The Social Foundations of the Bureaucratic Order

The Social Foundations of the Bureaucratic Order

Jannis Kallinikos

inadequacy woefully evident. Neither bureaucracy nor its asserted alternatives can properly be understood unless modernity and the social and organizational innovations it has brought about are sufficiently appreciated.

Despite, however, the oversimplified character of the discourse concerning the end of the bureaucratic age, it is crucial not to lose sight of the multifaceted economic, technological and social changes of which such a discourse is but one symptom. Such changes are associated with the diffusion of values, orientations or lifestyles that increasingly reflect the culture of individualism and the decline of legitimacy of public institutions (see, for example, Beck, 1992, 2000; Sennett, 1992). Coupled with globalization, the weakening of the nation state and the rapid growth of information and communication technologies, the culture of individualism keeps on redefining core characteristics of modernity. Though these developments remain far from being crystallized, there is indeed some evidence to suggest that current socio-cultural and technological changes converge to the redrawing of the prevailing boundaries of the private and public world, work from leisure, family and community.²

Under these conditions, the very forms by which individuals have traditionally been tied to social institutions are bound to change (Baumann, 1992; Beck, 1992; Sennett, 1992, 2000). New forms of individual involvement in organizations (e.g. flexible and temporary forms of employment) develop to accommodate the shifts in social institutions, values and lifestyles mentioned above. At the same time information and communication technologies become an important agent of organizational and occupational change (Rifkin, 1995, 2000; Castells, 1996, 2001). The precise ways, though, by which organization forms are connected to these developments remain a contested terrain. Neither the organizational forms nor the mechanisms that bring them about are sufficiently studied or clear. The appreciation of the impact of these changes on the prevailing forms of organization would seem to require the serious reconsideration of the social institution of bureaucracy in ways that step beyond stereotyped images of it.

Let it be clear that the critical attitude vis-`a-vis the claims that announce the demise of bureaucracy and the emergence of post-bureaucratic forms is not prompted by any dedicated commitment to the bureaucratic form of organization. The author neither condemns nor supports bureaucracy. Rethinking the foundations of bureaucracy serves the purpose of disclosing the social, cultural and economic roots of this institution and appreciating its limits against the background of a wider historical understanding. Undeniably, bureaucracy has entertained a number of tensions, most vividly shown in the contrast between the ideals of justice, meritocracy and egalitarianism by which it has been legitimized and the centralized and often awful practices with which it has been associated.

In the next section we take up the issue of rule-bound behaviour that is often seen as the epitome of bureaucracy. This shows that standardization is a ubiquitous aspect of contemporary organizations that contrasts sharply with the assumed malleability and flexibility of post-bureaucratic organizing, calling for the careful reconsideration of the bureaucratic form. The following section undertakes precisely that task. Bureaucracy is identified with the non-inclusive coupling of the individual to the organization consequent upon the separation of the role from the person. On the basis of this claim, the article ventures to dissociate bureaucracy from the dominant connotation of an inflexible and rigid form. It shows that the non-inclusive modulation of the individual–organization relationship makes bureaucracy the first and as yet sole organization form in which individuals are tied to organizations in selective, mobile and reversible terms. Such terms give the bureaucratic organization the possibility of addressing emerging contingencies. Then we attempt to show that the non-inclusive involvement of individuals into organizations is the primary matrix of relations out of which other derivative characteristics, often taken as the epitome of bureaucracy, such as standardization and centralization, emerge.

The Social Foundations of the Bureaucratic Order

Jannis Kallinikos

tions, duties and jurisdictions and regulating interaction patterns. Rule-bound behaviour has often been seen to be at the heart of the behavioural mechanics governing bureaucracy. The popular and artistic distaste for bureaucracy derives by and large from this alleged depersonalized functioning of formal organizations.

However, the historical background of the social relations out of which bureaucracy emerged suggests some important qualifications. Rule-bound behaviour was fashioned to accommodate the functioning of modern organizations whose operations could no longer be anchored in the normative certainty of Gemeinschaft. Being active on a spatial and temporal scale that extended far beyond the limited world of pre-modern communities, modern organizations needed both the legitimacy and new principles for controlling their operations. The construction of formal role systems for regulating human behaviour in organizations emerged as a historical innovation, essential in sustaining the egalitarian ethos of modern society. Formal role systems provide transparent motives and legible behaviour, on the basis of which one can decide whether citizens are treated equally and the employees of the organization protected from any abuse or arbitrary exercise of power (Perrow, 1986). By the same token, the bracketing of personal goals and the imposition of a formal order reflect the awareness that the pursuit of personal ends within the context of an organization may well undermine the goals and the adequate functioning of an organization (du Gay, 1994, 2000). The institutionalization of expectations and action patterns coinciding with rule-bound behaviour were thus an essential means for avoiding haphazard initiatives and opportunism, and improving the performance of public organizations.

Depersonalized behaviour is obviously not limited to public organizations. Although it may have been originally occasioned by major ideals of the emerging bourgeois society that were socially embedded through precisely the adequate functioning of state or public bureaucracies, the bureaucratic standardization of behaviour describes the entire organizational landscape of modernity. In one form or another, depersonalized conduct has been made an indispensable principle of all formal organizing in modernity. It has been applied with even greater severity in private firms and industrial organizations, where the design of jobs and the systems of surveillance tied to them emerged as the outcome of the relentless drive towards cost reduction, achieved through minute division of labour and economies of scale (Chandler, 1977). Rather than simply regulating behaviour, industrial organizations sought to regulate even the minutest details of workers’ bodily movements (Braverman, 1974; Noble, 1984). Standardization in that sense has come close to betraying the very ideals with which it has been associated. The inhuman, sometimes brutal, relations that prevailed in the industrial factories, in some cases as late as the outbreak of the Second World War, are indeed hard to reconcile with the ideals of liberty and justice.

Organization 11(1)

Articles

Managerialist thinking easily relegates the job-designing principles of industrial capitalism to the past, assuming that they are increasingly surpassed by important developments in the market environment and the technological infrastructure of production (Womack and Jones, 1996). It would be difficult indeed to deny the crucial organizational importance of major economic and technological innovations. Contemporary technology, in particular, has through extended automation changed significantly the conditions of production in industrial firms and other contexts of economic activity (Kallinikos, 2001). However, the status of these changes at the shop floor of the industrial factory and the outcome of these developments for the issue of behavioural mechanics that concerns us here are not entirely clear. There exist a number of studies to suggest a rather mixed and at times negative picture of recent trends, indicating greater control of employees, intensification of work (Zuboff, 1988; Kling, 1996) and novel forms of alienation (Harrison, 1994; Carnoy, 2000; Sennett, 2000; Murray et al., 2002).

Computer and information technologies do not make up an unequivocal ally to flexible patterns of behaviour, as the exponents of the post-bureaucratic or entrepreneurial form of organization often assume (Kallinikos, 1998). The expanding organizational involvement of software packages might indeed contribute to the standardization of behaviour. Software packages are currently major means for structuring and organizing work in contemporary organizations. Most of the time they are conceived and constructed by specialized software developing companies, in contexts other than those they are called upon to monitor. The construction of software packages by specialized organizations indicates that they are initially responses to abstract definitions of problems. Without this requirement the very development of programs would be impossible in the first place (Simon, 1969).

The increasing organizational involvement of de-contextualized systems of managing and monitoring information and the requirements of cross-contextual information comparability impose standardization in the form of a quasi-universal language of information processing. Undeniably, the introduction of software packages into organizations is a complex socio-technical process that involves deliberate and haphazard modifications. It is well known that software packages can be reconfigured to fit the demands of specific contexts or used in ways that were never thought of at the very moment of their inception (Ciborra, 2000). Yet the very substratum of software packages that reflects the logicist premises on which they are built and the cognitive mechanics of symbol processing cannot be undone (Lackoff, 1995) by reconfiguring or resisting the use and functionality of particular software packages. But even at the more macroscopic level of everyday work, software packages punctuate experience and impose uniformity through:

� The standardized definition of problems (or problem spaces).

� The specification of the tasks that the problem thus defined involves. � The delineation of the procedural steps necessary to enact the logic of

the software package (Zuboff, 1988; Kling, 1996).

Software packages are thus far from innocent means that support organizational operations. Rather, the three categories of effects stated above and the very cognitive foundations of software packages suggest that they hypostasize both meaning and action patterns. As perhaps any technology, software packages embody human experience, inviting particular modes of understanding and action that involve both the framing of the reality to be addressed, the determination of particular tasks and their sequential patterning. It is not by accident that information technology has now and then been considered as a mechanized contemporary alternative to the rule-bound behaviour of bureaucracy (e.g. Beniger, 1986).

Information systems undeniably promote standardization, even though homogeneity and uniformity may not be their sole effects. Current technology also enables a form of bricolage, exercised upon the huge number of information items and applications that constitute the landscape of most contemporary organizations (Ciborra, 2000). Software packages are in this way pregnant with various option possibilities that are produced by the play of permutations which the many information items underlying them can enter. Also, the development of links across organizations allows for the relatively easy crossing of institutional boundaries in ways that may promote alternative models of individual and organizational action (Castells, 2001). But these rather obvious effects should not hide the deep standardization of cognition and action, accomplished by the diffusion and organizational involvement of information systems. Standardization goes underneath the surface flexibility enabled by current technology. It involves uniform procedures for problem definition and problem solving, imposed by automated models of cognition, as described above. It even entails the standardization of interfaces, protocols and various hardware and software components (Hanseth, 2000). The ubiquitous character of standardization in contemporary life provides evidence in support of the claim that it cannot be made the exclusive arbiter of the bureaucratic form of organization.

The evidence that standardization represents a viable and widely diffused job and process design methodology is furthermore supported by the rapid and impressive diffusion of the standards of the ISO series during the last two decades. It is obvious that the ISO standards lead to rule-bound and standardized patterns of behaviour. But standardization represents an ideal even in models of management underlying the entrepreneurial or network form of organization, such as total quality management (TQM). Product quality in TQM models is by and large defined in terms of its capacity to conform to a set of pre-specified standards. Therefore, the careful design of these standards as well as the standardization of the process by which the product (or a service) is produced represent the fundamental requirements for achieving this goal. Indeed, TQM is a methodology of predictability, achieved through the minute description of various work tasks and processes and carried out in ways that demand active involvement of the employees of the organization throughout the production process (see, for example, Womack and Jones, 1996).

Standardization of behaviour seems furthermore to represent a viable job design methodology in types of organization that may be intuitively thought to represent a contrast to the popular image of rigid bureaucracies. Everyday observations suggest that organizations that are commonly assumed to be market-driven and customer targeting, such as airlines, banks or insurance companies, use standardized and largely stylized modes of conduct in abundance. Standardized behaviour not only continues to play a crucial role in many types of service organizations but some service, ‘consumer-responding’ organizations have extended the regulative domain of standardization, allowing it to shape such spontaneous behaviours such as smiling, wording, eye contact and other minute aspects of human conduct.

In sum, this brief review and re-contextualization of recent developments does not lend immediate support to the liberal neo-romanticism of individual creativity and initiative taking, following the decline of bureaucracy. Standardization is as ubiquitous as the very changes that may be taken to suggest the emergence of new forms of organization. There is a need, therefore, to turn to a more thorough and systematic study of bureaucracy. The current controversy between bureaucratic forms of organization and the asserted alternatives must be placed in a wider and coherent framework that may give it its appropriate meaning.

The Social Foundations of the Bureaucratic Order

Jannis Kallinikos

aggregate of persons but of roles and the patterns brought about by the interdependence of roles (Luhmann, 1995; Tsivacou, 1997). The inescapable corporeality of the human condition should not be taken to imply that individuals enter organizations in their full-blown cognitive, emotional and social complexity.

It is crucial to appreciate the distinctive status of relations out of which bureaucracy emerges as the modern form of organization. The role, not the person, constitutes the fundamental structural and behavioural element of modern formal organizing. Organizations are not made of individuals distributed over a complex landscape of job positions but of patterns built by those abstract operational requirements we call roles. Roles are enacted by the intrinsically modern capacity of contemporary humans to systematically and consistently suspend all other personal or organizational aspects that do not bear upon the role and to undertake action (e.g. machine operator, accountant, salesman, secretary) along delimited and well-specified paths. The living energy and the general communicative capacity of humans are essential resources for organizations (as for all social life) but this should not lead one to assume that formal organizations are made of individuals qua persons.

Bureaucracy thus introduces an abstract conception of work as a set of delimited behavioural choices (i.e. duties) that can be dissociated from the totality of the lifeworld and from the distinctive mode of being of every person. A major objective and an important consequence of the bureaucratic modulation of the individual–organization relationship is that individuals join the organization on the basis of considerations that relate to their ability to assume a role, that is, on the basis of merits provided by education, working experience, and so on. At the same time, other aspects of an individual’s life are severed from bureaucratic regulation. The non-inclusive involvement of individuals in organizations is sustained by the adequate differentiation of both individual and social life. That is, the characteristics that derive from education, professional specialization and working experience cover only a part (admittedly a very important one) of the totality of an individual’s roles and projects. In the social context of modernity, other organizations and institutions that are clearly and unambiguously differentiated from work organizations (for example, family and community) represent a crucial outlet for the individual’s interests and activities. The segmentation of life into separate and relatively independent spheres is an essential requirement for the forms of human involvement upon which bureaucracy is predicated (Kallinikos, 2003).

The bureaucratic form and the non-inclusive way of modulating the individual–organization relationship coincided with the gradual dissolution of class stratification and the immobile, hereditary social relations characteristic of the late feudal world (Luhmann, 1982, 1995, 1996; Gellner, 1983, 1996). Bureaucracy emerged as the dominant modern

Organization 11(1)

Articles

organization form out of the overall rational and functional preoccupations of modernity. However, the novel way of orchestrating the individual–organization relationship represented itself an important vehicle for constructing a new organizational form, premised on universalism and meritocracy as major principles for realizing and embedding the bourgeois ideals of individual liberty and justice (du Gay, 2000). By standardizing the requirements of role performance and formalizing the process of role taking, the bureaucratic organization became the vehicle through which jobs became potentially available to anyone that fulfilled the requirements of job specification. It is through the very separation of the role from the person that such an availability can be rendered possible and a labour contract be signed that makes the terms of the agreement legible and law enforceable (Weber, 1947, 1970, 1978).

Bureaucracy and modernity are therefore inextricably bound up with one another. Bureaucracy is the organization form of modernity. It is closely associated with the overall cultural orientations of modern man, the social mobility that coincided with the gradual dissolution of premodern stratification, and the burgeoning bourgeois ideals of individual freedom and justice, which it helped itself to embed. In this respect, bureaucracy contrasts sharply with pre-modern forms of organizing that relied by and large on the principle of inclusion for regulating the relationship of people to organizations. The differentiation of personal, social and working aspects of identity was elementary in the agricultural, feudal world. As a consequence these aspects of people’s life could not be separated from one another, as is the case in modernity (Gellner, 1983). In the pre-modern, segmentary societies, the social position—defined by a fixed and basically immobile social stratification regulated and reproduced as a rule by hereditary relations—determined by and large the working identity of people. The relatively open space of bourgeois democracy and the social mobility associated with it formed the basic conditions for the emergence of the organizational form that Weber came gradually to designate as bureaucracy.

The far-reaching significance of modulating the individual– organization relationship in terms other than inclusive emerges clearly in the background of the comparison of bureaucracy with the organizational form that Goffman (1961) once called total organizations, that is, mental hospitals, prisons, monasteries, army barracks, religious sects, and so on. In contrast to the non-inclusive coupling of the individual to the organization underlying bureaucracy, total organizations are based on the structural principle of inclusion. Individuals are contained in the organization, they are in other words ‘inmates’. Total organizations impose their austere order on the entire personality of their members. They do not distinguish between personality and collective. The term ‘individual’, as we know it, is alien to this form of organization. Total organizations thus provide an instructive contrast to bureaucracy. The latter may owe part of its administrative practices to the meticulous discipline worked out in the austere world of monasteries (Mumford, 1934), yet an abysmal chasm separates the bureaucratic organization from monasteries or other total organizations. Such a chasm is precisely produced by the non-inclusive, conspicuously modern and abstract mode of regulating the individual–organization relationship.

The profound differences separating inclusive and non-inclusive modes of regulating the individual–organization relationship are perhaps obscured by the impressive diffusion of bureaucratic principles and the withdrawal of total organizations to the fringes of modern everyday social encounters. Being the ‘normal’ yardstick of functional ability and institutional legitimacy, bureaucracy and the non-inclusive modulation of the individual–organization relationship tend to be taken for granted. Strange as it may seem at first glance, Foucault’s (1977) highly innovative use of panopticon (and the prison) may be taken to suggest total organizations as the model for understanding the modern organizational and institutional order. As indicated by the relevance which monastic life may have assumed for bureaucracy, there exist undeniable affinities in some of the forms and mechanisms by which these two widely differing archetypes of organization assure compliance and construct conformity. However, despite these affinities, bureaucracy and total organizations differ in some substantial ways. Modern discipline in institutional life presupposes the anthropological distinction of the role from the person and the structural device that embeds such a distinction. Without such a separation, the objectification of one’s contributions and the self-monitoring along measurable or governable dimensions would be impossible or, at least, hampered substantially. A careful reading of Foucault re-enhances this claim (Foucault, 1977, 1980, 1988). The tangle of behaviours, orientations and techniques that constitute humans qua persons must be dissolved to become the target of measurement, examination and control (Kallinikos, 1996; Hasselbladh and Kallinikos, 2000).

Organization 11(1)

Articles

through the detailed design of roles and the rules tied to their performance. In this light, rule-bound behaviour is inter alia motivated by the project of adapting to contingent demands, rather than the other way around, which the conventional understanding of bureaucracy as simply routine seems to suggest. To what extent organizational and social roles can be severely decoupled from the totality that makes the distinctive mode of being of every human (Tsivacou, 1997) remains of course a highly delicate issue. However, both bureaucracy and modern society were built on the premise that such a severe or adequate separation is possible (Mangham, 1995; Sennett, 2000).

The cardinal bureaucratic premise of non-inclusiveness is bound to produce a relationship of the individual to the organization that is marked by selectivity, mobility and reversibility (Luhmann, 1982, 1995; Gellner, 1983, 1996). Selectivity is the outcome of the fundamental assumption that individuals taking organizational roles are expected to suspend non-role demands and act on the basis of a well-specified and delimited set of criteria that constitute the role (job description and specification, duties and jurisdictions, field of responsibility). The mobility in the individual–organization relationship is produced by the fact that a role, being an abstract set of functional requirements, can be unleashed from the particular circumstances into which it is embedded, and be transferred across various organizational contexts (Hassebladh and Kallinikos, 2000). Mobility is furthermore enhanced by the reversible terms of individuals’ involvement in organizations. Reversibility implies that jobs can be altered or redesigned and the organizational sanctioning of job positions modified or even withdrawn. The relationship is also reversible from the point of view of the individual who can invoke several reasons for quitting an organization, even though this may bring negative pecuniary or legal consequences.

Against the background of these observations, bureaucracy emerges as an altogether different institution from the conventional image that identifies bureaucratic organizational forms with a sort of behavioural mechanics and incapacity to change. Indeed, the demands of the current age for contingent (local and functionally adaptable), mobile and reversible (temporary) patterns of behaviour cannot but be satisfied by an organization form that strengthens the bureaucratic premises whereby individuals are tied to the organization in terms other than inclusive. Shorter time frames for organizational action, employment forms other than those implied by lifetime contracts, diffuse tasks that demand the constant redesigning of roles, virtual relations, all presuppose that individuals are coupled to organizations in terms that are characterized by selectivity, mobility and reversibility.

As construed here the modulation of the individual–organization relationship in selective, mobile and reversible terms represents the cornerstone of bureaucracy. However, while rule-bound behaviour and role specialization can be directly associated with the selective, mobile and reversible terms by which individuals are coupled to organizations, it is not entirely clear how the hierarchical constitution of bureaucracy that represents one of its crucial attributes can be related to these terms. Hierarchy, though, is a pervasive trans-cultural and trans-historical characteristic that cannot be exclusively attributed to bureaucracy (Dumont, 1967). What is perhaps distinctive in the hierarchical configuration of bureaucratic organization is the relatively clear and rule-bound regulation of the exercise of authority. The jurisdictional domain of authority is specified and delimited while its exercise is governed by rules to constitute what Weber (1947) referred to as rational-legal authority. The delimited and rule-bound regulation of authority is precisely the outcome of the selective, mobile and reversible terms of modulating the individual–organization relationship. Authority is tied to the office (i.e. the role), exercised upon a specific domain of organizational operations while its organizational sanctioning can be withdrawn at every moment.

The understanding of bureaucracy in the terms described so far represents only a very brief description of the complex social and cultural processes underlying it. The interpretation of bureaucracy solely in terms by which individuals are tied to organization leaves out of consideration vital social and cultural processes that sustain the bureaucratic form of organization and other modern institutions. It has already been noted that a number of factors contributed to the emergence and diffusion of the bureaucratic organization and the ability to modulate the individual– organization relationship in a non-inclusive fashion. Crucial among them were the decoupling of the social status of individuals from their position in the system of production, the consequent social mobility brought about by bourgeois democracy and the relatively clear separation of private life from the public world. Closely related to these changes were the development of capitalism and the market economy. Other crucial factors involved the very anthropological foundations of individualism and the subtle cultural shifts by means of which the world was increasingly understood as the very object of human intentions, and calculability and rationality were rendered the universal mode of cognition (Heidegger, 1977). The modulation, therefore, of the individual– organization relationship in selective, mobile and reversible terms evolves within the wider context of a social and cultural universe that we here cannot but take largely for granted.

Organization 11(1)

Articles

forms of employment other than those implied by stable, clear-cut job arrangements and lifetime contracts, and flexible production systems can all be invoked as major cases that break with the classical bureaucratic principles of organization.³ Do not these examples really suggest bureaucracy to be on the verge of concluding its historical circle?

An important corollary that emerges from the preceding analysis is that standardized behaviour represents a ubiquitous element of all modern organizing that seems unlikely to be suspended by the trends subsumed under such catchwords as network, virtual or entrepreneurial forms of organization. Standardization is essential to all non-haphazard human action and communication that transcend the limits of particular contexts (Hasselbladh and Kallinikos, 2000). As a consequence, organization forms that claim some sort of context independence and universality cannot but exhibit at least a minimum of standardization. A second and perhaps more crucial insight following the analysis undertaken in the preceding pages concerns the relation of standardization to flexibility and the dissociation of bureaucracy from standardization. The bureaucratic form of organization cannot be simply equated with routines and standard operating procedures, no matter how important or ubiquitous they happen to be. Standardized behaviour constitutes the substratum, as it were, of which organizational action is made. Routines, procedures and roles are the elements of an organizational ars combinatoria capable of constantly reassembling these elements to address the ceaseless emergence of contingencies, underlying the contemporary condition.

The far-reaching consequences of this modular, as it were, constitution of bureaucracy emerge against the background of the metaphorical associations provided by standardized notation such as alphabetic writing or music notation (Kallinikos, 1996). The ability to bring together the characters of a notational system in various combinations is ultimately dependent on the standardization of every character and its clear distinction from other characters. Unless the standardization and finite differentiation of the notational elements of a system are accomplished, no rules can be worked out to govern their combinations (Goodman, 1976). A notational system that is composed of elements that dissolve into their substratum or melt constantly into one another eludes semiosis, and betrays the very purpose that it is supposed to serve (Kallinikos, 1996). Standardization forms a fundamental condition for the huge number of combinations (certainly under a number of constraints) that the standardized elements of a system can enter. Looked upon this way, standardization is a prerequisite for semiotic flexibility. Any system that is based on similar principles of architecture is bound to exhibit behavioural or semiotic flexibility.

However, to avoid the conspicuous logicist associations that could arise from the comparison of organizational action with symbolic notation (Simon, 1969), we could pursue the analogy of bureaucracy with one of the most complex and resilient institutions of social life, namely,

The Social Foundations of the Bureaucratic Order

Jannis Kallinikos

natural language. The picture of bureaucracy painted in this article suggests that the bureaucratic form of organization could be conceived as being governed by what structuralists once called the principle of double articulation of language.⁴ Like language (phonemes), bureaucracy is composed, at a first level, by a fairly standardized number of elements

(e.g. routines, tasks, procedures, roles) that, seen separately from oneanother, may convey the impression of a fixed and limited population of behaviours, devoid of meaning or signification. This is precisely produced by the ability to bring these elements together to various larger patterns (i.e. the second level) that acquire meaning and purpose due to their capacity to respond to emerging social projects. The abstract character of the organizational architecture that coincides with the double articulation of bureaucracy contrasts sharply with the readily observable reality of routines and standardized procedures and may perhaps be held responsible for the widely diffused association of bureaucracy with standardization and rigidification.

The many configurations that can be produced by a system organized in a way resembling the double articulation of natural language provide an indication of the enormous potential such a system disposes for adapting to changing conditions. Through reshuffling, recombination and occasional redesign of its elements the system becomes capable of addressing emerging situations that may demand responses different from those the system has produced so far. Despite the many constraints that may be thought to underlie the combinability of the elements (Kallinikos, 1996, 1998) a huge number of combinations are indeed available that may give the organization a broad repertoire of structural and behavioural options. The quest for flexibility cannot therefore be addressed in a genuine fashion unless contemporary forms of organization are built on the very foundations of bureaucracy as outlined in this text. Only by being able to reassemble the standardized elements by which it is made of, to produce new configurations (i.e. products, services, action patterns, and so on), can an organization hope to cope with the constant springing up of contingencies. A prerequisite for doing so is to allow individuals to be tied to systems in selective, mobile and reversible forms.

The organizational involvement of humans qua roles makes bureaucracy capable of unleashing the process of organizing from the inescapable context-embeddedness of the human body and the intractable complexity of experiences that underlie human beings qua persons. In so doing, it sets the premises for a form of organization free from the restricted mobility of the human body and the languid process of personal and psychological reorientation. Bureaucracy’s abstract principles of organization are the hull out of which virtual relations are bound to emerge. Virtual organization is indeed contained as a germ in the very separation of the role from the person, and the design and enactment of action patterns that are only loosely coupled to the corporeality and

Organization 11(1)

Articles

psychological complexity of humans (Baudrillard, 1983, 1988). The social and psychological conditions necessary for the emergence of bureaucracy give an indication of the profound institutional change necessary to build formal organizations on premises other than those implied by bureaucracy.

The claims advanced so far do not necessarily deny some perhaps crucial differences that seem to be underlying the institutional context (late modernity) in which many organizations currently operate. We mentioned in the introduction several factors (e.g. frequent economic and technological change, individualism and globalization) that distinguish the current environment into which most organizations operate from the asserted stability of bureaucratic contexts. It is a widespread assumption that these factors drive the current shift away from the bureaucratic form of organizations. However, such an assumption makes sense only if bureaucracy is identified with a rigid and inflexible organization form. Placed against the background of the preceding analysis, such an interpretation of the organizational order of modernity emerges as regrettably inadequate. The fast pace and unpredictability of social, economic and technological change indeed make necessary the strengthening or elaboration (rather than the abandonment) of the social and organizational innovations introduced by modernity and bureaucracy. The mobile, selective and reversible relations such innovations promote constitute the sine qua non, the virtual matrix, as it were, out of which concrete instantiations addressing specific sets of circumstances emerge.

There will certainly be occasions of radical environmental change that may require the reinvention and redefinition of the primary organizational substratum of routines, standard operating procedures and roles rather than their mere recombination. New technologies produce a different segmentation of organizational tasks. This possibility is undeniably present. The comparison of language with forms of organization is, after all, metaphorical. However, in the social realm of organizations, the de novo invention of the entire population of jobs, positions and procedures is a rather rare case and is not supported by empirical facts. Most of the time, novel situations can be managed through a complementary elaboration (elimination or addition of new roles, jobs and procedures) of a basic population of primary elements and the reassembling of these into new configurations. At other times, the comprehensive remaking of the first organizational substratum itself may not have such far-reaching implications as may seem at first glance. The remaking of organizational tasks, roles and procedures in such cases is based on a very similar logic for segmenting and structuring the totality of organizational work. Flexible manufacturing represents a case in point, producing its allegedly customized products through variable standardization of a basic population of components reassembled into final products or services. Contemporary architecture too nowadays produces variable design forms by exploring the possibilities of recombining a population of standardized building components (Kallinikos, 2001).

On the other hand, the shorter time frames of change and the crisscrossing of the organizational horizon by contingencies of various sorts undeniably threaten to shatter the modular constitution of bureaucracy into the very pieces from which it is composed. But it is of utmost importance to uphold that the centrifugal forces produced by incessant change arise precisely out of bureaucracy’s capacity to respond to the emerging contingencies and not by its inability to act in novel ways.

Organization 11(1)

Articles

(Simon, 1969; Galbraith, 1973; Weick, 1979) and not infrequently to the idiosyncratic patterns of behaviour March and Olsen (1976) summarized with the suggestive term ‘organized anarchies’. It may not have been by accident that none of these authors ever claimed to have discovered the foundations of a new organizational order. Furthermore, (de)centraliza-tion takes various forms. Decentralized initiative-taking in certain domains may well go hand in hand with centralization in other domains. For instance, operational decisions may well be decentralized while crucial decisions are subjected to increasingly tight control and regulation (Power, 2002). The net outcome of similar patterns are hard to predict and it is even harder to weigh their relative significance in empirical contexts.

The fact that formal organizations exhibit a variable centralization suggests that the issue of the distribution of the loci of decisions may be difficult to make the sole criterion for the demise of bureaucracy and the assumed emergence of alternative forms of organization. Neither the issue of centralization/distribution of authority nor that of standardization can be made the yardstick for deciding the limits of the bureaucratic form. No matter how important, these characteristics develop within the constitutive framework of relations established by the non-inclusive forms of human involvement and the selective, mobile and reversible terms by which individuals are tied to organizations. These terms form the very foundation of the bureaucratic organization. They constitute the primary relation out of which other secondary or derivative characteristics emerge. It is important to uphold the distinction between primary and derivative characteristics. Routines and standardized behaviour and centralization are undeniably important yet derivative characteristics of the bureaucratic organization. They emerge on the very foundation, established by the clear separation of the individual from the organization, and the selective, mobile and reversible terms by which individuals are tied to organizations.

The cardinal importance of the modular, non-inclusive mode for regulating the individual–organization relationship comes to the fore against the background of another fundamental relation. That is, the non-inclusive mode represents the outcome of a choice that involves a binary alternation (Kallinikos, 1988; Luhmann, 1995; Tsivacou 1997). You have either an inclusive or a non-inclusive mode of regulating the individual– organization relationship. It is impossible or, perhaps more correctly, inconsequential to have little of the former and little of the latter. It may be possible to have different modes regulating separate aspects of an organization but each aspect can be regulated by only one mode. We referred above to total organizations that regulate the life of inmates through an inclusive, non-modular relationship. However, administration in, say, prisons or mental hospitals is organized according to the non-inclusive, modular relationships and administrative staff occupy and enact distinctive roles. The selection of either of the two alternatives introduces a code that brings into being a very distinctive mode of organization. By contrast to this binary choice, characteristics like standardization and centralization exhibit a graded intensity. It is always possible to have more or less standardization or centralization. In this sense neither of these characteristics can become the ultimate arbiter for deciding whether an organizational form is bureaucratic or not.

These claims therefore suggest that it is necessary to make a distinction between constitutive and variable characteristics. The non-inclusive mode of human involvement in organizations provides the fundamental condition on the basis of which it is possible to decide whether a form of organization is bureaucratic or not. The significance of variable characteristics, by contrast, develops within the framework established by the constitutive function of the non-inclusive mode of human involvement in organizations. Variable characteristics precisely allow for distinguishing the contextual variation of the basic bureaucratic form.