actual conduct of class relations/managerial work and Chem Co. For a more developed critique of Giddens' theory of structuration as well as Nichols and Beynon's Living with Capitalism, see Knights and Willmott, 1985. [13] To be more precise, it is an example of managerial work as accounted for by Brown and mediated by Nichols and Beynon. [14] With reference to these structural properties, it is worth repeating the point that the distinction between systems of signification, domination and legitimation is an analytical one. As Giddens (1979; pp. 106-7) stresses, 'if signification is fundamentally structured in and through language, language at the same time expresses aspects of domination, and the codes that are involved in signification have normative force. Authorization and allocation are only mobilized in conjunction with signifying and normative elements and, finally, legitimation necessarily involves signfication as well as playing a major part in co-ordinating forms of domination'. [15] Of course, for most employees, for most of the time, the comparative loss of discretion or control over their labour power is not challenged through a collective struggle between management and labour. This is because meinagerial and organizational discipline is widely regarded as more or less legitimate insofar as the contract is perceived to be entered into more or less freely. However, the fact that organized resistance is exceptional certainly does not imply that contradictions do not exist. Nor, relatedly, does it suggest that there are no endemic conflicts of interest. 268 HUGH WILLMOTT [16] Potentially, union organization provides the means of countering, or at least curtailing, managers' powers to exploit the weak position of atomized workers. Strong union organization can protect the position of labour by making it less dispensable or substitutable, and thereby exploiting the fundamental dependence of capital upon it. Simply because collective resistance presents the greatest potential challenge to management control, it is this source of power that Brown and the more progressive managers at Chem Co. strove to undermine by stealth. The policy of incorporating trade unionism into management is well illustrated in the stagemanaged meeting between Brown, Smith and the steward. Instead of seeking to exclude the trade union representative, Brown invites him to play an 'active' part in the disciplinary proceedings. Through a petty conspiracy between Brown and the steward, Smith is given the impression that his steward has successfully defended his interests, and the illusion of sound union representation is maintained. At the same time, instead of taking the hard line by carrying out his threat to put Smith 'out on the road', Brown engineers a low-cost demonstration of the fair and enlightened nature of Chem Co. as an employer. At the very worst, from Brown's point of view, he will be obliged to suffer Smith's poor time-keeping for a few more days before 'putting him on the road', with the steward's consent. Balanced against this minimal cost, his 'spoofmg', if undetected, plays a mundane but nonetheless essential role in managing the fundamental contradiction between private appropriation and socialized production. REFERENCES ARCHER, M . (1981). 'Morphogenesis versus structuration'. British Journal of Sociology 31, 2, 346-61. BARAN, P . and SWEEZY, P. M. (1968). Monopoly Capital. Harmondsworth: Penguin. BERLE, A. A . and MEANS, G . C . (1932). TheModem Corporation and Private Property. New York: Macmillan. BHASKAR, R . (1978). 'On the possibility of social scientific knowledge and the limits of naturalism', youma//or M« Theory of Social Behaviour, 8, 1, 225-52. BLAU, P. (1965). The Dynamics of Bureaucracy. University of Chicago Press. BuRNHAM, J. (1941). The Managerial Revolution. Harmondsworth: Penguin. BURAWOY, M . (1985). The Politics of Production. London: Verso. CAMPBELL, J. P., DUNNETTE, M . D . , LAWLER, E. E. and WEICK, K . E . (1970). Managerial Behaviour, Performance and Effectiveness. New York: McGraw-Hill CARCHEDI, G . (1975). 'On the economic identification ofthe new middle class'. Economy and Society, 4, 1-86. CHILD, J. (1977). 'Management'. In Parker, S. R., Brown, R. K., Child, J. and Smith, M. A. The Sociology of Industry, 3rd Ed. London: Allen and Unwin. CLEGG, S. (1975). Power, Rule and Domination. London: Routledge and Kegan Paul. CLEGG, S. (1979). The Theory of Power and Organisation. London: Routledge and Kegan Paul. COWARD, R . and ELLIS, J. (1977). Language and Materialism. London: Routledge and Kegan Paul. CUTLER, A. J., HINDESS, B . , HIRST, P. Q. and HUSSAIN, A. (1977). Marx's Capital and Capitalism Today. Vol. 1. London: Routledge and Kegan Paul. DALLMAYR, F . (1982). 'The theory of structuration: a critique'. In Giddens A. (Ed.), Profiles and Critiques in Social Theory. London: Macmillan. DALTON, M . (1959). Men Who Manage. New York: Wiley. STUDYING MANAGERIAL WORK 269 Fox, A. (1973). 'Industrial relations: a social critique of pluralist ideology'. In Child, J. (Ed.), Man and Organization. London: Allen and Unwin. Fox, A. (1974). Beyond Contract. London: Faber. GALBRAITH, J. K. (1967). The New Industrial State. London: Hamish Hamilton. GANE, M . (1983). 'Anthony Giddens and the crisis in social theory'. Economy and Society. 12, 368-98. GIDDENS, A. (1976). New Rules of Sociological Method. London: Hutchinson. GIDDENS, A. (1979). Central Problems in Social Theory. London: Macmillan. GIDDENS, A. (1982). 'Power, the dialectic of control and class structuration'. In Giddens, A. and Mackenzie, G. (Eds.), Social Class and the Division of Labour. Cambridge University Press. GIDDENS, A. (1984). The Constitution of Society. Cambridge: Polity Press. GLOVER, I. (1977). 'Managerial work: a review of the evidence'. Social Science Research Council Report. GLOVER, I. eind MARTIN, G . (1986). 'Managerial Work - An empirical and cultural contradiction in terms.' Mimeo, Dundee College of Technology. GOFFMAN, E . (1959). Asylums. Harmondsworth: Penguin. GOLDING, D . (1979). 'Symbolism, sovereignty and domination in an industrial hierarchical organization'. Sociological Review, 27, 1, 169-77. GOLDING, D . (1980). 'Establishing blissful clarity in organizational life: managers'. Sociological Review, 28, 4, 763-82. GoLDTHORPE, J. (1982). 'On the service class, its formation and future'. In Giddens A. and Mackenzie, G. (Eds.), Social Class and the Division of Labour. Cambridge University Press. HABERMAS, J . (1972). Knowledge and Human Interests. London: Heinemann. HERMAN, E . S. (1982). Corporate Control, Corporate Power. Cambridge University Press. HINDESS, B . and HIRST, P. (1977). Mode of Production and Social Formation. London: Macmillan. KELLY, J. E. (1982). Scientific Management, Job Redesign and Work Performance. New York: Academic Press. KNIGHTS, D . and WILLMOTT, H . C . (1985). 'Power and identity in theory and practice'. Sociological Review, 33, 1, 22-46. KNIGHTS, D., and WILLMOTT, H . C . (Eds.). (1986). Managing the Labour Process. Aldershot: Gower. KNIGHTS, D . , WILLMOTT, H . C , and COLLINSON, D . (Eds.). (1985). JoA Redesign. Aldershot: Gower. KOTTER, J. P. (1979). Power in Management. New York: Amacon. KOTTER, J. P. (1982). The General Managers. New York: McGraw-Hill. LAYDER, D . (1981). Structure, Interaction and Social Theory. London: Routledge and Kegan Paul. LITTLER, R . C . and SALAMAN, G . (1982). 'Bravermania and beyond: recent theories of the labour process'. Sociology, 16, 2, 251-69. MINTZBERG, H . (1973). The Nature of Managerial Work. New York: Harper and Row. NICHOLS, T . (1969). Ownership, Control and Ideology. London: Allen and Unwin. NICHOLS, T . and BEYNON, H . (1977). Living with Capitalism. London: Routledge and Kegan Paul. PECHEUX, M . (1982). Language, Semantics and Ideology. London: Macmillan. PETTIGREW, A. (1973). The Politics of Organizational Decision-Making. London: Tavistock. PETTIGREW, A. (1975). The Awakening Giant. Oxford: Basil Blackwell. RANSON, S., HININGS, B . and GREENWOOD, R . (1980). 'The structuring of organiza- tional structures'. Administrative Science Quarterly, 25, 1-17. REED, M . I. (1984). 'Management as a social practice'. Joumal of Management Studies, 21, 3, 273-85. 270 HUGH WILLMOTT ROBERTS, J. and SGAPENS, R . (1985). 'Accounting systems and systems of accountability - understanding accounting practices in their organisational contexts'. Accounting, Organisations and Society, 10, 4, 443-56. SALAMAN, G . (1982). 'Managing the frontier of control'. In Giddens A. and Mackenzie, G. (Eds.), Social Class and the Division of Labour; Essays in Honour of Ilya Neustadt. Cambridge University Press. SAYLES, L . R . (1979). Leadership. New York: McGraw-Hill. SCOTT, J. (1979). Corporations, Classes and Capitalism. London: Hutchinson. SILVERMAN, D . and JONES, J. (1976). Organizational Work. London: Collier-Macmillan. SOFER, C . (1970). Men in Mid-Career: A Study of British Managers and Technical Specialists. Cambridge University Press. STANV^ORTH, P . and Giddens, A. (1974). Elites and Power in British Society. Cambridge University Press. STOREY, J. (1983). Managerial Prerogative and the Question of Control. London: Routledge and Kegan Paul. STOREY, J. (1985a). 'The means of management control'. Sociology, 9, 2, 193-211. STOREY, J. (1985b). 'Management control as a bridging concept'. Joumal of Management Studies, 22, 3, 269-91. THOMPSON, J. B. (1984). Studies in the Theory of Ideology. Cambridge: Polity. TINKER, A. M. (1984). 'Theories of the state and the state of accounting: economic reductionism and political voluntarism in accounting theory', yourna/ of Accounting and Public Policy, 3. 55-74. TURNER, B . A. (1971). Exploring the Industrial Subculture. London: Macmillan. WEBER, M . (1949). The Methodology of the Social Sciences. New York: Free Press. WILLMOTT, H . C . (1979). 'The structuring of organizational structure: a note'. Administrative Science Quarterly. 26, 470-4. WILLMOTT, H . C . (1984). 'Images and ideals of managerial work'. Joumal ofManagement Studies, 21, 3, 349-68. WILLMOTT, H . C . (1986). 'Unconscious sources of motivation in the theory of the subject; an exploration and critique of Giddens' dualistic models of action and personality'. Joumal for the Theory of Social Behaviour, 16, 1, 105-22. WiNKLER, J. T. (1974). 'The ghost at the bargaining table: directors and industrial relations'. British Joumal of Industrial Relations, 12, 191-212. WRONG, D . (1961). 'The over-socialised conception of man in modem sociolog/. American Sociological Review, 26, 2, 183-98. ZEITLIN, M . (1974). 'Corporate ownership and control: the large corporations and the capitalist class'. American Joumal of Sociology, 79, 1073-119.