The DOT-system as an actor-network
Student Joline Boudewijns Student number 4122232
Department Faculty of Economics Master specialization Accounting & Control Supervisor Drs. R.H.R.M. Aernoudts
Date July, 2016
Master thesis
Abstract
Over the last decades, thoughts based on the philosophy of New Public Management (NPM) have been the driver for changes within public sector organizations. The rationale behind the concept of NPM is that prior concerns regarding the legitimacy and efficiency of the services of the public sector could be met by the introduction of private sector business methods. In health care, the philosophy that NPM encompasses has been the driver for the introduction of a cost-effective accounting system. This system, based on ‘diagnosis-treatment combinations’ (DTC) helps to codify performed clinical activities in a systematic manner, thereby covering the complete process of care into an end product to invoice. Over the years, the DTC-system has received a lot of critique. Therefore, in 2012 an improved program has been introduced, referred to as ‘DTC on its way to transparency’ (DOT). This qualitative research examines how the DOT-system in hospitals performs as an actor-network. It does so by incorporating an actor-network approach, which allows to provide insight in how human actors and objects of technology may combine to provide networks within organizations. The results show that because the DOT-system is able to establish itself as an obligatory passing point medical professionals have to pass, medical professionals have to interact with the system in order to achieve a long-term interest. The system establishes a frame which determines the identity and room of maneuver for medical professionals. It affects them as they cannot escape from the frame without forsaking their long-term interest for the Hospital and the exercise of their profession. Their common goal with the DOT-system on the long-term and the people who speak for it create a strong connection with the medical professionals and this makes them keep connecting with the system. Practically, this means that a back-log in registration is done at another moment, whether this is during a break, in the evening at home, or even during weekends.
Preface
This thesis reflects the competences gained over the past years studying at the Radboud University. It is written as a completion of the master Accounting & Control. Five years ago I started a bachelor Business Administration, which marked the start of dynamic years that followed. Although the main focus has been on passing my bachelor’s and master’s degree, I also have had the change to develop myself in several ways, of which being the President of the student association was one of the highlights. I am very thankful for all the opportunities I have had.
Over the years and mainly during the process of writing a Masterthesis, there are several people who have supported me. First of all, I would like to thank my supervisor, Roel Aernoudts. His critical comments provided me valuable input during the process of writing. I also would like to thank Ed Vosselman for being the second reader of this thesis. Furthermore, I would like to express my thanks to Frederike Berendsen, Head of the Department Gynecology in the Jeroen Bosch Hospital. Her enthusiasm for my thesis and the practical support she provided to me to contact the medical professionals helped me a lot. Besides, I have really enjoyed interviewing the medical professionals. Their openness and answers provided me a lot of insight regarding the DOT-system and their work.
As this thesis is the final accomplishment of my Master, I would like to take the opportunity to thank my parents for all the financial and mental support they gave me over the years. They have learned me to work hard and I am so proud of what they do and who they are. Dad and Mom, you mean so much to me! Moreover, I would like to thank my boyfriend. Although he studied abroad over the last year, the quantity of cards, chocolates, and messages he sent me reflects his involvement and love. At last, I would like to thank my sisters and friends who were there for me whenever I needed it.
I hope you will enjoy reading this thesis.
Table of Contents Abstract...2 Preface...3 1. Introduction...6 1.1. Introduction...6 1.2. Research problem...8 1.3. Relevance...9 1.4. Outline...10 2. Theoretical background...11
2.1. New Public Management...11
2.1.1. Developments regarding the role of the government...11
2.1.2. The rise of New Public Management...12
2.2. Implications of New Public Management...14
2.2.1. DTCs and DOTs in the Netherlands...14
2.2.2. DTC on its way to transparency (DOT)...15
3. Theoretical framing...17
3.1. Prior research...17
3.2. Actor-network theory...18
3.2.1. Characteristics of actor-network theory...19
3.2.2. The establishment of an actor-network...22
3.2.3. The continuation of an actor-network...27
4. Methodology...33
4.1. Research methodology...33
4.2. Research object...34
4.3. Research method...34
4.4. Methodological guideline: Latour’s rules of method...36
4.5. Difficulties within the study of an actor-network...40
4.5.1. Temporal bracketing...41
5. Results...44
5.1. The actor-network of the DOT-system...44
5.1.1. The DOT-system as an inscription...44
5.2. Requirements for the ordering and continuation of the DOT-system as an actor...46
5.2.1. The creation of a network around the DOT-system...46
5.2.2. The frames and overflows the DOT-system engenders...49
5.2.3. The presence of translations...53
5.3. The effects of the system on the work of professionals: the tension between registration and helping patients...54
5.3.1. The continuation of the network...54
5.3.2. The impact on the work of medical professionals...55
6. Conclusion...58
7. Discussion...60
7.1. Discussion...60
7.2. Limitations of the research...61
7.3. Directions for further research...62
1. Introduction
1.1. Introduction
In the last decades, the public sector has been subject to significant changes. Instead of the former clear separation between the public and the private sector, a number of OECD-countries moved towards the philosophy of New Public Management (NPM) during the 1980s (Dunleavy & Hood, 1994; Hood, 1995). NPM is described as “a set of assumptions and value statements how public sector organizations should be designed, organized, managed and how they, in a quasi-business manner, should function.” (Diefenbach, 2009, p. 893). The drive for the public management reform was fundamentally a response to an increasing dissatisfaction and concern related to the legitimacy and efficiency of government performance (Speklé & Verbeeten, 2014). By implementing business-like accountability concepts, the public reform intends to make the public sector more performance-, cost-, efficiency- and audit oriented (Diefenbach, 2009; Speklé & Verbeeten, 2014). The philosophy of NPM comprises a broad span of dimensions of change that the public sector should entail, which are categorized by Hood (1991; 1995) in seven elements. These elements include a hands-on professional management, explicit standards and measures of performance, a greater emphasis on output controls, a shift to disaggregation of units, a shift towards greater competition, stress on private sector styles of management practice and stress on greater discipline and parsimony in resource use (Hood, 1991).
One of the major parts of public services in which the impact of the philosophy of NPM is visible, is the health care. In the Netherlands, till 2005 the financing of hospitals was based on budgeting to control health care expenditures. This type of financing, however, did not disclose information regarding the output and performance of hospitals. Besides, the amount of money provided on a yearly basis was determined beforehand and did not depend on performance. Incentives to increase production or to produce in a more efficient manner were therefore mainly absent, leading to an expanding level of health care expenditures over the years (Oostenbrink & Rutten, 2006). Based on the thoughts of NPM, a call arose to introduce a new way of organizing in hospitals. One of the initiations, referred to as diagnosis-treatment combinations (DTC) started in 1994 and was introduced in hospitals in 2005. The system based on DTCs represents an accounting registration system that codifies clinical activities in a systematic manner, thereby covering the complete process of care (Lowe & Doolin, 1999).
As it enables to classify the type of care hospitals provide, it allows transparency and insight in health care activities. The system is based on the comparable international diagnosis-related groups (DRG) system introduced in the 1980s, although it has much more detail than the international applied DRG-system has. All activities that medical professionals perform have to be registered, in which the sum of the clinical activities leads to an end product that reflects the care provided by the hospital. Instead of invoicing all separate clinical activities performed, the costs are allocated into an end product that covers those activities. When a product is determined, the care provided can be invoiced to health insurance companies (NZa, 2015).
Although the implementation of NPM business-like concepts intends to make the public sector more performance-, cost-, efficiency- and audit oriented, the DTC-system has received a lot of critique over the years. The critique mainly focuses on concerns regarding the complexity and the undesired bureaucratic work the system engenders. Although the intention of the system was to create transparency through simplicity, over the years the system has been turned into a labyrinth of possible end products. As this also provided leeway for fraudulent usage, a new program has been introduced in 2012 (NZa, 2015). This program, referred to as ‘DTC on its way to transparency’ (DOT) reduces the diversity of end products set by the DTC-system from 30,000 to 4,400 products and changes the discretion of professionals to determine the type of product. Like the DTC-system, medical professionals register the performed clinical activities into the DOT-system, which are then transferred via predetermined rules into a certain end product to invoice. To determine the amount to invoice these products, prices are negotiated between the health care provider and health insurers on a yearly basis.
The reforms that have been taking place in response to the problems in health care regarding the expanding level of health expenditures, the measurement of the outputs of the services provided by the health care and the lack of financial responsibility of medical professionals make health care a special case to study (Kurunmäki, 2004). It is therefore remarkable that research regarding the changes in the context of the NPM reforms for hospitals is limited, even because the new structures in health care are faced with critique both in literature and from the professionals working with the changed systems (Diefenbach, 2009). From a positivistic viewpoint, Speklé and Verbeeten (2014) focus on one of the dimensions of change the public sector should engender according to NPM-thoughts by examining the new performance measurement systems in public sector organizations, including hospitals.
They argue that critique regarding those systems is the result of the fact that the new quantitative style of performance measurement can only work under certain conditions. Based on the framework of Hofstede (1981), they distinguish three conditions that determine the type of control that is appropriate when looking at the nature of the activities. Speklé and Verbeeten (2014) show that if these three conditions are not met, this could lead to certain errors within the organization. Visser (2016) applies the arguments of Speklé and Verbeeten (2014) by stating that the type of control introduced in public sector organizations does not match with the nature of public sector activities. As the nature of public sector activities is characterized by medical professionals who need space for professional judgment (Murphy & Skillen, 2015), this collides with the new quantitative style of organizing and controlling in public sector organizations. Visser (2016) therefore argues that resistance regarding the NPM concepts in public sector organizations can be explained by the current mismatch between the type of control and the nature of the public activities.
Other studies have taken an interpretative perspective to study the changes in the context of the NPM reforms for hospitals. Kurunmäki, Lapsley and Melia (2003), Kurunmäki (2004) and McGivern, Currie, Ferlie, Fitzgerald and Waring (2015) study the acquisition of management accounting expertise by medical professionals. These studies show how the role of medical professionals changed by the health system reform. Instead of medical professionals being occupied with the achievement of clinical activities, they were gradually faced with the need to adopt financial and management administration skills. Their research describes how the work of medical professionals changes by the introduction of NPM thoughts into hospitals.
1.2. Research problem
Prior studies have contributed to our understanding of the reforms within health care by providing an explanation of the rising critique or by examining the influence of the reforms on medical professionals. The studies of Kurunmäki et al. (2003), Kurunmäki (2004) and McGivern et al. (2015), however, do not explicate why the described facts have come to be settled as they are. This thesis does so by describing how the DOT-system within hospitals has been settled as it is. Specifically, it examines how the DOT-system performs as an actor-network with medical professionals. Medical professionals in general include doctors, nurses, hospice workers, emergency medical technicians and other trained caregivers, but this thesis focuses on doctors as medical professionals.
The main question addressed in this thesis is:
‘How does the DOT-system perform as an actor-network?’
In order to address this question, an actor-network approach is taken. This socio-philosophical approach is particularly applicable for the study of technology as it replaces humanism with heterogeneity, which means that agency is assigned to human and non-human. Its focus is on the onset and shaping of actor-networks and the interactions between the actors within a network (Czarniawska & Hernes, 2005). By incorporating an actor-network approach into this research, it enables to provide insight into both the detail of how the DOT-system is performed within hospitals as well as the manner in which human actors and objects of technology may combine to provide networks within organizations (Lowe, 2001).
1.3. Relevance
Although accounting writers have adopted ideas from actor-network theory in a variety of areas, literature in which heterogeneity is brought together in the same analytical view is still limited (Lowe, 2001). This thesis builds on the work of John Law, Bruno Latour and Michel Callon by showing the many interactions and parties involved during ‘accounting in action’ in hospitals (Chua, 1995). It contributes to current literature by responding to the claim of Latour (1994) that the technical should be brought in to understand the social and provides theoretical understanding of social interaction with technical mediation. Prior accounting research taking an actor-network perspective has tended to study networks of human actors instead of taking heterogeneity into account (Lowe, 2001). In order to provide rich material for actor-network theory studies, Lowe (2001) states that accounting information systems are the obvious candidate for further actor-network research. This thesis does so by studying the DOT-system as a medical accounting registration system.
Secondly, the theoretical contribution of this thesis is that by taking an actor-network approach, this thesis allows to obtain an understanding of the complex interactions of a network of actors (Chua & Mahama, 2007). A greater explanatory power can be achieved than could be through other forms of study by tracking connections within a network, in which both human and non-human are included. This implies that actor-network theory provides a lens through which to view technology also as an actor that shapes social processes. It is interesting to study the DOT-system via actor-network theory as it alerts us to the significance of technological developments. Actor-network theory allows to create consciousness of actors that otherwise would be overlooked by focusing on their content and
their context (Latour, 1987). It creates visibility, an awareness of human and objects, by revealing why they are as they are and how their identities affect how other actors act. This implies that actor-network theory opens up a way of seeing the unintended consequences that arise during the shaping and continuation of actor-networks. This is an aspect that is lacking in other forms of study. Incorporating an actor-network approach into this thesis contributes to current literature as it not just focuses on describing interactions, but rather opens up a new way of perceiving why actors act the way they do and how their actions lead to (un)intended consequences for other actors connected within the network.
Furthermore, this thesis provides a specific elaboration of the NPM philosophy, namely the DOT-system in the Netherlands. Although NPM received widespread attention since its introduction, leading to one of the dominant paradigms nowadays for public management across the world (McLaughlin, Osborne & Ferlie, 2002), specific and detailed descriptions of its widespread use are limited. This study shows an example of the elaboration of this philosophy within health care.
Practically, this study creates an understanding regarding the impact of the NPM reform, and the DOT-system specifically, for policy makers. The rationale behind the DOT-system and its consequences for the work of professionals are often criticized in the literature and by medical professionals. Their critique focuses on the complexity and the undesired bureaucratic work the system engenders (Diefenbach, 2009). Instead of coming up with solutions to respond to this critique, this thesis shows why the system affects the work of medical professionals. An understanding of the essence of why the system works the way it does may provide useful insights for policy makers. Rather than attempting to tackle existing critiques, this thesis allows policy makers to grasp the underlying operations of the system, thereby providing them with an understanding of the working of actor-networks in practice.
1.4. Outline
The remainder of this thesis is structured as follows. Chapter two provides an overview of the rise of the NPM reform and its main elements. The implications of the reform shown by the introduction of DTCs and DOTs in the Netherlands is then described. In order to answer the research question, the rationale behind actor-network theory and its main elements are discussed in Chapter three. Chapter four provides the research method and the object of study. In Chapter five, the results of the interviews are described. Chapter six provides the main conclusions regarding the research question.
2. Theoretical background
This chapter describes the movements that have led to the introduction of the DOT-system. It does so by starting to discuss the changing perception over the past decades about what the government should offer its citizens. Onwards the 1970s, the efficiency and effectiveness of public services became questionable. This has led to the rise of NPM, a concept which is elaborated in this chapter. The DTC- and DOT-system in hospitals show an application of the thoughts of NPM. These systems are described in the latter part of this chapter.
2.1. New Public Management
2.1.1. Developments regarding the role of the government
The nature of public services, and the research and theory about these, has changed significantly over the last decades. This is mainly attributed to the changing perception of the provision the government should offer to its citizens (McLaughlin et al., 2002). After a period where the state just provided a basic minimum of essential provision, the period onwards the 1950s was characterized by the welfare state, referring to the idea that the government should provide all the basic needs of its citizens. However, from the late 1970s, dissatisfaction related to this line of thought arose. The arguments mainly referred to the need for services designed to meet individual needs instead of a basic level of service for all. Moreover, service users increasingly demanded participation in the design and delivery of public services as their efficiency and effectiveness was questioned (McLaughlin et al., 2002).
Mainly the health care, one of the largest parts of public services, has faced numerous challenges regarding its criticism of questionable efficiency and effectiveness. Over the 1950s and 1960s, the philosophy of the welfare state led to an expanding level of health expenditures and a perceived ambiguity of health care objectives (Kurunmäki, 2004). Moreover, the prior focus of the health care was on process accountability. This form of accountability enables medical professionals to have an ultimate freedom to choose their own best practices. However, their actions could not be verified and controlled by this type of accountability, whereof mainly health care financiers lacked information to appropriately measure and compare the outputs of health care providers. In line with this lack of measurement, the ethics regarding the financial responsibility of the medical professionals became a matter of concern for these financiers and other parties (Jones & Dewing, 1997;
Kurunmäki, 2004). The concerns of the managerial practices and financial control systems that were employed in hospitals therefore gave rise to an increased demand for greater accountability, a better management to diminish the professional power and a more efficient use of services/resources (Kurunmäki, 2004).
2.1.2. The rise of New Public Management
In the 1990s, the debate about the public sector services gained saliency with the introduction of NPM. Although this movement began in the late 1970s and early 1980s, the common characteristics of this reform were labeled by the term NPM in the 1980s (Gruening, 2001). The rationale behind this concept is that the concerns of the public sector services could be met by the introduction of private sector business methods. The heart of this change is a focus on “accountingization”, a term coined by Power and Laughlin (1992, p. 133). It implies an emphasis on accountability to lessen the differences between the public and private sector and to shift the emphasis from process accountability towards accountability in terms of results (Hood, 1995). The encompassing changes involve a shift in two basic designs of the public sector (Dunleavy & Hood, 1994). On the one hand, the public sector is moved ‘down-grid’. This means that the extent of discretionary power is reduced by general and uniform rules of procedure for staff, contracts and money. The second shift for the public sector comprises the movement ‘down-group’. Instead of the former seclusion of personnel, structure and business methods, the public sector is pushed towards the same methods that are used within the private sector, thereby making it less distinctive from the concepts used in the private sector. Hood (1995) categorizes seven elements that embrace the philosophy of NPM, which are depicted in Table 1. The first element is the unbundling of the public sector into separately managed units. This disaggregation into cost center units which are organized by ‘products’ intends to make the units better manageable. The second change is the exposure of the public sector to market forces. The introduction of competition into this sector should lower costs and improve standards as it is then subject to the rivalry of the market. The third and fourth element focus on management style and resource use, referring to the need to introduce private-sector management tools and a focus to make a more effective use of resources. The former elements depict the extent of the distinction betw een the public and private sector in its organization and methods of accountability (Hood, 1995). On the other hand, the philosophy of NPM also encompasses ideas about the managerial and professional discretion
within the public sector. Hence, the fifth element advocates more active control for public organizations by visible management control.
Table 1. Doctrinal components of new public management No
. Doctrine Typical justification Replaces Operational significance Some possible accounting implications 1. Unbundling of the
PS into
corporatized units organized by product
Make units manageable; and focus blame; split provision and production to create anti-waste lobby
Belief in uniform and inclusive PS to avoid underlaps and overlaps in accountability
Erosion of single service employment; arms-length dealings; devolved budgets
More cost centre units
2. More contract-based competitive provision, with internal markets and term contracts
Rivalry as the key to lower costs and better standards, contracts as the key to explicating performance standards Unspecified employment contracts, open-ended provision, linking of purchase, provision, production, to cut transaction cost Distinction of primary and secondary public service labor force
More stress on identifying costs and understanding cost structures; so cost data become commercially confidential and cooperative behavior becomes costly
3. Stress on private-sector styles of management practice
Need to apply proven private-sector
management tools in the public sector
Stress on PS ethic fixed pay and hiring rules, model employer orientation centralized personnel structure, jobs for life
Move from double imbalance PS pay, career service, unmonetized rewards ''due process'' employee entitlements Private-sector accounting norms. 4. More stress on discipline and frugality in resource use
Need to cut direct costs raise labor discipline, do more with less
Stable base budget and establishment norms, minimum standards, union vetoes
Less primary employment, less job security less producer-friendly style
More stress on the bottom-line 5. More emphasis on visible hands-on top management Accountability requires clear assignment of responsibility not diffusion of power Paramount stress on policy skills and rules, not active management
More freedom to manage by discretionary power
Fewer general procedural constraints on handling of contracts, cash, staff; coupled with more use of financial data for
management accountability 6. Explicit formal measurable standards and measures of performance and success Accountability means clearly stated aims; efficiency needs hard look at goals
Qualitative and implicit standards and norms
Erosion of self-management by professionals
Performance indicators and audit
7. Greater emphasis
on output controls Need for greater stress on results Stress on procedure and control by collibration Resources and pay based on performance Move away from detailed accounting for particular activities towards broader costs centre accounting; may involve blurring of funds for pay and for activity.
Before the NPM reform, the control of public organizations was mainly driven by a ‘hands-off’ mentality. This mentality, however, conflicts with the concept of accountability, which is the driver of the reform. More active control reflected by assignment of responsibility therefore represents the rationale of the fifth element. In line with accountability, the sixth and seventh dimensions of change are related to the performance of the public sector. Instead of the former focus on process accountability, NPM intends to drive public sector organizations towards output accountability (Speklé & Verbeeten, 2014). This implies more explicit formal standards and measures of performance replacing the former trust in professional expertise.
2.2. Implications of New Public Management
2.2.1. DTCs and DOTs in the Netherlands
In a relatively short period, NPM received widespread adoption, leading to being one of the dominant paradigms for public management across the world (McLaughlin et al., 2002). However, as NPM embraces many elements and is described as a concept or philosophy rather than a predetermined framework, the characteristics assigned to NPM differ. Gruening (2001) distinguishes characteristics which are often attributed to NPM and the characteristics that are not recognized by all observers. As Table 2 shows, most of the undisputed characteristics are embraced by the categorization by Hood (1995). However, this list also comprises characteristics that are not explicitly mentioned by Hood (1995), which shows the span of the concept. In practice, this has led to differences how the reform is implemented into public sector organizations.
In hospitals, the culture has long been marked by a bureaucratic structure in which medical professionals had the ultimate freedom to choose their own best practices. Driven by an increased need for transparency and efficiency, substantial changes were envisaged (Lapsley, 1994). For the health care in the Netherlands, the elements revealed by NPM were one of the drivers for the introduction of a cost-effective accounting system. The resulting system based on ‘diagnosis-treatment combinations’ (DTCs) was introduced in February 2005, aiming to register and reimburse hospital and medical specialist care (Oostenbrink & Rutten, 2006). The system, however, received a lot of critique over the years. This critique, mainly regarding its complexity and the undesired bureaucratic work it engenders, was the driver for the update of the system. In 2012, a new program has been introduced, referred to as ‘DTC on its way to transparency’ (DOT). This system reduces the prior 30,000 predefined DTC products to 4,400
possible end products. Although the system is referred to as the DOT-system, the products it includes are described as DTC-products.
Undisputed characteristics (identified by
most observers) Debatable attributes (identified by some, but not all observers)
Budget cuts Legal, budget, and spending constraints Vouchers Rationalization of jurisdictions
Accountability for performance Policy analysis and evaluation Performance auditing Improved regulation
Privatization Rationalization or streamlining of administrative structures Democratization and citizen participation
Customers (one-stop shops, case management)
Decentralization
Strategic planning and management Separation of provision and production Competition
Performance measurement Changed management style Contracting out
Freedom to manage (flexibility) Improved accounting
Personnel management (incentives) User charges
Separation of politics and administration Improved financial management More use of information technology
Table 2. Characteristics of New Public Management (Gruening, 2001, Table 1 in original)
2.2.2. DTC on its way to transparency (DOT)
The DOT- and the prior DTC-system characterize themselves as the accounting registration systems of hospitals in which all clinical activities can be codified in a systematic manner. Medical professionals have to register all clinical activities they perform. This varies between giving an injection and a consult, from x-ray photography to anesthesia. To determine a price to invoice, it is not just the case that all activities performed to a patient are summed up. Rather, the price is determined via a central web service, defined as the Grouper. This system analyses the data and determines via predetermined rules an end product that could be invoiced. A care product thus represents the care delivered in a certain period. With the introduction of the DOT-system, the number of possible end products is reduced from 30,000 products to 4,400. When an end product is determined by the Grouper via a predetermined decision tree, the care provided can be reimbursed by health insurance companies (NZa, 2015).
For the establishment of the prices that could be reimbursed for each care product, two segments are distinguished. The first segment, the A-segment, comprises 20% of the care provided. For this segment, maximum prices are determined by the NZa on a yearly basis. The second segment, the segment, is subject to free market forces. Both for the A- and B-segment, health care providers and health insurers negotiate on a yearly basis (NZa, 2015).
3. Theoretical framing
This chapter elaborates on actor-network theory, the perspective that is taken in this thesis to study the DOT-system as an actor-network. First, prior research is discussed in order to show that these studies offer a limited understanding of the complex reality of practices. Then the concepts embraced by actor-network theory are described in detail in order to obtain an understanding of its main assumptions and focus of study.
3.1. Prior research
Prior studies contribute to our understanding of the accounting changes that arose in the health care by the infiltration of NPM thoughts. The study by Speklé and Verbeeten (2014) helps to focus on potential mismatches that could emerge by using a management control system that is not designed for the nature of the activities in a sector that is characterized by professionalism. However, their study offers limited insights into the complex nature of NPM. Speklé and Verbeeten (2014) implicitly assume that dysfunctional conflicts can be managed through the design of appropriate controls by taking the assumption of an objective reality as the starting point (Chua, 1986). It could, however, be questioned whether such an objective and external reality exists.
An alternative to the discussed line of thought, referred to as the mainstream perspective, is the interpretative perspective. One of the most influential proponents of this alternative is Schutz (1953), who argues that reality has no meaning until human beings turn their attention and ascribe meaning to it. Through a process of continuous social interactions a social reality is formed (Chua, 1986). This reality does not describe an objective status quo, but relies on subjective meanings that arise during study. Instead of the assumptions taken by mainstream economics that there is an objective reality that exists independent of the observer, and that knowledge is gained by studying phenomena in this objective and external world, the interpretative perspective assumes a social reality that is emergent, subjectively created and objectified through human interaction (Chua, 1986).
Different studies have taken an interpretative perspective to study the accounting changes by the infiltration of NPM thoughts. Kurunmäki et al. (2003), Kurunmäki (2004) and McGivern et al. (2015) show how the role of professionals hybridizes by describing the changing roles from ‘pure’ professionals into medical professionals in managerial roles. Although these
studies provide interesting insights on ‘accounting in action’ (Chua, 1986), they still offer a limited understanding of the complex reality of practices related to NPM.
In order to get a richer understanding of how the DOT-system as an example of the elaboration of NPM performs as an actor-network, this phenomenon is studied by using an actor-network theory approach.
3.2. Actor-network theory
Actor-network theory, also known as the enrolment theory or sociology of translation, has its origins in the mid-1980s in the work of John Law, Bruno Latour and Michel Callon. It characterizes itself by claiming that the shaping of reality is a process which arises through the interaction between heterogeneous actors. The aspect of heterogeneity is a distinct aspect of actor-network theory, as it means that from an analytical view human and non-human are symmetrically treated. To obtain an understanding of the interactions between those heterogeneous actors, actor-network theory takes a network of actors into account (Chua & Mahama, 2007). These networks form the focus of study. Actor-network theory aims to examine and theorize how networks “come into being, how they move, how actors are enrolled into a network, how parts of a network form a whole network and how networks achieve temporary stability.” (Cresswell, Worth & Sheikh, 2010, p. 2).
Although ‘actor-network theory’ suggests that it is a theory, it can be referred to as a theoretical set of ideas, an approach of how to study phenomena. As explicated by Wacker (1998), a theory must meet four basis criteria. First of all, it has to contain conceptual definitions. Secondly, it needs a domain specification to identify where the theory is to be applied. In the third stage of theory-building, possible logical relationships between variables are deductively determined, often derived from fundamental laws. Based on these relationships, theory predictions are derived and tested (Wacker, 1998). A theory assumes linear and derivative associations, which could be predicted beforehand and be verified or falsified during empirical testing. Actor-network theory counters these characteristics by assuming agnosticism, which implies that a priori assumptions of the nature of networks or causal conditions are abandoned (Ritzer, 2005). It states that knowledge is a social product rather than something that is generated through scientific methods (Law, 1992). The focus of actor-network theorists is on examining these social products from an interpretative perspective. The term ‘theory’ thus might be somewhat confusing when describing
actor-network theory. It is therefore important to keep in mind that the term reflects an approach rather than that it describes a theory.
3.2.1. Characteristics of actor-network theory
Terminology: actants vs. actors, action nets, networks and actor-networks
The study of actor-network theory is concerned with the shaping of actor-networks. These actor-networks are not the same networks as we perceive an old-boys network, a telephone network, or a personal network of friends, colleagues and family. Those networks assume that actors who exist a priori to the network start to connect with each other, whereby the structure of the network determines the processes within. Actor-network theory takes another perspective by examining how these particular actors emerge, why they are connected with each other and how a network is created (Czarniawska & Hernes, 2005). Instead of assuming that actors, organizations and networks exist a priori, actor-network theory rather perceives them as products or stabilized effects of organizing activities. The resulting actor-network is described as the provisional result of processes of translation, performed by the actors out of which they are composed (Ritzer, 2005).
Within actor-network theory, different conceptualizations are used to describe the elements of an actor-network. Czarniawska and Hernes (2005) distinguish between actants and actors, whereby actants are perceived as those which act and are acted upon. Actants become actors if they have transpired a series of translations that enclose them into a network. John Law, Bruno Latour and Michel Callon, however, do not make this clear distinction between actants and actors. To avoid confusion regarding the terms, this thesis refers to actors, whether they are already enrolled into a network or not. The same holds for the distinction that is drawn by Czarniawska and Hernes (2005) between action nets and actor-networks. The term action net derives its nature from organization studies and is allocated to emphasize the study of the construction of connections. Although Czarniawska (2004) agrees that the concept of action nets is close to the concept of actor-network, she argues that actions might produce actors, networks and macro-actors, such as an actor-network. This means that an actor-network is one possible product of translation. Czarniawska (2004) therefore claims to focus on the action net as the starting point to further study how actors, networks and macro-actors derive their substance or essence. On the other hand, John Law, Bruno Latour and Michel Callon mainly focus on the resulting actor-network and explore and characterize this network. The term actor-network itself does, however, not capture the richness of its ideas, which is also
recognized by Callon (1999). The term ‘action net’ better reflects the richness of the ideas within actor-network theory. However, as the intertwined use of the terms may cause confusion, the term actor-network is used in the remainder of this thesis. Although it is recognized that it does not capture the richness of the concepts actor-network theory engenders, this is dealt with by describing the concepts in detail.
The assumption of heterogeneity
What distinguishes actor-network theory from other ways to study phenomena, is its assumption of heterogeneity, the fact that science (humans and their knowledge) and technology (artifacts or its processes and use) are not differentiated. This means that not only humans are perceived as actors that change the world around them, but that also non-human ‘technical’ factors such as systems, machines, texts and hybrids have the ability to drive actions. The fact that human and non-human are not perceived as a duality, but rather as equal important shapers of reality, is referred to as ‘generalized symmetry’ by Callon (1986). Although the term suggests that the equality of actors could be generalized, it rather refers to an ontological approach that no primacy is given to either of these. Instead of viewing human and non-human as different entities, actor-network theory explores these specificities without prejudging their form or shape. They all are perceived as equal actors that are able to shape reality. Latour (1996) shows the scope of the term actor by describing what and who could be actors as follows:
“An ‘actor’ in ANT is a semiotic definition -an actant-, that is, something that acts or to which activity is granted by others. It implies no special motivation of human individual actors, nor of humans in general. An actant can literally be anything provided it is granted to be the source of an action.” (p. 373).
Agency and relationality
The assumption that both human and non-human have the ability to change the world around them, is referred to as agency (Law, 1992). As human and non-human, hereafter referred to as actors, together create a social product at a certain moment, interactions between them form the major focus of study in actor-network theory. The semiotics of actor-network theory suggests that actors take their form and acquire their attributes as a result of their relations with other entities (Law & Moser, 1999). This means that actor-network theory assumes that actors themselves do not have a predetermined identity on forehand. Rather, it considers actors as fundamentally indeterminate, implying that they have no a priori substance or essence. It is via network associations that they derive their substance or essence (Law &
Moser, 1999). These network associations define, shape and name them, and provide them with substance, action, intention and subjectivity (Ritzer, 2005). Network interactions thus create the actor’s substance and essence. Besides, also the capacity of actors to act depends on their relation with other actors. The term ‘agency’ is thus not a characteristic of an actor per se (Chua & Mahama, 2007). Rather, Chua and Mahama (2007) and Creswell et al. (2010) argue that it is an effect of a dispersed network consisting of co-actors, connections and settings that give an actor the capacity to act. Actors themselves are relational effects (Law & Mol, 1995). An example to illustrate that actors derive their actions via network associations, is vividly described by Law (1997). In a parable of the manager and his powers, Law shows that everything is made by relations. A manager-accountant may derive its source of power by a big office, a computer and people working for him. If all these elements are disposed, it could be questioned whether the manager is still a manager. When the computer is removed, and spreadsheets, budgets and calculations are gone, he suddenly knows nothing about the finances of its firm. Without these devices and elements, he no longer is a manager-accountant. What Law (1997) makes clear with this example, is that the manager-accountant is relativity powerless when studied in isolation, nothing more than a ‘naked ape’. It shows that the manager-accountant as an actor has no substance or essence without its network interactions and that its capacity to act also depends on the interactions with other actors that take place.
Network associations
Latour (1986) argues that because of the focus of actor-network theory on interactions between various actors, there should also be a rethinking of the nature of society. He argues that because actor-network theory assigns agency both to human and non-human instead of just focusing on human, the study of an actor-network should shift beyond providing explanations in the form of motives, interests and intentions. Rather than focusing on causes, Latour (1986) claims that the study of actor-network theory should focus on the interactions that take place and on the effects they have to obtain an understanding of the network and its strength to hold things in their place. By doing so, things that are neglected during other ways to study phenomena are made visible. The focus is to obtain an understanding of the network associations, how they act and in which manner they affect actors.
This is described by Ritzer (2005) by stating that actor-network theory is interested in
“the ways in which networks overcome resistance and strengthen internally, gaining coherence and consistence (stabilize); how they organize (juxtapose elements) and convert (translate) network elements; how they prevent actors from following their own proclivity (become durable); how they enlist others to invest in or follow the program (enroll); how they bestow qualities and motivations to actors (establish roles as scripts); how they become functionally indispensable (as obligatory passing points).” (p. 1).
3.2.2. The establishment of an actor-network
Actor-network theory assumes that the activities performed in a network or during the shaping of a network are established through processes of translation. Actor-network theory is, however, not the only approach that takes a translation perspective into account. Wæraas and Nielsen (2016) argue that within organization and management disciplines, the theory of translation is used within institutional theory, actor-network theory and knowledge based theory. The concept of translation, however, has its origins in actor-network theory. It refers to “a complex process of negotiation during which meanings, claims and interests change and gain ground.” (Wæraas & Nielsen, 2016, p. 2). The concept encloses elements that distinguish the concept of translation in actor-network theory from other disciplines. First, it concerns the pursuits of interests, and therefore it has a political meaning. On the other hand, it contains a geometric element as it encompasses the mobilization of actors in different directions. Third, it has a semiotic meaning, as it concerns the process of transformation of meanings that occurs during the movement of objects (Wæraas & Nielsen, 2016). In order to study the complexity of negotiations which form the basis of how networks are built and ordered in a systematic manner, Callon (1986) distinguishes four moments or phases during which the identity of actors, the possibilities of interactions and the margins for maneuver are negotiated and delimited. This description of the processes of building and ordering a network, referred to as the process of translation, provides a tool for the study of a techno-economic network (Knights, Murray & Willmott, 1993) which is so for this thesis as it concerns technical actors. However, by referring to ‘moments’, it is suggested that it comprises some static events in time. To emphasize its dynamic nature, the term ‘phases’ is used here to describe the process of translation. The four phases are described by Callon (1986) as problematization, interessement, enrolment and mobilization. It is argued that a stable network, defined as an actor-world, is achieved when these phases all are transpired.
The process of ordering a network: the process of translation
The first phase mentioned by Callon (1986) is the phase of problematization. This term refers to the definition and exploration of a problem by an actor or group of actors. The definition of the problem is so that other actors also recognize it as their own problem. They acknowledge that they cannot reach their goals by themselves due to respective obstacles, and therefore they need to align their interests with other actors through cooperation. The actor or group of actors, which can be referred to as the prime actor, establishes itself as an ‘obligatory passage point’. The obligatory passage point refers to a situation through which all parties involved must proceed, which makes the prime actor indispensable (Callon, 1986; Knights et al., 1993). Interessement forms the second phase, referring to the series of processes of locking other actors into roles that were proposed for them in the program set by problematization. Those who claim to have a relevant solution persuade others to accept their definitions of the situation and to collaborate in pursuing their favored solution. The prime actor attempts to impose and stabilize the identity of other actors (Knights et al., 1993). Interessement is achieved by cutting or weakening the links between an actor and its related parties who want to link themselves to this actor (Callon, 1986). Each actor that is enlisted by the problematization can submit to being integrated into the initial plan. However, it also could refuse the transaction and define its identity, goals, project, orientation, or interest in another manner, referred to as an anti-program (Callon, 1986). The device of interessement therefore does not necessarily lead to actual enrolment. The phase of enrolment is described as the set of strategies in which the prime actor seeks to define and interrelate the various roles it has allocated to others (Callon, 1986). Those roles are not fixed and pre-established and hence multilateral negotiations may take place to achieve the definition and distribution of roles and the identity of the actors. Related parties that are excluded could counteract by interrupting the process of the forming relationships. Enrollment therefore does not take place per se. It is at first an endeavor, which might be achieved later on. Callon (1986) shows four possible ways in which actors are enrolled, namely by physical violence, seduction, transaction or consent without discussion. The fourth phase is described as the mobilization of allies. As in the first phases representatives are used, it is important to examine if they adequately represent the masses and whether the masses will follow. Mobilization describes the tactics used by the prime actor to ensure that supposed representatives are able to represent those collectivities and do not betray them (Callon, 1986). It refers to the methods deployed by actors to sustain commitment to and the maintenance of the network by persuading them that their interests are the same as those of the prime actor.
Facts
In order to understand the nature of a technology, the processes involved should be examined in its translation (Latour, 1987). Lowe (2001) argues that technical artifacts, like machines or instruments, are not perceived as being part of a pre-existing order, waiting to be discovered. He rather states that “facts are the result of an elaborate process of translation, which relies on the ability of a principal to bring others in to accept their conception and solutions.” (Lowe, 2001, p. 331). As is described in the process of translation by Callon (1986), whether something will evolve as a fact depends on the ability of the prime actor to bring others in. A fact is only established as such when a following is built up which accepts the ideas and does not modify it. To illustrate this claim, Lowe (2001) argues that a statement that is made but never is used or an academic paper that is written but never is taken up by others, may as well never have been written. Latour (1987) refers to this by stating that:
“Fact construction is so much a collective process that an isolated person builds only dreams, claims and feelings, not facts. One of the main problems to solve is to interest someone enough to be read at all; compared to this problem, that of being believed is, so to speak, a minor task.” (p. 41).
Fact construction does not have to be confused with the concept of facts from a positivistic viewpoint, referring to a mental state that something is real and can be verified. It rather has to be perceived from the viewpoint of the process of translation by Callon (1986). Only by seeking to explain the way how things have been developed, and how they have be turned into a fact, a real understanding of how things come to be can be developed (Latour, 1987). The role of translation is therefore of particular interest in the production of facts.
Inscriptions and action at a distance
While it is appropriate to conceive (accounting) systems as technologies that could evolve into facts in the hands of later users, it is an interesting adding to this line of thought to focus on the information the technologies contain. This focus is known as the concept of inscriptions. The term inscription is developed by Robson (1992), informed by Latour (1987), and refers to the material and graphical representations that constitute writing, numbers, lists and tables (Robson, 1992). It is the material translation of any setting that is to be acted upon. As is argued by Robson (1992) most of our knowledge does not come from our own experience of the world, but from books, newspapers etc. An interesting element of inscriptions is that the actor who wants to convey its interest to others does not necessarily has to be present. Inscriptions thus permit action from a distance, which is prized by those who
would seek to persuade, enroll and control others (Chua, 1995; Lowe, 2001). Inscriptions, for example in the form of accounting, allow distant control. It must not be concluded, however, that the exercise of such control is rendered unproblematic (Lowe, 2001). Lowe (2001) argues that the use of inscriptions therefore may result in unintended consequences instead of intended actions and reactions.
The concept of power in actor-network theory
Although the phases are described as different phases of a process of translation, they can overlap. A successful translation of an actor-network depends on the capacity of the network to define and enroll entities which might challenge the definitions and enrolments they are subject to. In first instance, actor-network theory does not assign primacy to any actor, meaning that it does not perceive a director as more influential than an administrative assistant or a computer. This is at odds with the traditional idea of power, in which power is perceived to be the cause of events and actions. The way actor-network theory perceives power, is by taking it as the effect or result of its persuasion via the number of entities networked (Czarniawska & Hernes, 2005). No actor has inherent power, but this must be generated through translations. According to actor-network theory, the fact that some actors are able to establish themselves as obligatory passing points through which all information must pass determines their influence and relative position. This also explains why constraining actor-networks could hold. As long as an actor or group of actors is able to render itself as indispensable and establishes and maintains itself as the ‘obligatory passage point’, other actors are forced to pass if they want to continue to exist, which holds the network together.
Black-boxing
Although an actor-network is referred to as an ‘end product’ by Law (1992), it is important to emphasize the temporality of the stable form due to the ongoing ordering of network associations. When studying an actor-network, which might be any type of organization, social order, technical innovation or scientific discover, the connections and interactions between the heterogeneous actors that are enrolled within the network form the focus of study. These network connections are not restricted to certain boundaries, such as a department or the structural boundaries of an organization. Instead, actor-network theory takes the interconnected collective action of many dispersed actors in a social context into account (Chua & Mahama, 2007). The analytical focus is on all actors that affect action within the network, covering as well social actors as textual, conceptual and technical actors.
Moreover, the actors and phenomena studied consist of heterogeneous networks themselves that form their essence. Each actor itself is the effect or product of a heterogeneous network, in which the elements in their turn also consist of networks (Law, 1992). Actor-network thus states that the actor is both a network and a point therein. To cope with endless network ramifications and complexity when studying an actor-network, simplification is required. Actor-network theory does so by referring to punctualization. Punctualization refers to the process by which complex actor-networks are black-boxed, thereby converting an entire network into a single point or node in another network. The term ‘black-box’ is derived from cybernetics who refers to it when a piece of machinery or a set of commands is too complex (Latour, 1987). In its place a little box is drawn about which they need to know nothing but its input and output. This black-box is taken for granted, meaning that it is perceived as an accepted idea, method or instrument. There are no doubts about what it does, until it is broken down. For actor-network theory this means that because the actor is not perceived as a network in itself during study, but as one actor or black-box, the mapping of the network is simplified. A network of simplified entities which in turn consists of other networks, is defined as an actor-world. An example of black-boxing in an actor-world is depicted in Figure 1. This figure shows how the entire network of an actor is black-boxed into one parameter. Although an actor-network can be simplified in an actor-world, it still consists of composed entities. For that reason, in the remainder of this thesis the term actor-network instead of actor-world is used.
Figure 1. An example of the anatomy of an actor-world
3.2.3. The continuation of an actor-network
Framing and overflowing
The above description of the process of ordering a network could give rise to the suggestion that once the four phases described by Callon (1986) are transpired, a stable network has been set up. This stable network consists of interacting actors, who act more or less independent of their surrounding context. However, it could be questioned if a secluded network exists. As actors are the creators and shapers of a network, actor-network theory assumes that any actor that is removed from, or is added to the network, affects the functioning of the network. This is because the actor-network is the context which gives each entity its significance and defines its limitations (Callon et al., 1986). If one of the elements is removed or added, the whole structure shifts and changes. Also social effects such as changes in power (i.e. power perceived as a persuasion via the number of entities networked rather than power as a possession) have an impact on the network.
Based on the work of Goffman (1971) and his thoughts on translation, Callon (1998) develops the concept of framing/overflowing to study how interactions are organized. The concept of framing means that certain boundaries are established within which interactions take place roughly independent of the surrounding context. The rationale behind establishing a frame is because it serves as the basis to study the network. The fact that a frame is established, however, does not imply that the network is completely secluded from the outside world, meaning that actors have no connections with the outside world. When actors in a network are bracketed or framed, boundaries are drawn between those actors and the rest of the world, but it is recognized that outside connections are present (Callon, 1998). The framing process thus does not just depend on the actors themselves, but it is rooted in the outside world, in physical and organizational devices (Christensen & Skærbæk, 2007). The actors that cross or break through the frame’s boundaries, referred to as intermediaries, give rise to overflows. This term, a sociological revision of the economists’ concept of externalities, refers to the positive and negative externalities that are produced during the process of framing (Christensen & Skærbæk, 2007). So, because of the rooting of actors in the outside world, the barriers of the frame become permeable.
Two opposed arguments can be distinguished to perceive the interactions between the frame and the outside world, so between the relation of framing and overflowing (Callon, 1998).
The first point of view argues that framing is the norm and that overflows are exceptions. When overflows arise, they must be contained and channeled with the help of appropriate investments. The second point of view states that framing is expensive and always imperfect. Therefore, instead of framing, overflows are the norm. Framing is not perceived as something that happens by itself, in which overflows are accidental and which must be put right. It argues that the process of framing is always incomplete. This is explained by Callon (1998) by claiming that actors should not be viewed in the way the economic or sociological perception does, but as a network of interdependencies in which the actor is enmeshed and continuously contributing. It is the second viewpoint that reflects the concept of ongoing ordering recognized by actor-network theory. When assuming an ongoing ordering, it is a violent effort to disentangle the actors concerned from the network of interactions and to push them onto a clearly demarcated ‘spot’ (Callon, 1998). This is because they all have links with the outside world which cannot be wholly detached. Moreover, overflows have many sources and can flow in many directions, making frames quite problematic (Callon, 1998). To illustrate the concept of overflows, Callon (1998) refers to a research contract between an academic research unit and a commercial enterprise. The contract attempts to frame the interaction of the parties by defining the performance of certain actions and the conditions for monitoring. Although the contract can specify and structure the frame within which it will be performed, a number of happenings outside the terms of the contract may occur. At the same time as all elements help to structure and frame the interactions which more or less form the substance, each of the terms of the contract simultaneously provide potential conduits for overflows (Callon, 1998). The different actors thus have a dual role: on the one hand, they frame interactions; on the other hand, they represent openings onto wider networks, to which they give access. Framing in itself is hence not a complete process, as overflows simply are the consequence of the omnipresent links with the surrounding context (Callon, 1998). In order to frame, overflows should also be taken into account.
The concept of framing/overflowing can be used to explain phenomena when taking an actor-network theory perspective. Christensen and Skærbæk (2007) do so and explain why public sector performance reporting that emphasizes external accountability may turn out differently from the official stated aims by using the concept of framing/overflowing. They show that a change may be a fragile accomplishment that might produce outcomes that are not compatible with the aim of introducing a new concept.
Although actors thus can frame the behavior of other actors by prescribing activities other actors should perform, a number of interactions outside the terms of the frame may occur. These unexpected outcomes can be referred to as overflows.
Whereas Callon (1998) further discusses the measurement of overflows from an economic and sociological perspective and the subsequent reframing of interactions, this thesis confines on the frames and overflows the DOT-system establishes and is subject to. As overflows distort framing attempts, it is important to identify who is responsible for them and who is affected by them (Christensen & Skærbæk, 2007). However, although overflows are unavoidable during framing attempts, this does not mean that the frame an actor tries to set will be abolished (Callon, 1986). Christensen and Skærbæk (2007) argue that networks may stabilize if the overflows do not produce heated situations. This refers to a situation of continuous overflowing in which everything becomes controversial, thereby threating the framing attempts. This implies that as long as not everything becomes controversial, framing attempts could hold. The fact that a network may stabilize despite overflows could also be explained by recapping the process of translation by Callon (1986) and the concept of power discussed. By elaborating on these concepts, a stable situation is achieved as long as the actor that sets a frame is able to establish itself or another actor as an obligatory passage point.
Boundary objects
The focus thus far has been on describing the assumptions of actor-network theory and how the building and ordering of a network could be explicated by the sociology of translation of Callon (1986). Based on this model, Star and Griesemer (1989) introduce a concept that extends the concept of interessement, namely the development of boundary objects. By drawing on an example of scientific work that deals with researchers from different social worlds, they argue that scientists and other actors need a certain kind of reconciliation. Based on the perspective of Callon, this means that actors enlist others, re-interpret their concerns into their own program and establish themselves as obligatory passing points. Star and Griesemer (1989), however, emphasize that actors from more than one social world try to conduct such translations simultaneously and hence they argue for an ecological approach that takes multiple viewpoints into account, whereby no primacy is assigned to either of them. Although they further move on towards ecology rather than towards the rationale behind actor-network theory, the concept they introduce provides an interesting addition to this thesis and therefore this concept will be elaborated.
Figure 2 conceptualizes the concept of translation by Callon (1986). The figure shows that interessement can be perceived as a kind of funneling, as the concerns of several actors are reframed into a narrower point, the obligatory passage point (Star & Griesemer, 1989). An actor-network, which is referred to as the alliance by Star and Griesemer (1989), then can be perceived as the result of the translations that make the prime actor indispensable for its allies, the other actors. Instead of stating that through translation one alliance is formed, Star and Griesemer (1989) argue that there could be several obligatory passing points that are negotiated with several kinds of allies if there are actors with different goals. Figure 3 illustrates this idea. The concept of boundary objects refers to “an analytical concept which inhabit several intersecting social worlds […] and satisfy the informational requirements of each of them.” (Star & Griesemer, 1989, p. 393). Boundary objects are the devices that are able to mediate different actor-networks, tying together actors with diverse goals (Briers & Chua, 2001). Star and Griesemer (1989) argue that boundary objectives characterize themselves as weakly structured in common use, thereby allowing to have enough robustness to maintain a common identities across the social worlds, for example a common goal. However, in individual use they become highly structured (Star & Griesemer, 1989). Extending this to actor-network theory, this means that an actor could be described as a boundary object when it is able to establishes itself as an obligatory passing point for different social worlds.
Figure 2. The concept of interessement in the sociology of translation by Callon (1986)
(Star and Griesemer, 1989, Figure 1 in original)
Figure 3. The concept of interessement introduced by Star & Griesemer (1989)
(Star and Griesemer, 1989, Figure 2 in original)
