Volume 12(3): 437–456 ISSN 1350–5084 Copyright © 2005 SAGE (London, Thousand Oaks, CA and New Delhi)

Epistemic Objects, Artefacts and Organizational Change articles

Reijo Miettinen and Jaakko Virkkunen University of Helsinki, Finland

Abstract. One of the key concepts of the neo-institutional studies of organizations has been routine—an established, rule-governed pattern of action. The concept of routine creates difficulties when used for making sense of the emergence of new practices or change in organizations and institutions. There are two reasons for this. First, routine was introduced originally to account for the continuity of organizational life. Second, it is based on theories of action and behaviour that focus exclusively on the pre-reflective and embodied aspects of human practice. This paper seeks an alternative approach by using the concepts of epistemic object and artefact mediation of human activity. It argues that representational artefacts, such as concepts and models, are instrumental in inducing change in human practices. Using the work of occupational health and safety inspectors as an example, it is shown how a practice or set of routines is made into an object of enquiry in order to generate a working hypothesis for an alternative practice. The hypothesis is further objectified by designing a set of informational tools and procedures that carry on the new practice. Key words. epistemic object; organizational change; representational artefacts; routine

Neo-institutional studies of organization have primarily focused on the problems of continuity, rationalization and structural isomorphism of organizations in organizational fields (Powell and DiMaggio, 1983; Scott, 1995). In these studies, a key concept has been routine, a stabilized way of acting. The notion of institutional contradiction, on the other hand, has been suggested to account for the sources of institutional change DOI: 10.1177/1350508405051279

http://org.sagepub.com

Organization 12(3) Articles (Powell and DiMaggio, 1991). However, the relationship between these two concepts has, thus far, not been clarified (Clemans and Cook, 1999). In this paper we suggest that the cognitive and microsociological foundations of the concepts ‘routine’ and ‘rule-governed action’ do not allow for making sense of the emergence of new practices or change in organizations. One of the pioneers of the new constructivist sociology of knowledge, Karin Knorr-Cetina (1997), suggests that, in today’s knowledge society, the construction of epistemic objects or objects of enquiry (such as a molecule, a production system, a disease or a social problem) is becoming an ever more important part of any expert work. These objects are not things with fixed qualities but rather are open-ended projections oriented to something that does not yet exist, or to what we do not yet know for sure. For this reason, they are also generators of new conceptions and solutions and can be regarded as a central source of innovation and reorientation in societal practices. The concept of an epistemic object was introduced in the history and philosophy of science by Hans-Jorg ¨ Rheinberger (1997). According to him, a knowledge object is synonymous with a research object or an epistemic thing (1997: 28). These objects embody what one does not yet know. At the time of their inception, they are still relatively undefined. An epistemic object can be understood only as a part of historically evolved experimental systems or practices. Rheinberger makes a distinction between technical objects and epistemic objects. The former are well defined and black boxed, and are more or less permanent and repeatable parts of an experimental practice. They, however, determine the realm of possible representations of an open-ended epistemic thing. Rheinberger states that the difference between an epistemic object and a technical object is functional in nature. The status of an entity ‘depends on the place or “node” it occupies in the experimental system’ (1997: 30). Experimentation, as an engine for making the future, has to engender unexpected events. However, the significance of these events will ultimately be derived from their potential to become integral parts of future technical conditions. Although the concept of epistemic object was developed in the context of studying experimental natural science, we suggest that it supplies an insightful vehicle for analysing how a practice (including its technologies and rules), or critical aspects of a practice, can be made into an object of enquiry in order to produce novel and alternative ways of acting. To develop the point, we will critically discuss the theoretical foundations and limitations of the concept ‘organizational routine’. We will then introduce the concepts of objectification and artefact mediation as keys for understanding organizational change, drawing conceptual resources from activity theory, science and technology studies, and the historical epistemology of Marx Wartofsky. We will elaborate our approach by analysing how the contradictions and developmental possibilities of a

438

Epistemic Objects, Artefacts and Organizational Change Reijo Miettinen and Jaakko Virkkunen collective practice—the activity of occupational health and safety inspection in Finland—were made into an epistemic object, and how this led to both a new kind of inspection practice and the creation and implementation of a new set of tools necessary for carrying it out.

Habits and Routines as Carriers of Organizational Knowledge and Continuity Whereas the traditional epistemology defined knowledge in terms of symbolically represented, declarative knowledge (theories, models, concepts, facts), behavioural theories of organization define competencies in terms of established ways of action using the concept of routine (Winter, 1986). This concept has its own history. In the following, we will deal with the three successive generations of the idea of routine as the carrier of organizational knowledge and tradition. In his landmark work Administrative Behavior (first published in 1949), Herbert Simon deals with two key concepts to explain organizational decision-making: ‘memory’ and ‘habit’. He takes the latter directly from John Dewey and William James (Simon, 1976: 88) and he also refers to Edwin Steen’s early suggestion (1940) to use the term ‘organizational routine’ to refer to the organizational counterpart of the ‘habit’ (Simon, 1976: 88). Psychologically, a habit has the function of narrowing the attention to certain aspects of the environment and it is, therefore, the central mechanism of behaviour persistence. In addition, organizations and institutions supply such standardized ‘attention-directors’ and by so doing ‘largely determine the mental sets of the participants’. As one of the founders of the new cognitive science, Simon suggested that human thought and decision-making can be analysed using an analogue of a computer program (Newell et al., 1972). Decision-making is more about following rules than about calculating outcomes. In their book An Evolutionary Theory of Economic Change (1982), Richard Nelson and Sidney Winter present the second-generation concept of routine.1 The book deals with ‘the capabilities and behavior of business firms operating in a market environment’ (1982: 3). The authors conceptualize capabilities in terms of routines, that is, ‘regular and predictable behavior patterns’ (1982: 14). In their theory, routines play the same role as genes play in the biological theory of evolution. Like Simon, Nelson and Winter suggest that there is an analogy between individual skills and organizational routines. That is why they start their theorizing by studying the nature of skills. In their view, a skill is like a computer program in that it involves a sequence of steps. It also involves the making of numerous ‘choices’, often automatically. The authors adopt two classical concepts from cognitive psychology, namely ‘plan’ as defined by Miller et al. (1960) and ‘script’, introduced by Schrank and Abelson (1977) as a means of defining individual knowledge in nonsymbolic, operative terms. The knowledge that serves as the foundation

439

Organization 12(3) Articles of skilful performance is in large measure tacit, ‘in the sense that the performer is not fully aware of the details of the performance and finds it difficult or impossible to articulate a full account of those details’ (1982: 73). The authors take this idea directly from Polanyi (1958). Routines are the skills of an organization. The routinization of activity in an organization constitutes the most important form of storage of the organization’s operative knowledge: ‘Basically, we claim that organizations remember by doing’ (Nelson and Winter, 1982: 99). Skills and routines are maintained by being exercised. Nelson and Winter include artefacts in the concept of routine (1982: 105): ‘They include forms of external memory—files, message boards, manuals, computer memories and magnetic tapes—that complement and support individual memories but that are maintained in a large part as a routine organizational function.’ The third generation of the concept of routine combines cognitive psychology, ethnomethodology and phenomenology. Powell and DiMaggio maintain that together they supply a ‘new theory of practical action’ that constitutes a microsociological basis for the new institutionalism (1991: 22). According to them, Harold Garfinkel’s ethnomethodology supplies new theoretical resources for understanding ‘the routine, taken-for-granted aspects of behavior (practical activity)’ (1991: 22). Ethnomethodology studied the preconscious ‘practical reason’ (knowledge without concepts) governed by rules that are recognized only when they are breached. No pre-established normative or cognitive consensus is needed, because of trust and the ‘willingness of the participants to use normalizing techniques to sustain encounters’ (1991: 22). Ethnomethodology maintains that people do not have an end in mind before action, but they rather justify their actions after the fact by referring to culturally available legitimating ‘accounts’. Powell and DiMaggio also refer to Anthony Giddens’ concept of structuration and to the concept of habitus by Pierre Bourdieu as a potential means of further developing the ‘new theory of practice’. To them, the internal schemes and scripts of cognitive psychology of the 1970s and the ethnomethodological concept of rulegoverned action seem to constitute the two complementary foundations of routines as an explanation of the continuity of organizational behaviour.2 Routines are maintained both by pre-reflective consent by individuals and by the control systems and legitimation set up by organizations and institutions. But how are they renewed? Nelson and Winter (1982: 129) refer to the facts that the puzzles and anomalies of the prevailing practices may raise questions that contribute to problem-solving and to innovations. They adopt Schumpeter’s (1934) idea that an innovation consists, to a substantial extent, of a recombination of previously existing routines. Neo-institutional theory has further discussed the problem of change using the notions of institutional contradiction (Powell and DiMaggio, 1991), disruptive event (Hoffman, 1999) or jolt (Meyer et al.,

440

Epistemic Objects, Artefacts and Organizational Change Reijo Miettinen and Jaakko Virkkunen 1990; Greenwood et al., 2002), and horizontal interaction between heterogeneous actors within and between institutional fields (Powell et al., 1996; Clemans and Cook, 1999; Morril, 2005). The proponents of the institutional theory have been particularly interested in the mechanisms by which new, locally created practices, or innovations, become generalized and institutionalized into a new organizational field (Powell and DiMaggio, 1983; Morril, 2005). Neo-institutionalist studies have, however, not dealt much with the very onset of the change process, e.g. how the transformation of ways of acting takes place locally. It seems to us that it is here that the limits of its theory of practical action are confronted. The concept of routine (script, rule) was adopted to highlight the ‘essential continuity’ of the behavioural patterns to the point that one can speak about the ‘theory of persistence of the routine’ (Winter, 1986: 167).3 This is why the local emergence of new routines has received less attention. Ethnomethodologists are also more interested in the problem of social order and rule-following than in change. For them, attempts to reflect consciously and discursively on forms of activity are mostly about legitimating actions that already by themselves exhibit orderly structure and ‘rationality’ (Lynch, 1992). Phenomenologically oriented theorists of action have been primarily interested in the ‘prereflective intentionality’ embedded in everyday actions (Emirbayer and Mische, 1998). These actions are creative in the sense that they are always transformed to conform to the demands of a specific situation. In our estimation, however, in order to supply a satisfactory account of how historically new forms of practice emerge and stabilize, the study of what is distinctive about reflection is needed. We largely agree with the critics of this approach who suggest that it neglects the reproductive functions of habits and actions (Camic, 1998) and neglects the study of conscious attempts by human agents to understand and transform their routines (McGowan, 1998). The concept of script- or rule-based routine tends to deny any role whatsoever to the thought, future-oriented imagination and anticipatory moral reasoning of the subjects in activity, thus giving an unsatisfactory account of human agency in the change of practices.

Artefacts for Reflection and Organizational Change Many scholars in the fields of history of technology (Hughes, 1978), psychology (Wertheimer, 1959; Vygotsky, 1979), philosophy (Dewey, 1938/1991; Wartofsky, 1987) and sociology (Emirbayer and Mische, 1998; Crossley, 2001) suggest that reflection and the generation of alternatives to present routines are phenomena that merit being studied as special forms of action. Reflection and invention become necessary when the normal course of conduct is interrupted, or when a crisis, a functional failure or a critical problem is faced. Dewey’s theory of reflective thought and action is an early formulation of this idea (Dewey, 1938/1991; Miettinen, 2000). The formation of habits is a central mechanism of

441

Organization 12(3) Articles learning and transmission of know-how and cultural tradition. However, habits also carry the risk of becoming conservative, rigid and resistant to change. In a changing world, habits often do not work or confront serious difficulties (which might be called institutional contradictions) or shocks. In such a situation, according to Dewey, conscious reflection on the conditions of activity is needed. Such reflection leads to the formation of a working hypothesis for the reconstruction of the situation, which then will be tested in practice. It is this reflection and enquiry that make a habit intelligent. The scholars mentioned above seem to share three interests: the idea of the methodological significance of history and temporality in the study of action; the inclusion of thinking subjects and agency in the analysis (in terms of imagining future trajectories of action, moral reasoning, hypothesis formation); and an interest in the means, conditions and instruments of change. Crossley formulates the second point as follows (2001, 116): ‘There is something more to agency than the concept of habit can fully capture: The creative, generative dynamic which makes and modifies habits.’ The last point is put by Emirbayer and Mische (1998, 1006) as follows: ‘What kinds of context provoke or facilitate them toward gaining imaginative distance to habitual schematic responses and therefore reformulate past patterns through the projection of alternative future trajectories?’ These scholars distinguish between primary (or first-order, routine) actions and the reflective actions necessary for the transformation of practices (e.g. Dewey, 1925/1988; Vygotsky, 1979). This distinction does not imply a return to an intentional or cognitive conception of action. Models and concepts do not precede action nor do they legitimate actions only after the fact. They serve as means of making visible already developed possibilities, contradictions and emergent alternative solutions of a practice. They also aid their further development (Fujimoto, 2002) and, later, their stabilization and transfer to other places, as suggested by Strang and Meyer (1993).4 Another basic criticism of the concept of routine is that the reference to the internal processes or structures of the mind or to the unarticulated, taken-for-granted social rules is an insufficient account of continuity, let alone organizational change. Hasselbladh and Kallinikos (2000: 702) suggest that ‘the bureaucratic form of organizing is not a disembodied state that exists only in the level of intersubjective agreements in local contexts’. It is necessary to study the objectified, embodied and material forms of organizing and acting. Bruno Latour (1991) maintains that we should look for the foundations of social order and continuity not in the ‘abstract’ forms of sociality (norms, values, roles, shared meanings), but rather in enduring material objects, such as buildings, machines, traffic systems, laws, library collections, systems of classification, psychological tests and art works. Although some types of material artefact (such as classification systems, written rules) are recognized as parts or elements

442

Epistemic Objects, Artefacts and Organizational Change Reijo Miettinen and Jaakko Virkkunen of routines and habits in organizational theory, they do not have a recognized methodological role in the analysis of the preservation, transmission or change of routines or ways of action. Several theoretical approaches within philosophy, the social sciences and psychology suggest that norms of action and cognition are objectified into artefacts and that they decisively influence the course of action. Deweyan pragmatism (Dewey, 1938/1991), cultural historical activity theory (Vygotsky, 1979; Lektorsky, 1980) and actor network theory (Latour, 1991, 1992), as well as Wartofsky’s historical epistemology (1979, 1987), belong to these approaches. A tool is also a mode of language. For it says something, to those who understand it, about the operations of use and their consequences. . . . In the present cultural setting, these objects are so intimately bound up with intentions, occupations and purposes that they have an eloquent voice. (Dewey, 1938/1991: 52) The instrumental man-made objects function as objective forms of expression of cognitive norms, standards and object-hypotheses existing outside the individual. (Lektorsky, 1980: 137) We have been able to delegate to nonhumans not only force as we have known it for centuries but also values, duties, and ethics. (Latour, 1992: 232)

The ideas of script, plan and rule-governed interaction imply that forms of action are coded in the brains, nervous systems and bodies of individuals (Taylor, 1995). We think that the social forms of using tools, signs and instruments for human purposes are embodied in both individuals and artefacts. The human form of sociality is objectified in the use of shared artefacts. Consequently, a more elaborated theory of the role of artefacts in human activity is necessary in order to understand the underlying dynamics of organizational continuity and change. The founder of cultural-historical activity theory, Lev Vygotsky, formulated the concept of mediated action in the 1920s (Vygotsky, 1979). The interaction between the human agent and its object is mediated by cultural means. The basic types of means are tools and signs. An individual internalizes these means during socialization by participating in common activities with other humans. Consequently, consciousness does not exist as situated inside the head of the individual, but is rooted in the constant interaction between individuals and the world of objectified cultural artefacts. Accordingly, the activity-theoretical approach regards retooling, the shared creation of artefacts used as means of reflecting and practical transformation of activity, as a key to changing practices (Vygotsky, 1986). Subsequently, activity theorists focused on collective activities characterized by a division of labour (Leont’ev, 1978; Engestrom, ¨ 1987, 1995). The elements of a collective activity or an activity system are the subjects

443

Organization 12(3) Articles and the object of activity, the means as well as the division of labour and the rules of the community of actors involved. An object of activity is both objective (the raw materials and problems at hand) and projective (oriented toward the creation of a use-value) and is realized in the construction of products and services that constitute the outcome of the activity. It thereby also constitutes the societal motive of an activity. A collective activity is carried out by individuals’ actions, which in turn are composed of automatized, often tool-specific operations that are the means of action. Leont’ev (1978: 165) thinks that operations function like modules of action. Individuals reach various objectives and carry out various kinds of actions by combining these operations. He refers to psychological experiments showing that tacitly learned automatized operations and skills tend to remain more inert than consciously learned operations that are more susceptible to transformations, because they can be more easily made into objects of reflection. We find at least two merits in the concept of an activity system compared with the notion of routine. First, it highlights the interaction between actors and the object of their joined activity, thus introducing a dynamic element needed to understand changes in the ‘routines’. Second, it highlights the constant interaction and tension between the ‘general’, cross-situational nature of artefacts and local, situated actions. The analysis of this relationship is also vital for the understanding of change in actions. Marx Wartofsky is one of the few modern philosophers who has focused attention on the significance of artefacts in practice. According to Wartofsky, a practice is a coherent set of human actions (1987: 364) ‘characterized by a commonly understood object, or good—i.e. telos; and by a socially structured and commonly accepted repertoire of means, whether instruments or skills . . . as well as commonly understood norms’. These include explicit rules, exemplary models of procedures and modes of practical judgement. Moreover, Wartofsky makes a distinction between primary and secondary artefacts (1979: 201). Tools and related bodily skills directly used in production are primary artefacts. Secondary artefacts are ‘distinctive artefacts created for the purpose of preserving and transmitting skills, in the production and use of primary artefacts’. These artefacts are not representations of nature out there but representations of modes of action, that is, of the artefact-mediated, historically conditioned interaction of human beings with their environment (1987: 369). For a long time over the course of human history, the practice-relevant generalizations applied in productive practices were probably objectified solely in primary artefacts. For instance, Turnbull (2000) states that even the great Gothic cathedrals of the 12th and 13th centuries were planned without drawings and built using only templates, plumb-lines and other simple geometric tools to carry out the necessary measurement and

444

Epistemic Objects, Artefacts and Organizational Change Reijo Miettinen and Jaakko Virkkunen control operations. At first, external representations of practices might primarily have served the function of preserving the existing practice and only later acquired a function as tools for planning and designing something that does not yet exist, allowing what Bateson (1972) has characterized as second-order learning. Wartofsky (1979: 209) suggests that there are also tertiary artefacts that are abstracted from their direct representational function and constitute a domain in which a free construction, in the imagination, of alternative rules and operations is possible. This construction feeds back to the actual praxis, as a representation of possibilities that go beyond present actualities, serving thus as tools for imagining and proposing alternatives (Wartofsky, 1987: 370). Such artefacts might serve as means not only of finding solutions to problems within the present practice but also of finding a new object and societal telos for the activity.

Making a Practice into an Epistemic Object: The Developmental Contradictions of the Labour Protection Inspector’s Work in the 1980s In the following, we will analyse how the activity of a Finnish occupational health and safety inspectorate was made into an epistemic object, that is, an object of enquiry, in order to reconfigure it. The following four phases of this process will be discussed: (1) analysis of the historical formation of the practice and its inner contradictions, (2) the creation of a model for an alternative practice that might resolve the current contradictions revealed in the analysis, (3) the design of a new set of mediational artefacts with corresponding new procedures (routines and skills, as it were) to concretize and objectify the initial model for a new practice, and (4) the reorganization of the management of the inspectorate to conform to the new object and form of the inspection activity.5 The formation of any epistemic object presupposes the use of adequate concepts. If institutional contradictions are an essential source of change, they should be made into an object of enquiry. Both activity theory and Michel Foucault’s archaeology of knowledge suggest that this may be achieved by analysing the historical emergence and development of the elements of the practice in question. Understanding the circumstances of their emergence makes them more transparent, unravels their ‘taken-forgrantedness’ and shows their limitations in the face of the challenges of the present. In the activity-theoretical framework, an enquiry into institutional contradictions is made operative in terms of the contradictions between the elements of an activity system, typically between the changing object of activity and its historically formed tools, instruments and rules (Engestrom, ¨ 1987). The models of the identified contradictions and of an alternative practice constitute a working hypothesis about a possible new practice, which is tested and elaborated by designing and

445

Organization 12(3) Articles implementing new practical tools and procedures for its further objectification.

Analysing the Practice and Its Contradictions In Finland, there exists state legislation and norms concerning working conditions and the safety of employees in the workplace. The enforcement of this labour protection legislation is the responsibility of 11 district authorities—the occupational safety and health inspectorates. The Ministry of Health and Social Welfare (formerly the National Board of Labour Protection) issues instructions and guidelines concerning working conditions, machinery and equipment and supervises the safety enforcement work of the 11 inspectorates. The main form of the inspectorates’ activity has traditionally been the examination of working premises by inspectors. The purpose of their work is to ensure that labour protection regulations are being observed and to give advice about the improvement of working conditions and workplace safety. During the 1970s, the scope of workplaces to be inspected was radically broadened by new legislation. The newly established National Board of Labour Protection issued detailed safety regulations and standardized the inspection procedure. The Board also developed a system of strategic planning to determine priorities, and launched campaigns and projects to focus the inspectors’ work on important problems. In the 1980s, however, the optimism with which the reform work had taken place during the 1970s declined somewhat. There was a growing awareness among inspectors that their traditional workplace surveys did not effectively address the root causes of the health and safety problems. The inspectors were frustrated by the lack of progress being made in occupational health and safety and they complained that the same defects appeared repeatedly, despite the instructions they issued. In 1989, The National Board of Labour Protection initiated a project to develop the inspection practices (Virkkunen and Kuutti, 2000). The research-and-intervention process was inspired by activity theory and the theory of expansive learning (Engestrom, ¨ 1987). In order to recognize the historical roots of the present-day practices and to identify contradictions within the inspectors’ current activity, a task force of inspectors was established. The ensuing analysis showed that, despite the increased number of workplaces to be inspected and the rapid technological development that had taken place in production, the basic logic of the inspectors’ activity had remained largely the same. Each inspector had responsibility for examining hundreds of workplaces. In addition, inspectors encountered novel technologies and new types of hazards at workplaces. In its analysis, the task force used the activity system model to identify and define the contradictions between the elements of the inspectors’ activity system that called for developmental measures. First, there was a contradiction between, on the one hand, the changed object of the

446

Epistemic Objects, Artefacts and Organizational Change Reijo Miettinen and Jaakko Virkkunen inspectors’ activity—that is, the increased number of workplaces to be inspected and new kinds of safety hazards—and, on the other hand, the prevailing division of labour. An inspector working alone had a fixed but over-extensive and increasingly heterogeneous set of workplaces to inspect. As a consequence of the fixed division of labour, it was difficult to concentrate inspections on important problem areas. Second, the tools for analysing conditions in the workplaces were partly inadequate for dealing with the rapidly changing safety problems. Ideas for reforming the health and safety inspection activity were discussed in a workshop of inspectors and their managers. A recurrent theme was the necessity to concentrate inspection work on the most severe problems. In this seminar, a model for a new form of inspection work was outlined (see Virkkunen and Kuutti, 2000: 307). Instead of single individual inspectors focusing on a fixed set of workplaces to enforce the legislation and improve safety, the inspectors would work in teams on general and important safety problems, analyse their causes, and subsequently plan interventions to change the situation. The new model thus comprised a new concept concerning the object of the inspection activity and a new logic to guide it, as well as an idea of the kinds of artefact that would be needed as instruments to realize this logic and also a novel concept for the subject of the activity: the problemoriented team. The activity of the occupational safety inspectors was made into an epistemic object by using a model of an activity system. This functioned as a tertiary artefact for collective reflection on the current practice, for defining the inner contradictions of the activity, as well as for anticipating its possible future forms.

Elaboration of the Model for an Alternative Practice As a direct result of these reflections, a project was established in the Uusimaa Occupational Safety and Health Inspectorate to develop the new instrument for planning inspection activities, as outlined in the seminar, to overcome the two central contradictions identified in the previous analysis. The project set out to develop and test a new set of tools called the System for Depicting the Field of Activity (SDFA). The system was based on a three-layer, hierarchical depiction of indices revealing safety problems within the Inspectorate’s area of jurisdiction. It would allow both the managers and the inspectors to move between levels of detail while analysing the safety problems. First, an overall analysis of the whole field was undertaken, which used statistics to describe the labour hazards, accidents and occupational diseases as they existed in different branches of industry and different trades in the Inspectorate’s area. This description made it possible to identify and delineate a set of important problem areas for a more detailed study. The second layer of description focused on analysing the occurrence, types and causes of the labour safety hazards in the studied

447

Organization 12(3) Articles problem areas. The third layer of the SDFA was the development of an action plan to remove particular safety problems. Thus the system made it possible for the Inspectorate to specify objects of inspection in a stepwise manner. Application of the first stage of the analytical method indicated that the construction industry had a particularly bad safety record. Consequently, a project team was set up specifically to investigate the underlying reasons. The team explained its approach in a report (19 September 1992) as follows: ‘First, maps describing the construction activity in Uusimaa will be formed. Then the risk indices describing the working conditions will be inserted into them in a suitable way. Once the possible aggregations of the indices have been found, the reasons for the problems will be clarified.’ The team developed a database for recording data about labour safety in different types of construction and in different trades within the construction industry to be used as a second layer of depiction within the SDFA.

Designing and Testing the Instruments for the New Practice As the project progressed, the SDFA methodology gradually produced a series of new tools, techniques and procedures. Four new artefacts mediated the Inspectorate’s changing interaction with its object of activity, that is, occupational and health problems: (1) a conceptual map of the structure of the industrial field, coupled with a corresponding database; (2) a new kind of hypothesis-driven checklist for the inspection of individual workplaces in the industry; (3) a new ‘script’ for inspection discussions; and (4) a project report detailing the results of the studied field of industry. The way in which these artefacts were interconnected was essential for the new team-based inspection activity. Statistical data and knowledge from research on occupational diseases, accidents and the well-being of workers in the industry were combined in the same database with records of the inspectors’ own observations concerning defects in working conditions, thus creating a new kind of knowledge space (Turnbull, 2000). The database therefore served as a collective memory, a means of combining knowledge from different sources, and as an open and easily accessible platform for representing the state of the object of the team’s activity. The development and use of these new tools concretized and elaborated upon the model of the new form of activity produced in the seminar. Instead of inspecting individual workplaces, the inspectors now analysed the causes of unsafe working conditions in sets of workplaces and created measures to improve labour safety. By defining and providing information about a common object, and making the tools commonly available, the SDFA made the team a collective subject of both productive work and learning. The new SDFA-based inspection activity is presented in Figure 1.6

448

Epistemic Objects, Artefacts and Organizational Change Reijo Miettinen and Jaakko Virkkunen Figure 1. The new SDFA-based inspection activity and the related tools and actions.

Previously, individual inspectors used to prepare checklists to guide their inspections. Under the new system, the team of inspectors prepared a common checklist for the inspection project, which included the most important safety risks found in the data analysis. The use of these shared lists thereby became a means to verify and further develop the results of the data analysis. In this way the inspectors also were led to compare their observations systematically. The inspectors also changed the way in which they conducted their discussions during an inspection. The new approach created more dialogue with workplace representatives during the inspections, which brought up more multifaceted knowledge than before. Having inspected selected workplaces and collected data, the inspectors analysed and summarized their findings and then discussed these with their colleagues. In these discussions, the inspectors made further generalizations regarding their findings and decided upon further targets for their inspection work. For example, it was found that occupational risks were often connected with the poor quality of the construction machines used. These machines were often acquired from equipment rental companies, which did not always properly track the maintenance status of the machines they delivered for use. The Inspectorate therefore took action to encourage these firms to improve their quality control systems. When using the database, the inspectors found that the number of accidents was not a good indicator of labour safety because of its random variation. To overcome this problem, an assessment instrument was developed to provide a quantitative estimate of the state of the risk factors found to be most important within the working conditions under

449

Organization 12(3) Articles scrutiny. Inspectors began to collect standardized assessment data on conditions within construction sites using this instrument and, with the analysed data in hand, they could now demonstrate to the top managers the relative position of their firm regarding the quality of its working conditions.

Reorganizing Management to Conform to the New Object of Activity Because the project on the construction industry proved successful, the Inspectorate set new teams on to other problem areas. The new teambased inspection practice could not, however, be managed in accordance with the old procedures and organization of management. The problem areas that the teams worked on crossed and overlapped with the previously defined boundaries between units. Moreover, the spread of the project-based inspection system had created another contradiction: it was very difficult to combine a client service that demanded quick reaction to customer demands with the more proactive and self-regulated project work. In order to find a solution to these new contradictions, the management of the Inspectorate initiated a management development process. An activity system model was again used as an intellectual tool of collective analysis and design. As a result of this process it was decided, in January 1994, to transform the management system. It was decided that one organizational unit would specialize in client service and that each inspector would work for a certain period each year in this unit. These changes consolidated the new inspection practice.

Conclusion and Implications for Future Work This paper has suggested that an understanding of action that is based on the concept of habit or routine is unsatisfactory in explaining the relationship between continuity and change in organizations. None of the theoretical foundations presented for the concept—cognitive scripts (by cognitive psychology), bodily schemes (by Bourdieu and phenomenology) or pre-reflective shared rules (ethnomethodology)—offers a satisfactory account of how forms of activity are transformed and how new practices emerge and are formed. Neither does the thesis that routines change because they are constantly situationally emergent supply a satisfactory foundation for understanding the historical transformations of human practices. These concepts work well in accounting for the recursive nature of social life, but they tend to omit both the analysis of the historically changing cultural context of actions and the role of individuals, the changing object of their activity, and their shared futureoriented moral agency in the transformation of activities. Nor is the concept of routine easily reconciled with the concept of institutional contradiction used by organizational studies to explain the sources of change. When Herbert Simon introduced the concepts of habit and routine, he referred to the work of William James and John Dewey; but, in the further elaboration of the concept, this legacy has been

450

Epistemic Objects, Artefacts and Organizational Change Reijo Miettinen and Jaakko Virkkunen forgotten. Dewey’s basic thesis was that, although habits are necessary and constitute invaluable carriers of traditions and skills, reflective thought is needed when habitualized action confronts permanent difficulties as a result of changing circumstances. This paper suggests that the concept of artefact mediation helps to connect the ideas of institutional contradictions and human subjectivity in a more satisfactory manner. As suggested by Rheinberger, artefacts and objects can have two distinct roles in human activity: they can be either its objects or its means (i.e. technologies). In the latter role they have a ‘black-boxed’ nature and they are not in the sphere of conscious attention, i.e. they constitute the tacit dimension of activity (Polanyi, 1958).7 To crack open the previously hidden self-evidence and ‘givenness’ of ways of acting and to transform the activity, the routines themselves must be made into an object of enquiry, that is, into an epistemic object. This paper has discussed how this might be achieved. Rheinberger’s suggestion is that scientists (or specialists facing the challenge of understanding and/or transforming a practice) are, as L´evi-Strauss (1966) suggests, first and foremost bricoleurs (tinkerers), not engineers. The bricoleur improvises with a set of instrumentalities to adapt to the task at hand. The bricolage of scientists includes improvising, imagining, playing and searching for new, unexpected cultural resources. This activity runs counter to the standardization and routinization inherent in a technological system or in a human practice. Therefore, it also simultaneously constitutes a model of the solution to the basic tension between the necessary reproductive routinization and the transformative tendencies inherent in human practices. However, to our mind, this solution leans too much towards the motives or ‘mentality’ of the individual subjects (scientists, artists) and does not deal enough with the collective nature of working on epistemic objects and how its results are actually objectified in organizational practices. We have suggested that the solution should rather be sought in the development of new tools and organizational forms that make reflection on practices a part of an activity. Forms of reflection can be ‘built in’ as a new practice within the production of services or between R&D, design activities and actual production. Quality circles provide a limited example of this: enquiries into problems of production take place in a separate space and with specialized means different from the productive activity itself, and, during production, workers switch flexibly from the productive practice to a collective problem-solving practice and back again. Karin Knorr-Cetina and Urs Bruegger (2001) suggest that the use of information technology allows the creation of ‘forums of exteriorizing processes and activities’. Such ‘transparent regimes of information’ ‘keep knowledge current, alive and distributed in the respective organizations’ (2001: 180) and ‘preclude it from remaining local and from becoming implicit’ (2001: 193). Nathalie Lazaric and her colleagues (2003) present

451

Organization 12(3) Articles an excellent example of how ways of constantly enquiring into production problems and suggestions for improvements were included in the computer and network-based production control system in the French steel industry. The SDFA presented in this paper also supplies an example of such a transparent platform: enquiry is ‘built into’ the new practice and its instruments. Within each project, hypotheses (which are not known beforehand) are formulated about the most important safety problems and their causes in a selected area of production. These are then tested and verified by inspections, through comparison of the results, and are embedded into the information system, which is open to all the inspectors in the Inspectorate. The case analysed in this paper reveals how the practitioners, together with an outside researcher, used a tertiary artefact to reflect on and make their current organizational routines into an epistemic object, and then managed to create a new instrumentality, the SDFA, that made continuous collective reflection and learning possible.

Notes 1 We selected Nelson and Winter’s book as an object of analysis because of its well-structured definition of the concept of the routine. The definitions presented by Levitt and March (1988: 320), for instance, escape critical analysis by being both short and overly comprehensive: ‘The generic term “routines” includes the forms, rules, procedures, conventions, strategies, and technologies around which organizations are constructed and through which they operate. It also includes the structure of beliefs, frameworks, paradigms, codes, cultures and knowledge that buttress, elaborate, and contradict the formal routines.’ Although Nelson and Winter do not discuss the neoinstitutionalist approach to organizational studies, their evolutionary theory shares with it the leitmotif of finding a psychologically more realistic alternative to the model of rational man of neoclassical economic theory. 2 Powell and DiMaggio (1991: 21) maintain that Garfinkel’s rules ‘more closely resemble the “scripts” or “production systems” of cognitive psychology (Shrank and Abelson, 1977; Klahr et al., 1987) than Parsons’s norms and values’. Compare Suchman’s (1987) critique of the cognitive theory of skills and Williams’ (1999) analysis of the incompatibility of Wittgenstein’s concept of rule with cognitivist theories of mind. Bourdieu’s (1977) concept of habitus and practical reason bears a strong resemblance to ethnomethodology (Bohman, 1999: 130) and also has much in common with the pragmatist concept of habit (Aboufalia, 1999; Joas, 1996). 3 We find the attempts of evolutionary theory to redefine routine as ‘search capability’, ‘deliberative processes’ or ‘metaroutine’ (see Reynaud, 2000: 253) artificial because all these terms refer to conscious thought and reflection. Winter expresses this problem when he refers to the basic difficulty of formulating or even imagining ‘higher-level rules by which lower-level rules are modified’ (1986: 172). 4 One of the referees of this paper suggested that any distinction between primary and secondary experience indicates the adoption of action/reflection or routine/knowledge dualism. This is not the case. For instance, in Dewey,

452

Epistemic Objects, Artefacts and Organizational Change Reijo Miettinen and Jaakko Virkkunen who is known to have been opposed to any hint of dualism, this distinction grows from a monistic, anti-Cartesian stance based on the concept of practice. To him, representation and reflection grow from, and are about, human practices and, to avoid empty intellectualism, they should not be separated from practical action. Both are functionally interdependent aspects of human practice. 5 In principle it would be possible to compare these phases to the corresponding phases in the models of institutional change presented by Greenwood et al. (2002) and Morril (2005). This paper, however, operates on the level of a local community where change is deliberately looked for. 6 The model in Figure 1 was constructed in a discussion between the second author of this paper and the deputy manager of the Inspectorate. 7 The discussion about the relationship between explicit (or articulate) and tacit forms of knowledge is closely connected to what was dealt with in this paper in terms of the relationships between routine, actions and reflection. Space considerations have, however, precluded explicit discussion of the point.

References Aboufalia, M. A. (1999) ‘A (neo)American in Paris: Bourdieu, and Pragmatism’, in R. Shusterman (ed.) Bourdieu. A Critical Reader, pp. 153–74. Oxford: Blackwell. Bateson, G. (1972) Steps to an Ecology of Mind. Chicago and London: University of Chicago Press. Bohman, D. (1999) ‘Practical Reason and Cultural Constraint. Agency in Bourdieu’s Theory of Practice’, in R. Shusterman (ed.) Bourdieu. A Critical Reader, pp. 129–52. Oxford: Blackwell. Bourdieu, P. (1977) Outline of a Theory of Practice. Cambridge: Cambridge University Press. Camic, C. (1998) ‘Reconstructing the Theory of Action’, Sociological Theory 16(3): 283–91. Clemens, E. S. and Cook, J. M. (1999) ‘Politics and Institutionalism. Explaining Durability and Change’, Annual Review of Sociology 25: 441–66. Crossley, N. (2001) The Social Body. Habit, Identity and Desire. London: Sage. Dewey, J. (1925/1988) Experience and Nature. The Later Works of John Dewey, vol 1, ed. Jo Ann Boydston. Carbondale and Edwardsville: Southern Illinois University Press. Dewey, J. (1938/1991) ‘Logic’, The Theory of Enquiry. The Later works of John Dewey, vol. 12, ed. Jo Ann Boydston. Carbondale and Edwardsville: Southern Illinois University Press. Emirbayer, M. and Mische, A. (1998) ‘What Is Agency?’, American Journal of Sociology 103(4): 962–1023. Engestrom, ¨ Y. (1987) Learning by Expanding. Helsinki: Orienta Konsultit. Engestrom, ¨ Y. (1995) ‘Objects, Contradictions and Collaboration in Medical Cognition: An Activity Theoretical Perspective’, Artificial Intelligence in Medicine 7: 395–412. Fujimoto, T. (2002) ‘Evolution of Manufacturing Systems and Ex Post Dynamic Capabilities’, in G. Dosi, R. R. Nelson and S. G. Winter (eds) The Nature and Dynamics of Organizational Capabilities, pp. 244–80. New York: Oxford University Press.

453

Organization 12(3) Articles Greenwood, R., Suddaby, R. and Hinnings, C. R. (2002) ‘Theorizing Change: The Role of Professional Associations in Transformation of Institutionalized Fields’, Academy of Management Journal 45(1): 58–80. Hasselbladh, H. and Kallinikos J. (2000) ‘The Project of Rationalization: A Critique and Reappraisal of Neo-Institutionalism in Organization Studies’, Organization Studies 21(4): 679–720. Hoffman, A. J. (1999) ‘Institutional Evolution and Change: Environmentalism and the U.S. Chemical Industry’, Academy of Management Journal 42(4): 351–71. Hughes, T. P. (1978) ‘Inventors: The Problems They Chose, the Ideas They Have, and the Inventions They Make’, in P. Kelly and M. Kransberg (eds) Technological Innovation: A Critical View of Current Knowledge, pp. 166–82. San Francisco: San Francisco Press. Joas, H. (1996) The Creativity of Action. Chicago: University of Chicago Press. Klahr, D., Langley, P. and Neckes, R., eds (1987) Production System Models of Learning and Development. Cambridge, MA: MIT Press. Knorr-Cetina, K. (1997) ‘Sociality with Objects. Social Relation in Postsocial Knowledge Societies’, Theory, Culture and Society 14(4): 1–31. Knorr-Cetina, K. and Bruegger, U. (2001) ‘Transparency Regimes and Management by Content in Global Organizations. The Case of Institutional Currency Trading’, Journal of Knowledge Management 5(2): 180–94. Latour, B. (1991) ‘Technology Is Society Made Durable’, in J. Law (ed.) A Sociology of Monsters: Essays on Power, Technology and Domination, pp. 103–32. Sociological Review Monograph 38. Latour, B. (1992) ‘Where Are the Missing Masses? The Sociology of a Few Mundane Artifacts’, in W. Bijker and J. Law (eds) Shaping Technology/Building Society. Studies in Sociotechnical Change, pp. 225–64. Cambridge, MA: MIT Press. Lazaric, N., Mangolte, P.-A. and Massu´e, M.-L. (2003) ‘Articulation and Codification of Collective Know-How in the Steel Industry: Evidence of Blast Furnace Control in France’, Research Policy 32: 1829–47. Lektorsky, V. A. (1980) Subject Object Cognition. Moscow: Progress Publishers. Leont’ev, A. N. (1978) Activity, Consciousness, and Personality. Englewood Cliffs, NJ: Prentice Hall. Levitt, M. and March, J. G. (1988) ‘Organizational Learning’, Annual Review of Sociology 14: 319–40. L´evi-Strauss, C. (1996) The Savage Mind. Chicago: University of Chicago Press. Lynch, M. (1992). ‘Extending Wittgenstein: The Pivotal Move from Epistemology to the Sociology of Science’, in A. Pickering (ed.) Science as Practice and Culture, pp. 215–65. Chicago and London: University of Chicago Press. McGowan, J. (1998) ‘Toward a Pragmatist Theory of Action’, Sociological Theory 16(3): 292–7. Meyer, A. D., Brooks, G. R. and Goes, J. B. (1990) ‘Environmental Jolts and Industry Revolutions: Organizational Responses to Discontinuous Change’, Strategic Management Journal (special issue): 93–110. Miettinen, R. (2000) ‘The Concept of Experiential Learning and John Dewey’s Theory of Reflective Thought and Action’, International Journal of Lifelong Education 20(1): 54–72. Miller, G. A., Galanter, E. and Pribram, K. H. (1960) Plans and Structures of Behavior. New York: Holt, Rinehart & Winston.

454

Epistemic Objects, Artefacts and Organizational Change Reijo Miettinen and Jaakko Virkkunen Morril, C. (2005) ‘Institutional Change through Interstitial Emergence: The Growth of Alternative Dispute Resolution in American Law, 1965–1995’, in W. W. Powell and D. L. Jones (eds) How Institutions Change. Chicago: University of Chicago Press. Nelson, R. and Winter, S. (1982) An Evolutionary Theory of Economic Change. Cambridge, MA: Belknap Press. Newell, A., Shaw, J. C. and Simon, H. A. (1972) Human Problem Solving. Englewood Cliffs, NJ: Prentice Hall. Polanyi, M. (1958) Personnel Knowledge. Towards a Post-critical Philosophy. New York: Harper & Row. Powell, W. and DiMaggio, P. J. (1983) ‘The Iron Cage Revisited. Institutional Isomorphism and Collective Rationality in Organizational Fields’, Administrative Science Quarterly 48: 441–76. Powell, W. and DiMaggio, P. J. (1991) ‘Introduction’, in W. Powell and P. J. DiMaggio (eds) The New Institutionalism in Organizational Analysis, pp. 1–38. Chicago and London: University of Chicago Press. Powell, W. W., Koput, K. W. and Smith, K. (1996) ‘Interorganizational Collaboration and the Locus of Innovation: Networks of Learning in Biotechnology’, Administrative Science Quarterly 41: 116–45. Reynaud, B. (2000) ‘The Properties of Routines: Tools of Decision Making and Modes of Coordination’, in P. Saviotti and B. Noteboom (eds) Technology and Knowledge. From the Firm to Innovation System, pp. 249–62. Cheltenham: Edward Elgar. Rheinberger, H.-J. (1997) Toward a History of Epistemic Things: Synthesizing Proteins in the Test Tube. Stanford, CA: Stanford University Press. Schrank, R. and Abelson, R. (1977) Scripts, Plans, Goals and Understanding. Hillsdale, NJ: Lawrence Erlbaum. Schumpeter, J. A. (1934) Theory of Economic Development. Cambridge, MA: Harvard University Press. Scott, R. W. (1995) Institutions and Organizations. Thousand Oaks, CA: Sage. Simon, H. A. (1976) Administrative Behavior. A Study of Decision-making Process in Administrative Organization, 3rd edn. London: Free Press. Strang, D. and Meyer, J. W. (1993) ‘Institutional Conditions for Diffusion’, Theory and Society 22: 487–511. Suchman, L. (1987) Plans and Situated Action. The Problem of Human Machine Communication. Cambridge: Cambridge University Press. Taylor, C. (1995) ‘To Follow a Rule’, in C. Taylor Philosophical Arguments, pp. 165–80. Cambridge, MA: Harvard University Press. Turnbull, D. (2000) Masons, Tricksters and Cartographers. Amsterdam: Harwood Academic Publishing. Virkkunen, J. and Kuutti, K. (2000) ‘Understanding Organizational Learning by Focusing on “Activity Systems”’, Accounting Management and Information Technologies 10: 291–319. Vygotsky, L. S. (1979) Mind in Society. The Development of Higher Psychological Processes. Cambridge, MA: Harvard University Press. Vygotsky, L. S. (1986) Thought and Language. Cambridge, MA: MIT Press. Wartofsky, M. (1979) Models: Representation and Scientific Understanding. Dordrecht: Reidel. Wartofsky, M. (1987) ‘Epistemology Historicized’, in A. Shimony and D. Nails (eds) Naturalistic Epistemology, pp. 357–74. Dordrecht: Reidel.

455

Organization 12(3) Articles Wertheimer, M. (1959) Productive Thinking. New York: Harper & Brothers. Williams, M. (1999) Wittgenstein, Mind and Meaning. Towards a Social Conception of Mind. London and New York: Routledge. Winter, S. (1986) ‘The Research Program of the Behavioral Theory of the Firm. Orthodox Critique and Evolutionary Perspective’, in B. Gilad and S. Kaish (eds) Handbook of Behavioral Economics, pp. 151–87. London: JAI Press.

Reijo Miettinen is Professor of Innovations and Research Work in the Department of Education at the University of Helsinki and vice director of the Center for Activity Theory and Developmental Work Research. Before becoming an academic, he worked as a planning officer and researcher in the Technical Research Center of Finland, being in charge of the in-house training of the research scientists. His interest is in studying innovation and research practices using the resources of cultural historical activity theory (CHAT) in dialogue with the practice-oriented approaches of science and technology studies. Among his recent publications is National Innovation System. Scientific Concept or Political Rhetoric (Edita, 2002). Address: Center for Activity Theory and Developmental Work Research, Department of Education, PO Box 26, Teollisuuskatu 23, 0014 Helsinki University, Finland. [email: [email protected]] Jaakko Virkkunen is Acting Professor of Adult Education and senior researcher at the Center for Activity Theory and Developmental Work Research at the University of Helsinki. Before becoming an academic, he acted as a human resources development specialist and management consultant in state administration. His current research interests include developmental interventions aiming at the expansive transformation of work activities as well as forms of collaborative agency in the transformation of work practices. Address: Center for Activity Theory and Developmental Work Research, Department of Education, PO Box 26, Teollisuuskatu 23, 0014 Helsinki University, Finland. [email: jaakko.virkkunen@ helsinki.fi]

456