Kathrin Böhling

Opening up the Black Box

Organizational Learning in the European Commission

Doctoral thesis

Defended in October 2005

Technische Universität Berlin

Committee: Prof. Meinolf Dierkes, Prof. Ariane Berthoin Antal, Prof. Hubert Knoblauch

2

Also published as

Kathrin Böhling (2007)

Opening up the Black Box

Organizational Learning in the European Commission

Peter Lang Verlag, Frankfurt (M.)

3

To my grandmother Maria, and Luis

4

Acknowledgements

This book about organizational learning in the European Commission is the result of much boundary-crossing during most of my academic career so far. As a German sociology student at the University of Amsterdam in the Netherlands I acquired a European attitude that took further shape in the San Francisco Bay Area where I developed my initial research ideas. I started to look at the Com-mission as a black box across the Atlantic. We may be aware of its agenda and initiatives, its president and commissioners, but the Commission’s adminis-trative fabric usually escapes our attention. This book will hopefully be able to contribute to the understanding of the Commission’s agency in Europe. It is also a contribution to the organizational learning program in the research unit “Inno-vation and Organization” of the Social Science Research Center Berlin (WZB).

I would like to thank Professor Meinolf Dierkes, former director of the “Innovation and Organization” unit, who gave me the opportunity to explore my research ideas on boundary-spanning and organizational learning in the stimulating atmosphere of the WZB. His belief and trust in my capacities as a researcher were pivotal during the overall process. Moreover, I am particularly grateful to Professor Ariane Berthoin Antal, senior fellow at the WZB, for her great support throughout the entire, sometimes rather difficult research process. Reading patiently through numerous drafts, she has always provided me with challenging and constructive feedback. Beyond that, I also want to thank a number of colleagues; Mike Geppert (London), Christiane Kerlen (Berlin), Burkhart Lutz (Berlin) and Holger Strassheim (Berlin) for discussing my work with me. Making an empirical contribution to theory-development on organiza-tional learning rests on access to people in the research setting. I am therefore grateful to the many Commission officials and representatives from business and Member State governments for their time and attention in the interviews. Last but not least, I wish to acknowledge the Volkswagen Foundation, and especially Alfred Schmidt, for their generous support.

A final word to those who stayed close to me during the entire research process: I am profoundly grateful to my partner Christian and my friend Susanne. Both have always encouraged and supported me in finding a balance between the great joys of motherhood and the deep satisfaction of an academic career.

Kathrin Böhling, November 2006, Berlin

5

Content

List of Figures................................................................................................................ 8

List of Tables ................................................................................................................. 9

Abbreviations .............................................................................................................. 10

Interview Codes........................................................................................................... 10

1. Introduction............................................................................................................ 11

2. Organizational Learning from Boundary-spanning Activities ......................... 17

Bringing the Environment into Theories of Organizational Learning.................. 17

Linking External Changes with Internal Dynamics in Organizations ................................................................................... 18

The Notion of Boundary-spanning Activities ....................................................... 20

Processes of Organizational Learning ...................................................................... 24

Stages in Processes of Organizational Learning ................................................... 27

Weaknesses of Conceptualizing Organizational Learning in Terms of Stages.................................................................................. 29

Toward a Contextualized Model of Organizational Learning ............................... 30

Knowledge Acquisition through Boundary-spanning Activities .......................... 31

Boundary-spanning Activities and the Distribution of Knowledge ...................... 33

Boundary-spanning Activities and the Interpretation of Knowledge.................... 35

Boundary-spanning Activities and Organizational Memory ................................ 39

Emerging Theoretical Model and Implications for Research ................................ 41

3. Research Design, Context, and Methodology...................................................... 43

Research Design .......................................................................................................... 43

Research Context ........................................................................................................ 45

The IST Program ................................................................................................... 45

The Management of the IST Program................................................................... 48

Focal Boundary-spanning Activities ..................................................................... 52

Methodology ................................................................................................................ 53

6

Data Collection ...................................................................................................... 53

Analysis of the Data .............................................................................................. 57

4. Learning from Boundary-spanning Activities with ISTAG .............................. 59

The Role of ISTAG in the Management of the IST Program................................. 59

The Rise of Organizational Learning Needs ............................................................ 60

Acquiring Knowledge from ISTAG .......................................................................... 63

Providing a Vision for the IST Program ............................................................... 63

Tailoring Knowledge Provision to Learning Needs.............................................. 66

The F6 Unit as ISTAG’s Advocate ....................................................................... 70

Distributing the Knowledge Acquired from ISTAG ............................................... 72

The Linkage between ISTAG Activities and Internal Dynamics of DG INFSO.......................................................................... 72

Facilitators of and Barriers to the Distribution of the Acquired Knowledge................................................................................... 77

Interpreting the Knowledge Acquired from ISTAG............................................... 79

Making Sense of the Acquired Knowledge within the Context of DG INFSO.......................................................................... 79

The Impact of Organizational Politics on Knowledge Interpretation................................................................................. 81

The Organizational Memory and the Knowledge Acquired

from ISTAG................................................................................................................. 82

Challenges to Established Structures in DG INFSO............................................. 82

Inertia in DG INFSO ............................................................................................. 86

Summary of the Main Findings................................................................................. 88

5. Learning from Boundary-spanning Activities with ISTC ................................. 90

The Role of ISTC in Managing the IST Program ................................................... 90

Prescribed Learning Needs ........................................................................................ 92

Acquiring Knowledge from ISTC ............................................................................. 93

Gaining Access to Policy-oriented Knowledge..................................................... 94

Limits to the Knowledge-seeking Behavior of DG INFSO ................................. 96

Channeling of Policy-oriented Knowledge into DG INFSO ................................ 99

Distribution and Interpretation of the Knowledge

Acquired from ISTC................................................................................................. 102

7

Barriers to the Wide Distribution of the Acquired Knowledge in DG INFSO ................................................................... 102

Relevance of the Acquired Knowledge in DG INFSO ....................................... 106

Organizational Memory and the Knowledge

Acquired from ISTC................................................................................................. 109

Stability in the Implementation of the IST Program ........................................... 109

Taking Regulations for Granted .......................................................................... 111

Summary of the Main Findings............................................................................... 112

6. Discussion of the Empirical Evidence in Light of Theory............................... 114

Variations in Processes of Organizational Learning ............................................. 115

The Rise of Organizational Learning Needs ....................................................... 116

Knowledge Acquisition ....................................................................................... 118

Distribution.......................................................................................................... 122

Interpretation ....................................................................................................... 126

Organizational Memory....................................................................................... 129

Concluding Comments ............................................................................................. 131

Distinctive Qualities of Organizational Learning Triggered by Boundary-spanning Activities ....................................................... 133

Overlapping Stages in Processes of Organizational Learning ........................... 135

6. Conclusions........................................................................................................... 138

The Study in Retrospect ........................................................................................... 138

Processes of Learning in the European Commission ............................................ 139

The Learning Behavior of the European Commission....................................... 140

Policy Sector Matters for Learning from Boundary-spanning Activities ........... 142

Theoretical Implications of the Impact of Boundary-spanning

Activities on Organizational Learning.................................................................... 144

The Formation of Learning through Boundary-spanning Activities................... 145

Complementarities in the Contextualized Learning Model ................................ 147

Organizational Learning and the Institutionalization of Organizational Life ......................................................................................... 149

References .................................................................................................................. 152

8

List of Figures

Figure 2.1 Constructs and processes associated with organizational learning ............. 26

Figure 2.2 A contextualized model of organizational learning. ................................... 42

Figure 3.1 A simple model of social research. ............................................................. 44

Figure 3.3 Management process Information

Society Technologies (IST) Program .......................................................... 50

Figure 4.1 Support of ISTAG activities by the F6 unit. ............................................... 70

Figure 4.2 The F6 unit as an advocate of ISTAG......................................................... 72

Figure 4.3 The main DG INFSO actors involved in drafting

the annual work program. ........................................................................... 74

Figure 5.1 Dialogue with the Information Society Technologies

Management Committee.............................................................................. 94

Figure 5.2 Factors limiting the relevance of ISTC input. ........................................... 107

Figure 6.1 Overlaps and distinctions in processes of organizational learning. .......... 137

9

List of Tables

Table 2.1 Social constitution of boundary-spanning roles in

changing environments ................................................................................. 22

Table 2.2 Sources of power and how they are acted upon in organizations................. 37

Table 3.1 The Information Society Technologies (IST) Program................................ 47

Table 3.2 Human resources of DG INFSOa.................................................................. 53

Table 3.3 Distribution of interviews ............................................................................. 54

Table 3.4 Shifting emphasis in the data collection ....................................................... 56

Table 4.1 Advisory reports of ISTAG during the 5th Framework Program ................. 67

Table 5.1 Uncertainty reduction at two levels and its constituent elements............... 105

Table 6.1 Summary of the main findings ................................................................... 115

Table 6.2 Filtering and channeling relevant knowledge into DG INFSO .................. 120

Table 6.3 Facilitators and barriers to the wide distribution of knowledge ................. 124

Table 6.4 Interpretation of knowledge acquired through

boundary-spanning activities ...................................................................... 127

Table 6.5 Embedded learning from boundary-spanning activities ............................. 132

10

Abbreviations

ACTS Advanced Communications Technology and Services Program

DG INFSO Directorate General for Information Society

ESPRIT European Information Technologies Program

IST Information Society Technologies Program

ISTAG Information Society Technologies Advisory Group

ISTC Information Society Technologies Management Committee

RTD Research Technology Development Policy

TAP Telematic Applications Program

Interview Codes

Com00 Interview with Commission official in 2000

Com01 Interview with Commission official in 2001

Com03 Interview with Commission official in 2003

AG Interview with member of ISTAG

PC Interview with member of ISTC

11

1. Introduction

The European Commission is a “black box” (Lequesne 2000, Malek & Hilker-meier 2001). It may be clear why the Commission was created, but it is not clear how it carries out the broad range of activities it was intended for. Its mandate and competencies are based on the founding treaties of the Community (Bach 1999, Edwards & Spence 1997). Formally, the Commission has the sole right to propose legislation, and it is the executive arm of Community governance. It ensures that policies decided upon by the Council of the European Union and the European Parliament are implemented. The Commission also acts as the guardian of the Treaties and serves as the Community’s conscience by propo-sing ideas that reflect the European interest. Moreover, it promotes the consensus-forcing style of decision-making among the Member States, which tend to differ widely in their political and institutional arrangements. Lastly, the Commission represents the Community in third countries and in international organizations such as the World Trade Organization. The European Commission symbolizes the profound consequences that the establishment of the Single Market, the founding of the European Union (EU), and the recent enlargement of the Union have for everyday life. The common public stereotypes of the Commission as an enormous bureaucracy that tries to interfere in everything that people do or as a corrupt, complacent, and overpaid supranational institution (McDonald 2000) fuel skepticism and controversy about the Commission’s role as the motor of Europe’s integration process (Münch 1995).

Difficulties in classifying the European Commission’s organizational identity contribute to the fog surrounding its influence and performance in the European polity. The Commission is a supranational organization that assumes decision-making and implementation powers usually exercised only by public authorities in sovereign states (Feld & Jordan 1994), but it is not the government of a United States of Europe (Sbragia 1992). In one of the first attempts to capture the organizational identity of the Commission, Cram (1994) called it a “multi-organization.” It means that the Commission’s individual Directorate Generals differ in their efforts to expand the scope of their competencies and to put their preferred issues on the policy agenda of Europe. Laffan (1997) draws on this view, seeing the Commission as a “policy entrepreneur” with both the scope and capacity to influence outcomes in the policy process. To capture the dual function of the Commission as executive government and public adminis-tration, Christiansen (1997) introduces the term “politicized bureaucracy” (p. 76). It refers to the combination of (a) the Commission’s strong preference for rule orientation and rule coherence, hierarchical organization, and functional specialization, and (b) the politicization of the major fields of Commission activity, which are to propose legislation, implement policy, and ensure com-pliance. Bach (1999) shares Christiansen’s concern about the Commission’s

12

dual function and sees limited utility in regarding the Commission as a Weberian type of bureaucracy, which is essentially based on routine, repetitive, and orderly action. According to Bach, the Commission ought to be character-ized as a “transnational expertocracy” because of its high-caliber expert staff, its reliance on a technocratic decision-making style, and its entrepreneurship in developing and putting Community policies into practice. However, despite these differentiated approaches to the Commission’s identity as a corporate actor, thorough examinations of the Commission as a product of its organiza-tional dynamics are rare (Bach 1999, Lequesne 2000, Malek & Hilkermeier 2001).

Given the Commission’s weak resources, it is difficult to understand how the Commission can live up to its mandate and still retain a certain degree of autonomy from interest groups and the Member States (Lequesne 2000, Pollack 1998). “That the Commission would be able to ride roughshod over the interests of its opponents (which invariably include at least one Member State), or manage its own priorities in an impositional style, is unlikely” (Cram 1994, p. 199). With a staff of approximately 17,000 officials (Spence 1997) and a budget of €6 billion for 2004, the Commission is smaller than the municipality of Paris. Moreover, neither human nor financial resources has increased with the enlargement of the EU, which now consists of 25 Member States and 450 million people.

The involvement of external stakeholders from Member State adminis-trations, nongovernmental organizations, and expert communities is an impor-tant guide to the Commission’s pivotal role in the definition of policy problems, the translation of initiatives into clear proposals, and the management and application of Community policies (Mazey & Richardson 1997; see also Abels 2000, Bach 1999, Jachtenfuchs & Kohler-Koch 1996, Schmitter 1996). The Commission draws on an intricate system of advisory, preparatory, and regulatory committees to compensate for its limited resources (personnel, budget, expertise, and authority) and to gain legitimacy for its legal and adminis-trative tasks by winning approval from the governments of the Member States (Brücker, Joerges, Neyer, & Schlacke 1997; Haibach 2000). External stake-holder involvement is regulated by the Council’s Comitology Decision,1 which specifies the conditions for the delegation and control of the Commission’s implementation powers. Comitology gives rise to an institutional forum that facilitates intergovernmental discourse and promotes transnational social inte-gration (Neyer 2000). It simultaneously compels the Commission to remain

1 Council Decision 87/373/EEC of 13 July 1987 lays down the procedures for the exercise of

implementation powers conferred on the Commission, Official Journal of the European

Communities L 197/33 (replaced on 28 June 1999 by Council Decision 99/468/EC, Official

Journal of the European Communities L 184/23).

13

sensitive to Member State interests and to search for dialogue with social groups and domestic actors (Christiansen 1997). The Commission’s strong reliance on external stakeholders during the policy process can therefore also be seen as an impediment to its capacity for autonomous action (Gordenker & Saunders 1978, LaPalombara 2001). The Commission is deeply enmeshed in its environment, a condition indicating that it operates with permeable boundaries. The Commis-sion’s openness to its environment entails the risk that the external stakeholders will question the core of its corporate identity, notably its goals of action and underlying assumptions (Aldrich 1979, Haas 1990).

The proposed analysis of the European Commission in terms of organiza-tional learning provides a solution to the puzzle of its agency as an under-resourced supranational institution that operates in a demanding environment, trying to please the heterogeneous set of external stakeholders it depends on for much of what it does. Organizational learning is understood here as a social accomplishment that involves the distribution and interpretation of externally acquired knowledge and its integration with the organizational memory (Huber 1991). This analysis highlights how the Commission becomes aware of the environment in which it finds itself and how it transforms this knowledge about external demands, expectations, expertise, and causal understandings into an organizational property relating to its action. The main argument of this study centers on the process of integrating knowledge from external sources by means of organizational learning and the resulting enhanced ability to cope with changing environments. In short, the thesis is that the European Commission learns how to cope with the changes and developments in its environment through various forms of communicational interactions with external stake-holders.

By pointing out the interpersonal relations between the Commission and its stakeholders in the Member State governments, economy, civil society, and expert communities as a mechanism for environmental scanning, this study emphasizes a basic claim in the organizational learning discourse—that the response of organizations to changing environments triggers learning (Child & Heavens 2001, Dodgson 1993, Hedberg 1981, Huber 1991, Klimecki & Thomae 1997). Theories of organizational learning stress that the behavioral repertoire available to an organization is enhanced by the sharing and sense-making of knowledge across the organization’s different subgroups (Berthoin Antal 2003, Huber 1991). The ability of an organization to improve its performance or prepare for future activities by absorbing knowledge is an important outcome of organizational learning (Child & Heavens 2001, Levitt & March 1988, Wiesenthal 1995).

The treatment of the European Commission as an organization that may learn is not completely new. Brown (2000), for instance, applies the notion of organizational learning to environmental policy-making in the EU. She defines

14

organizational learning as the incorporation of policy feedback and causal understandings that result in “more effective policies, defined as more complex integrated, flexible, and implementable legislation that enhances goal achieve-ment” (p. 576). Scientific uncertainty is viewed as a trigger of organizational learning because it increases the willingness of the actors involved to seek out and accept information from inside and outside the organization. Malek and Hilkermeier (2001), too, turn to notions from the organizational learning discourse and apply them to the study of the European Commission. They direct attention to the conditions under which the European Commission reconsiders its guiding assumptions and develops new and integrated understandings about policy problems and solutions. Malek and Hilkermeier contend that the Euro-pean Commission has learned whether its staff has developed an “integrated understanding of problems and solutions” (p. 4) across departmental boundaries.

Scholars of international relations have also recently shown a growing interest in the accumulation and use of knowledge by international organizations as an important guide to the understanding of their dynamics. International organizations are conceived of as the “recalcitrant tools” of their architects, tools that have a life of their own, serving interests other than the rational and altruistic ends for which they were originally, at least publicly, created (Ness & Brechin 1988, p. 269; see Hoeser 2002). Barnett and Finnemore (1999) argue that the integration of technical (not necessarily scientific) knowledge produces autonomy for international bureaucracies. It gives them an authoritative voice when defining the content, direction, and scope of cross-national policies and programs. Other studies concentrate on the inception and involvement of think tanks, epistemic communities, and policy networks by the agencies of the United Nations and multilateral banks, indicating that this development con-tributes to the ascendance of technocratic decision-making in international politics (Haas 1992, 2000; Stone 2000). Yet the way in which the involvement of technocratic bodies reveals the agency of international organizations is left largely unspecified.

The main aim of this study, the research for which was conducted from 2000 to 2003, is to contribute to the understanding of the agency of supra-national organizations by analyzing learning processes that stem from the way in which these types of organizations are embedded within complex and demanding environments. Focusing on the relationships that supranational organizations have with external stakeholders in Member State governments, the economy, civil society, and expert communities as linkages to those environ-ments, the analysis emphasizes how seminal such relationships are to organizational learning. Their variety at the external boundaries of supranational organizations are examined from an organizational learning perspective, an approach that illustrates how these institutions manage what Lequesne (2000) deems to be their main characteristic: the balancing act between autonomy and

15

dependence. The linkages between supranational organizations and external stakeholders are conceptualized as boundary-spanning activities (Böhling 2001, 2002).

To cope with the dearth of literature on the European Commission as a product of its organizational dynamics and with the lack of empirical research on organizational learning, an exploratory research strategy is chosen for this study. It affects the development of the theoretical framework and the empirical research. In the theory chapter (Chapter 2), it meant that the analytical frame-work was in flux throughout the research process, shifting around two basic assumptions:

1. Processes of organizational learning differ depending on the quality of boundary-spanning activities between organizational members and external stakeholders.

2. Organizational learning is shaped by, and possibly shapes, situational factors in the social context of organizations, reciprocality that leads to potential changes in organizational behavior.

The development of the analytical framework benefited from alternating the focus between empirical evidence and theoretical ideas. The research design combines elements of deduction and induction in an effort to refine theory in the organizational learning discourse by providing theoretically informed empirical evidence (Chapter 3). The mode of inquiry is the fine-grained case study because it helps provide sufficient insight into the complex interplay of factors that drive learning processes in organizations in response to their perceptions of environmental changes.

Two case studies are conducted to demonstrate that different qualities of boundary-spanning activities between the Commission and external stakeholders affect organizational learning. The first case is about the consultation with an advisory committee consisting of prominent persons selected for their expertise and experience in the field. The second case is about the interactions with a regulatory and regulated committee consisting of delegates from the EU’s Member State governments (Chapters 4 and 5). The European Commission’s Directorate General for Information Society (DG X) exemplifies a supranational organization. The factors in the focal organization that influence what counts as relevant knowledge from external sources are examined and related to the condi-tions and circumstances that affect the channeling and absorption of the acquired knowledge at the administrative level.

The major findings of the empirical analysis are summarized and system-atically compared in the discussion section (Chapter 6) to provide solid ground for an evidence-based reconsideration of the study’s theoretical framework. It is shown that an element of choice exists in learning processes and that it engen-

16

ders both facilitators and barriers to the incorporation of externally sourced knowledge. Organizational learning through boundary-spanning activities thus cannot sufficiently be understood unless the issues at stake for the focal organization are inquired into.

The concluding chapter (Chapter 7) refers back to the initial aim of this study, which is to shed light on the agency of supranational organizations through the analysis of learning processes. The learning behavior of the Directo-rate General for Information Society indicates the Commission’s identity as supranational technocracy and politicized bureaucracy. It leads to appreciable gaps in Member State control and is wedged into the paradox of continuity and change that is typical of organizational life. The study’s findings can be genera-lized to other cases of the Commission’s Directorate Generals if the policy issues they are addressing are consistent with the EU’s principle of the internal market. Implications for future research conclude this study on the impact that boundary-spanning activities have on organizational learning.

17

2. Organizational Learning from Boundary-

spanning Activities

Bringing the Environment into Theories of

Organizational Learning

Organizations are usually modeled as bounded systems with external boundaries that distinguish them from their environments (Hernes & Paulsen 2003). The idea of organizations as boundary-maintaining activity systems stems largely from Talcott Parson’s influence on sociology, and in systems theory it is a central tenet that has allowed analysis of organizations as entities that interact with their environments (Luhmann 1964). In this realm of thought, boundaries tend to be treated as rigid and stable descriptions of static entities that uphold order and control. The open-organization model by Thompson (1967) is among the first approaches that question the stability of organizations as boundary-maintaining activity systems with fixed boundaries and rules and procedures that are equally adapted to all environments. Empirical evidence indicates that boundaries also play a role in effecting change and transience and that they need to be “conceptualized as subtle descriptions of entities that are dynamic and elusive” (Paulsen & Hernes 2003b, p. 304; see Wiesenthal 1995). Boundaries can therefore be treated as phenomena that give rise to their own dynamics with multiple ways of influencing organizational life (see Adams 1980, Ancona & Caldwell 1992, Jevnaker 2003, Tacke 1997, Tushman & Scanlan 1981).

This chapter offers an exploration of how the dynamic capabilities inherent in the interface between organizations and their environments affect organizational life. In keeping with the focus of the study, the emphasis is on the interactions between organizational members and external stakeholders as a trigger of organizational learning. According to Freeman (1984), a stakeholder in an organization’s external environment is broadly defined as “any group or individual who can affect or is affected by the achievement of the organization’s objectives” (p. 46). A theoretical framework is presented to specify the con-ditions under which the communicational interactions that span the external boundaries of organizations trigger organizational learning. Hedberg (1981) and March (1991) argue that the organization–environment nexus partly determines the way in which organizations transform their knowledge about external changes and developments into the refinement of existing practices and the invention of new ones. To assess the transformation of externally acquired knowledge into an organizational property that helps organizations deal with

18

future situations, organizational learning is defined as a social accomplishment. Learning involves the processing of knowledge in an organization in such a way that it expands the organization’s potential range of behaviors (see Huber 1991, p. 226; Brown 2000, p. 576). It is understood as a metaphorical conception that accents the relationship between the organizing and the social and cognitive processing of knowledge and between organizational action and thought (Gherardi & Nicolini 2001). Learning refers to both the cognitive and social processes of absorbing new knowledge and the outcome, which is the enhanced ability of organizations to cope with changes in their environments (Child & Heavens 2001, Weick 1991, Wiesenthal 1995).

Linking External Changes with Internal

Dynamics in Organizations

The contention that learning is triggered by processes of responding to changes and developments in the external environment is a staple in theories of organiza-tional learning (see Argyris & Schön 1978, p. 29; Hedberg 1981, p. 3; March & Olsen 1975, p. 157). Numerous authors share the basic rationale that dynamic and complex environments require adaptation in organizations and that success-ful adaptation is contingent on effective learning (see Berthoin Antal, Dierkes, Child & Nonaka 2001a; Dodgson 1993; Edmondson & Moingeon 1998; Huber 1991; Popper & Lipshitz 2000; Scherf-Braune 2000; Senge 1990). Although this argument has become mainstream in the organizational learning discourse, the mechanisms in the organization–environment nexus that trigger learning have remained largely unspecified. Klimecki and Thomae (1997, p. 14), for instance, suggest that the mechanisms driving organizational learning be related to the capacity that organizations have for agency, to their sociopolitical dynamics, or to their tendency to reduce complexity.

Gnyawali and Stewart (2003) take a cognitive perspective to bridge the gap in theoretical work on the impact of environmental forces on organizational learning, noting that the organization–environment alignment is often treated as a black box. They show that an organization’s efforts to formulate and imple-ment effective strategies is a function of their ability to create, refine, and validate shared understandings about the nature of environmental events, specifically, environmental changes critical to an organization’s activities. How-ever, Gnyawali and Stewart do not explain how these shared understandings evolve, and their exclusive focus on the perceptions managers have about the environment as the ultimate source of knowing leads them to neglect the fact that organizations are rarely monolithic cultures (Berthoin Antal, Lenhardt & Rosenbrock 2001). Although it is frequently meaningful to look at organizations as cohesive entities, they consist nonetheless of different groups that often have

19

different objectives and evaluate the same outcome differently (Hedberg 1981, Levitt & March 1988).

The significance of perceptual filters for processes of organizational learning is commonly stressed (Dierkes 1988, Dierkes, Hähner & Berthoin Antal 1997, Wiesenthal 1995). Because “organizational learning cannot be created and eradicated by varying external stimuli” (Dodgson 1993, p. 387), emphasis is placed on the communicational structures of organizations, which create for their members a mental box in which to think and work (March & Olsen 1975, March & Simon 1958, Weick & Ashford 2001). Perceptual filters are essential for making sense of the world. In this capacity, they underlie action inside and outside the borders of an organization (Hernes 2003, p. 40). Suggesting that learning results when organizations interact with their environments, Hedberg (1981), for instance, looks at the mechanisms in organizations that select and interpret stimuli and the mechanisms that enable organizations to store and assemble responses. He suggests that organizations influence the learning of their members through cognitive systems and memories, including standard operating procedures, customs, and symbols. According to him, the assembling of responses to external changes and developments requires the creation of meaningful environments that are relevant to the activities of an organization. In that sense, environmental events “sometimes may have little to do with what the organization does” (March & Olsen 1975, p. 153). The novelty of environmental changes is therefore not so much a property of the environment as it is of an organization’s reaction to it (March & Olsen 1989).

Assessing the factors that influence understandings that organizations have of the environment requires examination of the mechanisms that make them knowledgeable about it. Lane (2001), for example, focuses on inter-organizational networks as a trigger of learning because they enable organiza-tions to gain access to new knowledge. She argues that styles of organizational learning vary with the cultural and institutional environment of which organiza-tions are a part. A key mechanism that connects environmental events to individual cognition is the existence of interpersonal relations at the external boundaries of organizations, for such relations afford them opportunities to become knowledgeable about the wider social context they are embedded in (Child & Heavens 2001, Friedberg 1995, Wiesenthal 1995). The “interactive communication behaviors” (Sutcliffe 2001, p. 202) between organizational members and external stakeholders link the internal dynamics of organizations with external events. Comprehension of the organization–environment nexus in terms of communicational interactions implies the assumption that organizations exist within environments that become meaningful to their activities as a con-sequence of the perceptual filters that are used to assess and understand them. Information is therefore a variable in the communication process (Sutcliffe 2001). The beliefs, opinions, and views that organizational members develop

20

from perceiving and interpreting external events, occurrences, and messages indicate how externally sourced information becomes knowledge considered relevant for organizational activities.

The Notion of Boundary-spanning Activities

The linkages between organizational members and external stakeholders depict the openness of organizations to their environments. Theories in organizational learning suggest that increasing the permeability of external boundaries makes organizations more responsive to changing demands and conditions in their environments (Hedberg 1981, Huber 1991, Klimecki & Thomae 1997). More-over, “the ways in which openness to new information and knowledge is achieved . . . establish processes whereby that knowledge is handled within an organization [, and these ways] will impinge on the use made of it” (Child 1997, p. 69). The openness of an organization’s external boundaries therefore not merely brings about the transfer of external signals into the organizational con-text but also serves as an intermediary that affects the extent and direction of organizational learning about the environment (Child & Heavens 2001). If variations in the achievement of openness at an organization’s external bound-aries influence the scope and direction of learning, then criteria are necessary for this process (Tacke 1997, Wiesenthal 1995).

The notion of boundary-spanning activities has been introduced to capture the interactions across an organization’s external boundaries from the per-spective of organizational learning (Böhling 2001, 2002). Organizational boun-daries are defined as the interface between an organization and its stakeholders. This interface consists essentially of activities undertaken by persons (Adams 1980, p. 322, Paulsen & Hernes 2003b, p. 306). Boundary-spanning activities are conceived of as a particular kind of social practice at the external boundaries of organizations: the kind that triggers learning processes. To spell out Hedberg’s (1981) assertion that acting is the means of acquiring knowledge (p. 5), such activities are understood in this study to be a means of acquiring knowledge about changing conditions and developments in the environment. Integrating the notion of boundary-spanning activities with theories of organiza-tional learning contributes to a differentiated understanding of the impact that the organization–environment nexus has on organizational learning. In this line of thought, variations in the characteristics of boundary-spanning activities affect how organizations learn, for what counts as relevant knowledge and how it crosses the external boundaries of organizations depends on the actors involved and the nature of their dialogue. Crozier and Friedberg (1980, p. 81), for instance, note that the interaction of organizational members with external stakeholders varies in the way it is regulated and linked with intraorganizational

21

processes. Boundary-spanning activities may therefore be seen as more or less formalized interactions serving specific goals and embodying particular rights and obligations. Certain underlying orientations and axiomatic assumptions guide the behavior of the organizational members and external stakeholders.

Through the imagery of organizations as open systems, the notion of boundary-spanning is elaborated to stress that interactions with external stake-holders affect the behavior of organizations. Adams’s (1980) classification of different boundary activities (e.g., input-filtering, the search for information, and the representation of organizations to important groups in their environments) underscores the idea that transforming input from external sources into output within organizations has impacts on their environments and feedback effects on the organizations themselves. Tushman and Scanlan (1981) look at boundary-spanning in terms of information, bringing out the role of individual organiza-tional members in spanning the “communication boundary” (p. 290) between external areas and the internal information needs of the organization. They point out that those individuals must be well connected internally and externally and have to be knowledgeable of “local coding schemes and languages as well as the specialized conceptual frameworks” (p. 291) in order to translate the acquired information so that it is relevant to the organization. Ancona and Caldwell (1992) develop a typology of the external communication behavior of organiza-tions to show that the behavioral patterns of external activities rather than the frequency of communication affect organizational performance.

The ideas associated with the concept of boundary-spanning can be traced back several decades. The term “boundary spanning units” was coined by Thompson (1967, p. 170) and “boundary spanning roles” by Aldrich (1979, p. 243) to explore the conditions under which organizations adjust their structures and activities in response to environmental changes. Drawing on Thompson’s (1967) contention that organizational boundaries need to be conceptualized as permeable entities and as barriers to protect the internal equilibrium of an organization from external disruption, Aldrich (1979) argues that “the existence of boundary spanning roles is an important guide to environ-mental contingencies perceived as significant by authorities” (p. 243). He ap-plies Thompson’s underlying reasoning of the organization–environment alignment (structural isomorphism) to the rise of boundary-spanning activities, relating these processes to environmental characteristics and organizational size. “Environmental pressures or demands are responsible for much of the observed boundary role differentiation in organizations . . . As environments become more complex or differentiated, so do the relevant boundary roles” (Aldrich 1979, pp. 255–256). Heterogeneity and instability in the environment thus help explain the generation and formalization of boundary-spanning roles. As environments become more heterogeneous and unstable, organizations tend to assign the boundary-spanning roles to functionally differentiated units in an

22

effort to cope with the perceived complexity of the environment (Thompson 1967). They are more common in large organizations with a high degree of internal differentiation and specialization. As Thompson (1967) points out, large organizations operating in heterogeneous environments “seek to identify homo-geneous segments and establish structural units to deal with each” (p. 70). The monitoring of changes through boundary-spanning roles is more critical to organizational survival in unstable environments than in stable environments.

Along with the functional differentiation of boundary-spanning roles, the degree of their routinization is another major characteristic of their social consti-tution. Thompson (1967) points out that organizations try to control the behavior of their boundary-spanning personnel because they tend to develop attitudes that are similar to the external stakeholders they are dealing with (see also Crozier & Friedberg 1980). To perform the role of boundary spanner successfully, the incumbent must be able to represent to other members of his organization the expectations, preferences, values, and norms of external groups. However, by appreciating and understanding outside groups, boundary spanners can come under suspicion of loyalty to those groups (Adams 1980). Standard operating procedures and other limits to the discretion of boundary spanners are therefore employed to reduce the risk that they will exhibit attitudes and behaviors inconsistent with their organization’s objectives. Conversely, a minimal degree of routinization is necessary so as to enable incumbents of boundary-spanning roles to deal with exceptional cases in heterogeneous environments. As Aldrich (1979) explains, “the degree to which boundary roles are routinized is thus a function of both pressures to adapt to environmental contingencies and con-straints and the necessity of controlling the behavior of potentially deviant members” (p. 260). The isomorphic relationship between the environment and the social constitution of boundary-spanning roles is expressed in Table 2.1.

Table 2.1 Social constitution of boundary-spanning roles in changing environments

Environment Stable Dynamic

Homogeneous Standardized with prescribed limits

Centralized

Flexible with discretionary content

Centralized

Heterogeneous Standardized with prescribed limits

Functionally differentiated and decentralized

Flexible with discretionary content

Functionally differentiated and decentralized

23

The concept of boundary-spanning activities is relevant to theories of organizational learning when the environment is treated as a source of informa-tion. Boundary spanners gather intelligence about environmental changes and developments and make it available to other members of the organization. Aldrich (1979) points out that boundary spanners must be quite skilled in filter-ing out external events that may be relevant for the organization’s objectives and must summarize the acquired knowledge in terminology other members can understand. Acting as filters and facilitators, boundary spanners protect the organizational core from information overload and contribute to adjustment and renewal in the organization by making it aware of new developments in the environment. The influence that boundary spanners have on the channeling of knowledge from external sources into organizations is also acknowledged by Child and Heavens (2001). They note that “boundary spanners who work at the interface between their firm and its external environment play a critical role in the process of transferring information into an organization” (p. 320).

In addition to processing information, boundary spanners may also assume a function as representatives to the outside world. When boundary span-ners represent their organization in the environment, the flow of information is essentially outward. To represent their organization in the environment and mediate between it and important outside groups, boundary spanners provide information about the organization, its operations, and its management to the external environment, partly as a result of external demand and partly on their own initiative (see also Adams 1980). The provision of information to external client groups that are specifically adapted to the organization generates social support and legitimacy (Aldrich 1979). Ancona and Caldwell (1992) describe the external representation of boundary-spanning as “molding” because they see it as an organization’s attempt “to influence the external environment to suit its agenda by shaping the beliefs and behaviors of outsiders” (p. 638). Adams (1980) points out that organizations are selective in the transmittal of information about themselves because it makes them “vulnerable to competitors and to the manipulation of other organizations” (p. 345). However, the literature on boundary-spanning is not clear about how to achieve selectivity in the trans-mittal of information to external groups. The environment is not treated as a source of information that needs to be tapped for the purpose of representing the organization. Adams (1980) points to boundary spanners’ dependence on other organizational members to furnish important outside groups with information that helps generate external support. Tushman and Scanlan (1981) stress that the formalized purpose of the representational role—to focus on resource acquisition and disposal—reduces the incumbent’s “sensitivity to exceptions, opportunities, or new ideas from outside the organization [and may therefore] not serve as an information linking mechanism for areas within the organization” (p. 301).

24

From the angle of organizational learning, though, the processing of infor-mation on external demands and expectations depends upon prior understand-ings in the organization about external stakeholders and is thus arguably a precursor to the function of external representation. The information that accrues to an organization through boundary-spanning reduces uncertainty and may then become a means of gaining control over relevant entities in the environment. Huber’s (1991) stage model of organizational learning, for example, integrates the information-processing and external representation views on boundary-spanning. Assuming that the need for environmental fit is fundamental to organizational survival, Huber sees the employment of boundary-spanning personnel as contributing to the alignment. Boundary spanners may gain access to knowledge about critical contingencies in organizational environments and may channel it into the organization. As sensors of the environment, they may acquire information for enhancing the effectiveness of strategic management and may play the role of external gatekeeper in their organization. In addition, environmental scanning through boundary-spanning personnel may contribute to the monitoring of an organization’s performance. As a result of the external relations that boundary spanners have, they are able to assess how well the organization is meeting both its own standards and the expectations of external stakeholders. These contacts open possibilities for influencing the choice of the issues to address when external bodies assess how well the organization is doing. Organizations may acquire information about external demands and expectations and may process it to modify their behavior toward the stake-holders whose support is needed.

Processes of Organizational Learning

The availability of new knowledge regarded as relevant for an organization’s activities as a result of talking to external stakeholders may trigger organiza-tional learning but does not explain it. The absorption of the gained knowledge is also influenced by a number of other factors in organizations, including their structural, cultural, and regulatory properties and organizational politics. Distinctive challenges and opportunities are accompanied by learning processes that are informed and stimulated by interpersonal relations between organi-zational members and external stakeholders. They come to the fore when organizational learning is seen as a process. Processes of learning can be conceived of in a number of different ways. In the traditional paradigm, learning is treated as a stepwise process focusing on how knowledge is acquired, distributed, and stored in the memory of organizations (Berthoin Antal 1997, Berthoin Antal et al. 2001a, p. 5, DiBella, Nevis & Gould 1996, Huber 1991).

25

One of the most comprehensive overviews of the distinct stages in learn-ing is Huber’s (1991) aforementioned stage model of organizational learning. Synthesizing the work of different research groups, it is a response to what he perceived as a lack of cumulative and consolidated work in the organizational learning discourse (see also Gnyawali & Stewart 2003, Schreyögg & Eberl 1998, Wiesenthal 1995). His model, which postulates four elements (see Figure 2.1), suggests that learning does not occur simply by organizing the acquisition of new knowledge (see also DiBella et al. 1996). It encompasses internally based and externally oriented strategies for identifying new know-ledge and disseminating it widely and deeply enough for action to be taken. Having the four different steps in mind when looking at organizational learning makes the point that it “is neither an effortless nor an automatic process” (Berthoin Antal, Lenhardt et al. 2001, p. 865).

This illustration of the four distinctive learning stages may be misleading, for it seems to suggest simple linearity from one step to the other. In reality, learning does not follow the neat progression implied by Huber’s stage model, or any other. Instead, it entails feedback loops and iterative movements (Berthoin Antal 1997). Yet the model is useful for the systematical assessment of the role that each step has and of the ways in which all the steps are related in processes of organizational learning. It breaks the encompassing phenomenon of organizational learning down into a number of relatively small and observable processes (Büchel & Raub 2001). Following the line of thinking set out by Huber (1991) may then help explain why learning processes are often inter-rupted or incomplete (March & Olsen 1975). Processes of organizational learn-ing are incomplete when interpretations of environmental changes are divorced from action and therefore of little or no effect on organizational behavior.

26

Figure 2.1 Constructs and processes associated with organizational learning. From

“Organizational Learning: The Contributing Processes and the Literatures”, by G. P. Huber,

1991, Organization Science, 2(1), pp. 88–115, p. 90.

1.0 Knowledge

Acquisition

2.0 Information

Distribution

3.0 Information

Interpretation

4.0 Organizational

Memory

1.1 Congenital Learning

1.2 Experiential Learning

1.3 Vicarious Learning

1.4 Grafting

1.5 Searching and Noticing

3.1 Cognitive Maps and Framing

3.2 Media Richness

3.3 Information Overload

3.4 Unlearning

4.1 Storing and Retrieving Information

4.2 Computer-Based Organizational Memory

1.2.1 Org. Experiments

1.2.2 Org. Self-appraisal

1.2.3 Experimenting Organizations

1.2.4 Unintentional or Unsys-tematic Learning

1.2.5 Experience-based Learning Curves

1.5.1 Scanning

1.5.2 Focused Search

1.5.3 Performance Monitoring

Constructs and

Processes

Subconstructs

and Subprocesses

Subconstructs

and Subprocesses

27

Stages in Processes of Organizational Learning

Huber’s (1991) learning model commences with an examination of the pro-cesses through which an organization acquires knowledge from internal or external sources. Knowledge is often picked up tacitly and unintentionally rather than in a mindful and systematic way, and that difference is likely to affect the other stages of learning. In Huber’s detailed analysis, the variety of processes and mechanisms by which organizations can acquire knowledge encompasses congenital learning, experiential learning, grafting, vicarious learning, searching and noticing, and scanning. Congenital learning refers to the behaviors and practices engrained in an organization by its founders. Organizations may also learn by drawing on the direct experiences of their members. Vicarious learning is about transferring second-hand experience about strategies, approaches, and practices of other organizations and adapting them to the circumstances in the organization. Grafting refers to the knowledge to which an organization may gain access through the recruitment of new personnel. Processes of searching and noticing refer to additional mechanisms that contribute to the acquisition of knowledge from the environment. Organizations scan their environments for information on changes relevant to their objectives and operations, or they inter-act with external constituencies and stakeholders to assess how well they are meeting both their own standards and external expectations.

The second construct relates to information distribution as the process by which the knowledge gained is diffused among the units of the organization. The dissemination process is a determinant of the “occurrence” and “breadth” (Huber 1991, p. 100) of organizational learning. According to this model, learn-ing occurs when information distribution generates an awareness in organiza-tional compartments that there are potential synergies between different items of knowledge. New knowledge is commonly created by piecing together the obtained information with other items of knowledge. Huber contrasts informa-tion distribution that leads to “new organizational learning” with information distribution that leads to “more broadly based organizational learning” (p. 100, italics in original). The breadth of learning relates to a problem frequently found in organizations—the fact that they tend to know less than their individual members. The distribution stage leads to broadly based learning if the informa-tion is widely shared by organizational compartments and supported by retrieval systems that not only store knowledge but also help bring together across departmental boundaries those who possess knowledge and those who need it.

Because coupling knowledge from different sources may create not only new knowledge but also new understandings, the distribution stage is seen as a precursor to information interpretation. The diffusion of information is therefore linked with information interpretation, which is the third stage in Huber’s (1991) model of organizational learning. The interpretation process underlines the

28

development of shared understandings among organizational members and compartments about distributed information. In Daft and Weick (1984), interpretation is defined as “the process through which information is given meaning” (p. 294) and as “the process of translating events and developing shared understandings and conceptual schemes” (p. 286). The development of shared understandings about occurrences and events enables organizations to enhance cooperation and coordination and therefore to increase the range of their potential behaviors.

Discussing what affects the formation of shared understandings in organi-zations, Huber (1991) suggests a number of factors, including cognitive maps, information overload, and unlearning. A cognitive map is a specific way of representing a person’s beliefs about a limited domain and consists of concepts and the relationships between them (Axelrod 1976, Weick & Bougon 1986). Huber (1991) asserts that cognitive maps vary across organizational compart-ments, leading to differences in the way that distributed knowledge is framed or labeled. Information overload is an additional factor that complicates the forma-tion of shared understandings. It occurs “if the information to be interpreted exceeds the unit’s capacity to process the information adequately” (p. 105). The load of information on organizational units can be reduced through unlearning, which is the process through which organizational members discard knowledge if it is seen as obsolete or misleading (Hedberg 1981).

Lastly, organizational memory is the means by which knowledge is stored for future use. A great deal of the organization’s knowledge about how to do things is stored in standard operating procedures, routines, and scripts. Under-stood as ingredients of an organization’s structure and culture, they affect how the knowledge acquired from internal and external sources is stored in the organization’s knowledge base and then brought forth from it. Organizational memory can therefore be defined as the “repository of organized knowledge” (Walsh 1995, p. 286) that acts as a template for interpretation and action. Huber (1991) addresses the critical role of organizational memory in learning pro-cesses, covering two points. First, the “demonstrability and usability of learning depend on the effectiveness of the organization’s memory” (p. 104). Huber emphasizes the rise of computer-based organizational memories in this respect, believing that their accessibility and reliability make them superior to inter-subjective memories. Second, the process of organizational learning is indefinable without reference to the concept of organizational memory. Organizational memory indicates that the different stages of the process model are all interrelated because they are affected by previous learning, criteria used in decision-making, and frames of reference.

29

Weaknesses of Conceptualizing Organizational

Learning in Terms of Stages

Huber’s (1991) stage model of learning emphasizes the processing of informa-tion. But it directs little attention to the circumstances and conditions that pro-vide and sustain the context of learning and does not address learning barriers that stem from an organization’s structure, culture, and leadership (Berthoin Antal, Lenhardt et al. 2001). Yet the context of learning affects the quality of the learning process. Learning in organizations is not merely a cognitive phenol-menon involving a number of individuals; it is a social activity in which “any modification of the knowledge distribution is perceived as a way of subverting the established knowledge/power relations within a social context” (Gherardi, Nicolini & Odella 1998, p. 276). Organizational learning is not an automatic process but rather an accomplishment entailing cognitive and practical activities and influenced by organizational politics. The conceptualization of organi-zations as information-processing systems, which underlies Huber’s (1991) model, tends to treat information or knowledge as an objective, transferable good and rather neglects the symbolic connotations attached to it. As Child and Heavens (2001) note, “the significance of information for organizational learn-ing is not just what it literally says but also where it comes from and how its social implications are interpreted” (p. 322).

Taking knowledge acquisition as a starting point also appears to divorce organizational learning from the social context in which it occurs. Organizations do not learn at random. They learn because key individuals such as managers or administrators may see a need to do so. External experts, too, can play an impor-tant role in awakening the interest to learn. As Berthoin Antal and Krebsbach-Gnath (2001) conclude from their research on the specific roles played by con-sultants in organizational learning, knowledge acquisition is not the first step in learning processes. It is preceded by the awareness of a need to learn, by a willingness to learn. The contribution that consultants make to this step consists in the “challenging of assumptions and reframing of past solutions as inap-propriate for current problems” (p. 472). Kerlen (2003) closely examines the ways in which consultants and members in the client organization identify and agree on a definition of the issues to be addressed. She sees these processes as predecessors to the knowledge acquisition phase in the stage model. She shows that learning processes in organizations are based in the organization, not triggered exclusively by the outside consultant’s view. Groups within the organization engage consultants to create opportunities for tackling urgent issues or problems.

Although Huber (1991) gives a very detailed and elaborate account of the various processes and subprocesses in knowledge acquisition, his approach to information distribution and interpretation is less comprehensive. An important

30

reason for this weakness may be that he uses the terms information and knowledge interchangeably. By contrast, other scholars such as Nonaka and Takeuchi (1995) distinguish information as “a flow of messages, while know-ledge is created by that very flow of information, anchored in the beliefs and commitments of its holder” (p. 58). Gnyawali and Stewart (2003) hold that information becomes knowledge to an organization if interpreted and put into a meaningful context. Nonaka and his colleagues look at interaction, cooperation, and communication as crucial elements of knowledge creation in organizations. Knowledge is a dynamic asset, “for it is dynamically created in social inter-action between individuals both within and across organizations” (Nonaka, Toyama & Byosière 2001, p. 493). From this angle, social interaction becomes a central driving force behind learning processes. However, the underlying view of organizational knowledge creation as “a never-ending process that upgrades itself continuously” (p. 498) fails to take account of an organization’s inherent sociological factors that impinge upon learning processes. Nonaka and his col-leagues speak about knowledge creation through action and interaction (Nonaka et al. 2001, p. 492), but there is little treatment of the relationship between action and knowledge (see also Nonaka & Takeuchi 1995). Moreover, Nonaka and his team analyze processes of knowledge creation to the neglect of debate and negotiation as constituent elements of sense-making.

Toward a Contextualized Model of

Organizational Learning

To overcome the shortcomings of Huber’s (1991) stage model of organizational learning, it is necessary to situate learning processes within organizations as sociopolitical systems (Child 1997) and therefore attend to the specific context of their activities and social practices that sustain learning (Gherardi et al. 1998; Gherardi & Nicolini 2001). Granovetter’s (1985) notion of organizations’ “embeddedness” is a helpful heuristic for placing organizational learning in its social setting (Geppert 2000). It highlights networks of personal relations both within organizations and across their external boundaries as a way of explaining attempts that their members make to engage in purposive action. Organizational behavior is conceptualized in terms of social relations where the actors are neither atoms nor slavish adherents to a script. The concept of embeddedness implies that learning from boundary-spanning activities indicates the agency of organizations in constraining environments.

To explore the idea that organizational learning is embedded in the social context of organizations, this section presents a model that has boundary-spanning activities as the point of departure. Integrating boundary-spanning activities and theories of organizational learning helps embed learning pro-

31

cesses, for boundary-spanning activities underscore the social context of communicational interactions at an organization’s external boundaries and of the linkages between those interactions and internal dynamics. The learning approach to boundary-spanning addresses current debates about the nature and locus of organizational learning, emphasizing that “learning occurs, and know-ledge is created, mainly through conversations and interactions between people” (Easterby-Smith, Crossan & Nicolini 2000, p. 787, italics in original).

This section’s discussion of the impact that boundary-spanning activities have on each process of learning should illuminate the conditions and circumstances that affect the incorporation of externally acquired knowledge and lead to distinctive qualities of learning about how to cope with changes in the environment. It should help go beyond Aldrich’s (1979) claim that the boundary spanner’s creation of organizational intelligence tends to be accepted by the other members of the organization because “recipients of summarized information are in no position to check the original sources and therefore must trust the ability of boundary-spanners to correctly interpret the raw information” (p. 250). The organizational learning perspective on boundary-spanning activities suggests that other members of the organization will not necessarily accept inferences drawn about the environment during boundary-spanning activities. Against this background the full version of Huber’s (1991) definition cited in the introductory chapter can now be understood: organizational learning from boundary-spanning activities is, hence, defined as the creation of organiza-

tional knowledge that is a social accomplishment involving the distribution and interpretation of externally acquired knowledge and its integration with organizational memory. The contextualized stage model of learning is based on two assumptions:

1. Processes of organizational learning differ, depending on the quality of communicational interactions that evolve between organizational members and external stakeholders.

2. Organizational learning is shaped by, and possibly shapes, situational factors in the social context of organizations, reciprocality that leads to potential changes in organizational behavior.

Knowledge Acquisition through

Boundary-spanning Activities

The emphasis on boundary-spanning activities as a means of acquiring know-ledge about changes and developments in the environments of organizations draws attention to the processes whereby information from external sources becomes relevant knowledge for organizations. Boundary-spanning activities

32

filter knowledge regarded as relevant for organizational activities. The know-ledge acquired through boundary-spanning activities is relational, context-specific, and anchored in the intersubjectively shared beliefs of those acquiring and interpreting it (Gherardi et al. 1998). The filters entailed in boundary-spanning activities “enable some things to be seen more clearly by blending out others” (Berthoin Antal, Lenhardt et al. 2001, p. 868). They affect how informa-tion from external sources is perceived and interpreted by those involved (Aldrich 1979, p. 122). The filtering of boundary-spanning activities illustrates the duality of an organization’s external boundaries, which must simultaneously be maintained and kept permeable (Strassheim 2004, Wiesenthal 1995). The filtering limits the permeability of those external boundaries. It helps organiza-tions maintain a common corporate culture and identity by constraining the transfer of potentially contradictory information. To assist this effort, organiza-tions may respond by developing a Leitbild (Dierkes et al. 1997, Dierkes 1997, Kuusi & Meyer 2002). A Leitbild is a guiding vision that articulates an overall goal, offers orientation, and provides a basis for different groups in the organi-zation to work in the same direction. It may be seen as both an outcome and a trigger of organizational learning. In either case, the achievement of internal support for such a generalized perspective on current or future tasks is crucial for orchestrating collaborative action in the long run (Dierkes & Marz 1998, Dierkes, Marz & Teele 2001).

The social constitution of boundary-spanning activities shapes the way that externally sourced information is filtered and channeled into organizations (Aldrich 1979, Child & Heavens 2001). The literature on boundary-spanning activities suggests that an organization’s knowledge-seeking behavior in these activities varies with the perceived goals and underlying orientations that drive this process and with its regulation (Aldrich 1979, Crozier & Friedberg 1980, Thompson 1967). Highly formalized interaction with external stakeholders tends to stifle innovation and predefine the framework for monitoring the environment and handling potential gains in knowledge (see Jevnaker 2003). The regulation of boundary-spanning activities according to standardized rules and procedures leaves little maneuvering room to the actors who are trying to influence the terms for defining relevant knowledge and the ways it is channeled into the organization. Conversely, flexibility in the regulation of the interaction with external stakeholders gives the actors space (discretion) to draw inferences from environmental monitoring and provide them to the organizational members who are not directly involved in the boundary-spanning activities. More research is necessary to reveal further insights about the relationship between the filtering inherent in boundary-spanning activities and the transfer of the knowledge thought to be relevant for organizational activities across the external boundaries of organizations.

33

The boundary-spanning literature also suggests that activity at the inter-face between organizations and the environment may affect the power relations of the organizations. Discretion and expertise in interpreting external events and occurrences for other organizational members is a source of power for boundary-spanning personnel (Thompson 1967). Power accrues to boundary spanners because they absorb uncertainty “by selectively paring away the world’s native complexity” (Gordenker & Saunders 1978, p. 87). Moreover, boundary spanners have substantial power and status because they mediate a critical resource for organizational survival (Tushman & Scanlan 1981). The importance of the boundary spanner’s information-processing function increases when environments are perceived as heterogeneous and unstable (Aldrich 1979). Depending on the perceived importance of the information-processing function in those activities, the knowledge gained may be either recognized as a relevant contribution to decision-making processes or drowned in the loads of infor-mation the focal organization is trying to cope with. Flexibility in the regulation of boundary-spanning activities thus tends to increase the ability of boundary spanners to derive power from their access to knowledge deemed relevant for the organization and from their influence on its dissemination.

Because a willingness to learn is crucial to the development of learning processes, boundary-spanning activities have to relate to the rise of organiza-tional learning needs. The propensity of organizational members to draw on their relations with external stakeholders as a source of new knowledge about the environment is greater if there is a need to learn than if there is no such need. The notion that organizations need to exhibit a willingness to learn to engage in learning processes implies that their internal context sets the stage for their knowledge seeking behavior. The extent to which organizational learning can be stimulated and informed by external developments depends on an organization’s perception that strategies and operations need to be adjusted to safeguard survival or that expectations of important outside groups need to be identified to maintain the organization’s legitimacy.

Boundary-spanning Activities and

the Distribution of Knowledge

The importance of distributing knowledge acquired in processes of organiza-tional learning is acute. Huber’s (1991) model suggests that knowledge distri-bution determines the occurrence and breadth of learning in organizations. Hence, the impact of boundary-spanning activities is assessable in terms of two dimensions: (a) the degree to which they affect the creation of new knowledge about the environment and (b) the extent to which they facilitate broadly based understandings about the environment among the groups and units within the

34

organization. The occurrence and breadth of learning are not necessarily related. An organization may develop new perspectives on its environment or change its responses to external changes by restricting the knowledge-sharing process to top management (Stopford 2001). To understand the impact of boundary-spanning activities on the distribution of knowledge, it is worthwhile to look at their linkages with intraorganizational processes (see Crozier & Friedberg 1980). Boundary-spanning activities affect multiple decisions a priori or during processes of organizing (Jevnaker 2003).

The regulation of the linkages between boundary-spanning activities and intraorganizational processes determines how and how much the different groups in an organization gain access to externally acquired knowledge. Shrivastava (1983), for instance, notes that an organization’s regulatory system controls the flow of information among organizational members by guiding “exactly which information goes to whom and for what purposes” (p. 23). The rules that govern those linkages are understood as “widely shared notions about which behavior is appropriate under certain conditions. . . . When a certain rule applies, the person or the group activates a number of specific routines” (Kieser, Beck & Tainio 2001, p. 601). Organizational rules and routines are understood as “repetitive, recognizable patterns of interdependent actions, carried out by multiple actors” (Feldman & Pentland 2003, p. 95). Rules in practice may evoke routines, and routines may crystallize into rules. The rules and routines inscribed in the boundary-spanning activities and the way they are linked with intra-organizational processes may limit the access that other units have to the acquired knowledge. The impact of an organization’s regulatory system on the distribution of the knowledge gained through boundary-spanning activities varies with the degree of that system’s formalization. The more formalized the linkages are between boundary-spanning activities and intraorganizational processes, the less choice there is for the actors to determine how the ensuing communication behaviors are to be carried out and who gets what kind of knowledge (Fox 1974).

Even when the linkages between an organization’s boundary-spanning activities and internal processes are highly formalized, the distribution of know-ledge is not an automatic process. The rules and routines that govern knowledge distribution may “exhibit considerable variety and require effort from anyone intent on engaging in them successfully” (Kieser et al. 2001, p. 600). The involvement of different people in boundary-spanning activities makes it questionable whether rules and routines unfold the same way every time (Pentland & Rueter 1994). People bring in their own attitudes, focal points, and interests, all of which may vary over time and color the distribution and recap-tion of knowledge triggered by boundary-spanning activities.

Another important source of flexibility in the regulation of knowledge distribution is the adjustments and improvisations that people undertake to make

35

rules and routines work (Feldman & Pentland 2003). Agency in the distribution of externally acquired knowledge implies that organizational members may use the governing principles to influence the knowledge-sharing according to their interests. Restrictive use of these principles inhibits the wide dissemination of the knowledge deemed relevant for organizational activities, impeding the ability of the overall organization and its different internal groups to learn. Broadly based learning is improbable if the externally acquired knowledge is channeled mainly to the groups whose attitudes are similar to those of groups directly involved in the boundary-spanning activities.

Boundary-spanning Activities and

the Interpretation of Knowledge

Boundary-spanning activities are selective in the knowledge they pass on. The filtering inherent in these activities indicates that organizational learning is not merely about acquiring and distributing new knowledge but also about giving meaning to it. Because information from external sources becomes relevant to organizations only when put into a meaningful context, boundary-spanning activities may be viewed as inducing interpretation processes in organizations. People involved in boundary-spanning activities affect the relevance of the acquired knowledge, but they do not establish its meaning once and for all, particularly in organizations that are trying to accomplish broad organizational learning. Recall that organizations tend to be anything but monolithic cultures and defy treatment as unitary actors (Berthoin Antal et al. 2001b, Crozier & Friedberg 1980, Levitt & March 1988, Sutcliffe 2001). Organizations are seen as “the result of an intense activity of assemblage, boundary-making and identity-preserving” rather than as stable entities with defined boundaries (Easterby-Smith et al. 2000, p. 791). They consist of various groups and cultures with different goals, frames of reference, and interests that may lead to difficulties in understanding each other’s terminology, metaphors, or stories. Groups within the organization may not look at things from the same angle. They may follow unlike objectives, and their axiomatic assumptions may differ. Merely exposing an organization to what is perceived by some as relevant knowledge is therefore insufficient without efforts to internalize it (Kim 2001). The fact that organiza-tions have multiple realities and practices that are not necessarily coherent can make the interpretation of externally acquired knowledge across an organiza-tion’s groups an essential ingredient of learning (Weick & Ashford 2001).

As Huber (1991) points out, interpretation consists not merely in giving meaning to information and showing the relevance that external occurrences and events have for the organization’s strategies and operations but also in develop-ing mutual or shared understandings across an organization’s units. To absorb

36

new knowledge, organizations must relate the newly acquired knowledge to the routines and orientations that guide their actions. Knowledge acquisition through boundary-spanning activities develops what Cohen and Levinthal (1990) call an organization’s “absorptive capacity.” They explain that “the ability to evaluate and utilize outside knowledge is largely a function of the level of prior related knowledge” because such knowledge “confers an ability to recognize the value of new information, assimilate it, and apply it” (p. 128).

The absorption of knowledge that boundary spanners gain from their interaction with external stakeholders is associated with its interpretation across an organization’s internal boundaries. The politics of organizational knowledge play an important role in the way the internal barriers are bridged. Sense-making in organizations is not a purely democratic process; leaders are given a special prerogative (Coopey 1995, Czarniawaska-Joerges 1996, Filion & Rudolph 1999). Through control of strategic planning, senior management in organiza-tions may strongly influence not only the perceptions that insiders have of the external environment but also the reciprocal perceptions that outsider’s have (Coopey 1995, p. 209; Stopford 2001). Organizational members have different opportunities and follow different interests in determining the relevance that the acquired knowledge has for action. The close relationship between power and knowledge (Hardy & Clegg 1996) is also a salient issue because “key members of the organization have a stake in maintaining the structures and power relations that are advantageous to them” (Berthoin Antal, Lenhardt et al. 2001, p. 866). The incorporation of new knowledge about environmental changes and developments may call for a modification or at least a reconsideration of an organization’s behavior toward its environment. It may also defy or even threaten the internal status quo.

The establishment of shared understandings about the relevance that environmental knowledge has for organizational activities is a particularly pres-sing issue in organizations with different sets of boundary-spanning activities. Child (1997), for example, asserts that the multitude of social networks across an organization’s external boundaries “adds grist to the mill of internal political debate and negotiation” (p. 69). Because of their boundary-spanning activities, then, organizations may internally reproduce conflicts that originate in their environments. The likely proliferation of goals and actions of organizational groups complicates collaboration. Because learning processes “take place in a landscape of interests and differential power positions and relations” (Easterby-Smith et al. 2000, p. 793), the difficulties in achieving collaboration across an organization’s internal boundaries can be eased if the boundary spanners build coalitions with senior management. Such high-level access may encourage groups across an organization to adopt perceptions that boundary spanners have of environmental changes.

37

Hardy’s (1996) distinction between four dimensions of power is useful for detailing the interaction of boundary-spanning activities and the politics of organizational knowledge in interpretation processes (see Table 2.2). At a general level, Hardy defines “power” as a force that affects outcomes and “organizational politics” as power in action. Like Lukes (1974), Hardy distin-guishes resources, processes, and meaning as dimensions of power but includes the system dimension as well, which she defines as “the unconscious acceptance of the values, traditions, cultures and structures of a given institution . . . [It] is the backdrop against which all actions and decisions take place” (Hardy 1996, p. S8) She focuses on strategic change, arguing that it does not occur in a vacu-um but rather “takes place in a system in which a certain distribution of power is already entrenched” (p. S9). Viewing the interpretation of knowledge in learning processes as a precursor to potential changes in an organization’s behavior implies that it is driven by organizational politics.

Table 2.2 Sources of power and how they are acted upon in organizations

Dimensions of power

Location of

power Resources Processes Meaning System

Source of

power

Information, expertise, credibility, prestige, access to senior management, control of rewards and sanctions, etc.

Decision-making processes, participants and agendas, etc.

Symbols, rituals, language, etc.

Unconscious acceptance of values, traditions, cultures, and structures

Action of

power

Principles of behavior modification are used to influence specific actions.

New awareness is created by opening up processes to new participants, issues, and agendas.

Change is given new meaning, making it appear legitimate, desirable, rational, or inevitable.

Status-quo prevents change by capturing all organizational members in its web.

Adapted from “Understanding Power: Bringing about Strategic Change,” by C. Hardy, 1996, British Journal of

Management, 7 [Special Issue], S3–S16.

Organizational members draw on resources such as information, political access, credibility, and access to higher echelon members to influence decision-making and bring about desired behavior. This dimension of power works along the lines of behavior modification and resembles LaPalombara’s (2001, p. 562) definition of power as the ability of one person to make another do his or her bidding. The second dimension of power is that of processes. It is about the pro-cedures and routines of decision-making processes that may be invoked by dominant groups to determine outcomes by preventing other members of the organization from participating fully in decision-making. The factor that is used to shape perceptions, cognitions, and preferences so that organizational mem-

38

bers accept the status quo is power’s third dimension, meaning. It is embedded in the symbols, rituals, and language of an organization and is central to questions of choice in sense-making, that is, whether choice is perceived as determined by dominant groups or as an outcome of an inclusive democratic process (Coopey 1995). The fourth dimension of power, the system, stabilizes organizations, enables them to carry out their activities, and may impede learn-ing in some cases. Because this form of power is taken for granted, it is often beyond the reach of organizational members. Hardy argues that organizational members may change the status quo and, hence, transform the system if they act upon the other three dimensions of power.

On these grounds it is asserted in this study that organizational politics assume different forms and may therefore variously affect the sense-making of acquired knowledge. Organizational members draw on the different sources of power when they interpret the knowledge gained through boundary-spanning activities. Accordingly, the boundary spanner’s provision of new knowledge considered relevant for organizational activities affects, and is affected by, the resource dimension of power when the knowledge feeds into decision-making. Boundary spanners may play a vital role in decision-making by demonstrating that their perceptions about environmental changes matter for decisions to be taken. If boundary spanners and senior management come to hold similar views on environmental changes, their perceptions may appear to be significant. This kind of coalition illustrates how the meaning dimension of power relates to interpretation processes. It tends to shape perceptions of other organizational members.

The development of shared understandings about the acquired knowledge also depends on the regulation of the decision-making process where it comes to bear. Because organizational rules and routines guide decision-making pro-cesses, they determine which groups are involved at which stage and to what extent they may influence the outcome of the decision-making. The regulation of the decision-making where the knowledge gained from external sources feeds in tends to reinforce and reproduce the underlying structures and cultures of an organization. The system dimension of power does not fully determine sense-making processes in decision-making, for rules and routines are both a source of stability and a means of generating change (Feldman & Pentland 2003, Granovetter 1985). Like other social phenomena, rules and routines embody a duality of structure and agency (Giddens 1984). Though sense-making of the acquired knowledge is carried out against the backdrop of rules and expec-tations, “there is always the possibility of resisting expectations and doing otherwise” (Feldman & Pentland 2003, p. 102).

39

Boundary-spanning Activities and

Organizational Memory

The importance of an organization’s regulatory aspects for acquiring, distri-buting, and interpreting the knowledge gained from boundary-spanning activities indicates the impact that organizational memory has on learning processes. In addition to the norms and forms of an organization, its rules and routines play the role of memory because they are a way to encode knowledge and capabilities for future use (Berthoin Antal 2000, Huber 1991, March 1991). An organization draws on its memory to “preserve certain behaviors, mental maps, norms and values over time” (Hedberg 1981, p. 6) irrespective of person-nel turnover or changes in leadership. The concept of memory is similar to Hardy’s (1996) previously cited notion of the system dimension of power. The regulatory aspects of organizations, their specialization, their departmental-ization, and rituals and beliefs affect how knowledge that is thought to be relevant becomes integrated in the organizational knowledge base and the mechanisms for retrieving it. In stabilizing the activities of organizations, the regulatory, cultural, and structural properties of organizations also give rise to a certain inertia in learning processes. The stabilization of learning processes that results from organizational memory implies that the knowledge gained through boundary-spanning activities becomes relevant to an organization if that know-ledge is compatible with the dominant beliefs and practices and if it fits with the organization’s structures.

In their routine-based conception of learning, Levitt and March (1988) emphasize the salience of organizational memory for learning processes. The memory of organizations is sustained through the socialization of organizational members into the structure of routines and through control. Because much of what organizations do is accomplished primarily through rules and routines (March & Olsen 1989, March & Simon 1958, Thompson 1967), organizations are seen as learning by encoding inferences from history into routines that guide behavior. “Unless the implications of experience can be transferred from those who experienced it to those who did not, the lessons of history are likely to be lost” (Levitt & March 1988, p. 526). The generic term “routines” includes the rules, procedures, forms, rituals, and beliefs around which organizations are con-structed and through which they operate. The basic assumptions of the routine-based conception of organizational learning enjoy widespread recognition in the organizational learning discourse. Weick and Ashford (2001), for instance, point out that the acquisition of new knowledge tends to serve established beliefs and practices. Berthoin Antal, Lenhardt et al. (2001) suggest that the existence of a “not-invented-here syndrome” may stifle the communication among the various units in an organization. The grounding of organizational activities in the col-lective, interindividual knowledge base can prevent norms, objectives, and

40

structural arrangements from being questioned and can thereby block “double-loop learning” (Argyris & Schön 1978, p. 22).

There are structural difficulties in a routine-based conception of learning. The past is not always a perfect predictor of the future, and the appropriateness of routines is not fixed for eternity (Levitt & March 1988, p. 531). To stress the stabilizing effect that organizational memory has on learning from boundary-spanning activities is to imply that organizations are locked into “single-loop learning,” a process that leads to the refinement of existing practices and beliefs (Argyris & Schön 1978, p. 18; Nonaka & Takeuchi 1995). But the stabilizing effect that organizational memory has on learning contradicts the destabilizing effect that boundary-spanning activities have on organizational life. The inferen-ces that organizational members draw from their communicational interactions with external stakeholders entail “the risk of destabilizing the activity system” (Tacke 1997, p. 16). Jevnaker (2003), for example, points out that boundary-spanning activities give organizations access to knowledge that tends to be uncertain, equivocal, and plagued by partial approaches. Knowledge that is tap-ped from external sources may be inconsistent with conceptions of appropriate-ness in organizations (Levinthal & March 1993) and incompatible with their core beliefs (Wiesenthal 1995, p. 145).

It is therefore worth asking how learning can take place given the variety of new, but not necessarily compatible, ideas, perspectives, expectations, and forms of expertise that organizations come across during their boundary-spanning activities. Wiesenthal (2000) contends that theorists of organizational learning may benefit from institutionalist approaches by conceptualizing the constraints that affect an organization’s search, experimentation, and discovery of new behaviors. The institutionalist perspective on learning from boundary-spanning activities holds that societal norms, conventions, and values are brought into the internal context of organizations in the course of these efforts (Granovetter 1985, Scott 1992). Organizations incorporate societally legiti-mated, rationalized elements in their external relationships to maximize their legitimacy and increase their resources and survival capabilities (Meyer & Rowan 1977). But the communicational interactions at the external boundaries of organizations differ in the unquestioned acceptance and legalistic rules and routines that drive these interactions. Organizations with a set of functionally differentiated, highly discrete boundary-spanning activities may experience greater challenges to the stability of their activities than do organizations that pursue more homogeneous and restricted boundary-spanning activities. How-ever, the impact that externally fixed institutions have on learning from boundary-spanning activities is relative. They constrain and enable but do not determine boundary-spanning activities, for human agency creates, reproduces, and changes rules, norms, and values in the daily activities of organizations (see Geppert 2000, Giddens 1984, Risse 2000).

41

Emerging Theoretical Model and

Implications for Research

This section develops the thesis that processes of organizational learning need to be related to the social context that sustains them if one is to understand the emergence of distinctive qualities of learning in the course of the efforts that organizations make to align themselves with their environments (Child 1997, Easterby-Smith et al. 2000, Geppert 2000, Gherardi & Nicolini 2001, Gheradi et al. 1998). It is argued that the focus on boundary-spanning activities is instruct-ive for the contextualization of learning processes because it points out ways of learning from social interaction within specific sociocultural and material set-tings at the interface between organizations and their environments and links these interactions with intraorganizational dynamics. Huber’s (1990) stage model of learning, which distinguishes between the acquisition, distribution, and interpretation of knowledge and its storage in organizational memory, serves as the basis for integrating the boundary-spanning notion with theories of organi-zational learning. The resulting theoretical model emerges from discussion of the impact that boundary-spanning activities have on each stage of learning. The main aim of this discussion is to arrive at an understanding of the conditions and circumstances that influence the refinement of existing practices and the creation of new ones as a result of communicational interactions that organizations have with external stakeholders. The contextual dimension of learning directs attent-ion to the importance of perceptual filters, regulatory aspects, organizational politics, and the stability of an organization’s activity system for learning from boundary-spanning activities. Figure 2.2 outlines the contextualized model of organizational learning and summarizes its chief arguments.

The key issues of the analytical framework are translated into the follow-ing research questions to provide a starting point for examining the impact of boundary-spanning activities on processes of organizational learning:

1. Is there evidence of a need to learn from external stakeholders?

2. How does information from external stakeholders become relevant know-ledge to the organization? How is it channeled into the organization? To what extent is the acquisition of knowledge affected by power relations in the organization?

3. How is the acquired knowledge shared by the various groups of the organization? How does it feed into internal decision-making processes? How are these processes regulated?

4. How do people make sense of the acquired knowledge? To what extent is the initial definition of meaning subject to sense-making processes? How are these processes shaped by organizational politics?

42

5. How is the acquired knowledge related to the established beliefs and practices of the organization? What are the stabilizing forces that affect the storage of the acquired knowledge?

Figure 2.2 A contextualized model of organizational learning.

Boundary-

spanning Activities

Readiness to

Learn

Knowledge

Distribution

Knowledge

Acquisition

Knowledge

Interpretation

An organization is ready to learn from boundary-spanning activities if there is a perceived need to draw on the interactions with external stakeholders as sources of knowledge about the environment.

(a) The social constitution of boundary-spanning activities determines how external knowledge is filtered and channeled into an organization.

(b) Acquiring knowledge from boundary-spanning activities may influence the power relations of an organization.

(a) The regulation of the linkages between boundary-spanning activities and intra-organizational processes determines how and how much the different subgroups of an organization gain access to the acquired knowledge.

(b) Agency in the use of the principles that govern the knowledge distribution implies that specific interests may be pursued through knowledge-sharing.

(a) The interpretation of externally acquired knowledge across an organization’s subgroups with its multiple realities and variety of practices is essential for learning.

(b) Organizational politics assumes different forms and variously affects the sense-making of the acquired knowledge in different ways.

(a) The impact that organizational memory has on learning from boundary-spanning activities implies preference for knowledge compatible with the dominant beliefs and practices in the organization.

(b) Learning from boundary-spanning activities challenges the stabilizing impact of organizational memory because such learning is associated with knowledge gains that tend to be equivocal and incompatible with the dominant beliefs and practices in the organization.

Organizational

Memory

43

3. Research Design, Context, and Methodology

Exploration of the impact that boundary-spanning activities have on processes of learning in supranational organizations starts with two tasks. The first is to apply a theoretical concept (boundary-spanning activities) to the empirical reality in supranational organizations. The second task is to relate that concept to pro-cesses (organizational learning) that are invisible without the use of conceptual tools (Kissling-Näf & Knoepfel 1998; Kissling-Näf, Marek & Gentile 1994). A purely deductive approach to the theoretical question behind the research entails the risk of producing empirical evidence of a phenomenon that remains abstract to the practitioners in the field, including policy-makers and public adminis-trators in the EU. Relying on an inductive method instead makes it difficult to arrive at generalizations that help refine existing theory in the organizational learning discourse. Looking at learning processes through the abstract lens of boundary-spanning activities therefore calls for a research strategy that allows for both the interplay of conceptual ideas and the generation of insights about the dynamics in the European Commission. The major requirement of the research design is to advance understanding of the role that boundary-spanning activities play in learning processes and to deepen insights about the organizational dynamics of Europe’s public administration from an actor-centered perspective.

Research Design

Ragin’s (1994) model of social research provides a way to integrate deduction and induction into the research on the impact that boundary-spanning activities have on processes of organizational learning. He argues that most social research includes elements of both and that the interplay between ideas and evidence “culminates in theoretically based descriptions of social life and evidence-based elaborations of social theory” (p. 47). Figure 3.1 illustrates that analytic concepts derived from social theory help make sense of evidence as images are synthesized from the data.

44

Figure 3.1 A simple model of social research. Adapted from Constructing Social Research

(p. 57), by C. Ragin, 1994, Thousand Oaks: Pine Forge Press.

Ragin (1994) explains that analytic concepts call social phenomena into being by defining them and provide conceptual tools to differentiate them. The use of concepts in social research is associated with the act of naming phe-nomena to fix attention on them and to ask questions about them. An image forms when bits of evidence are linked in meaningful ways to bring coherence to the data. This process involves the selection of facts considered germane to the research focus and relates back to the ideas and frames that initially moti-vated the collection of data. As Ragin points out, “it is difficult to form an image from evidence without first using some sort of initial analytic frame to highlight or define relevant evidence” (p. 72). Because concepts and images do not necessarily fit, a dialogue between these two ends of the research process is necessary. According to Ragin, the dialogue between ideas and evidence drives the analysis of social phenomena. It means breaking the phenomenon into its constituent parts and looking at them in relation to the whole they form.

The present dissertation is based on a qualitative case-study approach to exploring the notion of boundary-spanning activities and its relation to processes of organizational learning. This mode of investigation affords sufficient insight into the complex interrelationships between the factors identified in the theo-retical framework (Child & Smith 1987). A number of empirical studies on organizational learning endorse the qualitative case-study approach by main-

Ideas/Social Theory

Evidence/ Data

Analaytical Concepts

Images

Representation of Social Life

Mostly

Deductive

Mostly

Inductive

Retroduction

45

taining that theoretical grounding is a key requirement for conducting case studies about learning in organizations (Krebsbach-Gnath 1996, p. 91; see Geppert 2000, Kerlen 2003, Scherf-Braune 2000). The design of the case studies used in the present work draws on basic ideas in Ragin’s model of social research, with the boundary-spanning notion serving as the analytic concept within which to portray the interpersonal relations as a trigger of learning at the external boundaries of organizations. This chapter presents two case studies providing evidence that learning processes differ depending on the social con-text in which they occur. To distinguish between the two cases, this study refers to the different functions of boundary-spanning activities: information-proces-sing and external representation. The emergence of organizational learning needs; the acquisition, distribution, and interpretation of knowledge; and organi-zational memory are the conceptual tools of the process model of organizational learning. They structure the analysis and presentation of the empirical evidence, aid in the outlining of the learning processes, and pave the way for a theory-guided discussion of the evidence of boundary-spanning activities.

Research Context

An inquiry into the impact of boundary-spanning activities on learning processes in supranational organizations requires the selection of a research context that reflects the basic ideas of the theoretical framework. This study focuses on the management and administration of the Research and Technology Development (RTD) program that promotes a “User-friendly Information Society” in the 5th framework program (1998–2002) of the European Community. Putting the management of the Information Society Technologies (IST) program at the center directs attention to the Directorate General for the Information Society (DG INFSO, hereafter also referred to as “the administration”). It serves as an example of a supranational organization for three reasons. First, it has a dynamic environment, which pressures it to engage in learning. Second, the nature of that environment requires this focal organization to possess competencies and resources that enable it to cope with changes in its environment. Third, DG INFSO exhibits interpersonal relations classifiable as boundary-spanning activities at its external boundaries.

The IST Program

Over of the last 20 years, the European Commission has learned how to accumu-late knowledge from external sources, to weave the knots between the powerful actors in the field, and to act as a think tank providing new solutions to the old

46

problem of competitiveness (see Abels 2000, Roobeek 1990, Schaper-Rinkel 1999). It has tried to respond to the changes in the European industries related to information and communication (IC) technologies, where computing services became one of the few areas of growth during the phase of recession, market saturation, and oligopolistic competition that plagued the hardware manufac-turers in the 1990s. The emphasis in European technology policy shifted from supporting the producers of high-tech products and services to supporting their users, that is, from the production to the diffusion and application of those goods and services. Moreover, a holistic view of technological innovation formed, promoting recognition of the importance of contextual factors such as skills, training, risk capital, and technology transfer (Peterson & Sharp 1998, p. 127; Peterson & Bomberg 1999, p. 219). The shift in emphasis coincided with the involvement of a wider range of program participants. Noting the waning inter-ests of the major information technology (IT) companies in European policy initiatives, the Commission became more responsive to the needs of small and medium-sized enterprises and began to reach out to the users of modern IC technology devices (Cram 1994, p. 207). The reconceived and broadened rationale for European technology policy did not change the fact that its design remained rooted mainly in economic thinking (Selwyn 2001).

Receiving €3.6 billion in funding, the IST program (see Table 3.1) is financially the largest of the four specific research programs within the Euro-pean Community’s 5th framework program,2 whose budget totals €14.96 billion. Unlike those three counterparts, which run under the Directorate General for Research (DG XV), the IST program is managed by DG INFSO. Building on the convergence of technologies, the IST program integrates the different foci of its predecessor programs in the 4th framework program, which are related to IC technologies. These forerunners are the European Information Technologies Program (ESPRIT), with its focus on information technologies; the Advanced Communications Technology and Services Program (ACTS), which centered on the development of communication technologies; and the Telematic Applica-tions Program (TAP). The acknowledgement of technological convergence drives the integration of the different programs. The mobile phone, for instance, demonstrates the convergence of single technologies into network technologies. With the possibility of sending and receiving e-mails, surfing the Internet, and downloading music and videos, categorization becomes a tricky issue. Is this mobile device a phone, PC, TV, video recorder, or play station? In the end, it may probably be seen as an example of a new generation of technical artifacts in

2 The other three programs of the 5th framework program are (a) Improving the quality of

life and management of living resources, (b) Promoting competitive and sustainable growth, and (c) Securing the energy supply, protecting the environment, and ensuring sustainable development. Additionally, the 5th framework program comprises three hori-zontal actions that span the research programs.

47

the evolution of the Information Society. The proclaimed convergence of technologies is accompanied by a blurring of the boundaries between what used to be historically distinct sectors in industry. Consumer electronics seizes upon the opportunities that arise from the integration of information-processing and electronic information transmission. User-friendliness is a recurrent theme in the outline of the IST program. It is translated into a number of overarching key requirements such as the utility, dependability, interoperability, and affordability as crucial elements of the Information Society.

Table 3.1 The Information Society Technologies (IST) Program: key actions, horizontal lines,

and budget breakdown

Type of activity Amount (millions

of Euros)

a. Key actions

(i) Systems and Services for the citizen (ii) New methods of work and electronic commerce (iii) Multimedia content and tools (iv) Essential technologies and infrastructure

b. Horizontal actions

(i) Research and technological activities of a generic nature: Future and emerging technologies

(ii) Support for research infrastructure: Research networking

646 (18%) 547 (15%) 564 (16%) 1,363 (38%)

319 (9%)

161 (4%)

Total 3,600a

a Of the total budget, 10 percent is reserved for “cross-program actions”; 2 percent, for “integrated application platforms.” The fact that neither spending title appeared in the table of the IST program decision implied that the decisions on the budget sources for cross-program actions and integrated platforms had to be taken once the program commenced.

Adapted from the IST Program Decision of the Council of Ministers, Official Journal of the European Commu-

nities, 12 March 1999 (L 64/20) – 1999/168/EC.

The European RTD policy is one of the few areas where a substantial policy role of the Community is relatively uncontroversial among the Member States (Peterson & Sharp 1998, p. 126). Despite the Euroscepticism during the 1980s, the budget for the framework program more than tripled in the eight years following its first launch, growing from €848 million in 1984 to €2.842 billion in 1992 (Peterson & Bomberg 1999, p. 208). By the time the IST program was launched at the end of the 1990s, European research policy was one of the most developed of all EU policies. This spillover to the supranational level may to some extent be explained by the way the Commission uses the

48

knowledge acquired from its advisory committees and other expert panels (see Elizalde 1992, p. 324; March & Olsen 1989, p. 30; Pierson 1996, p. 137). It recognized early on that building and nurturing relationships with represent-tatives from industry and academia strengthens its own voice when it seeks approval for its initiatives from the Member States (Peterson & Sharp 1998, p. 177). Moreover, the incorporation of external expertise helped the Commis-sion cope with the widening information gap that opened between private and public actors because of the technological complexity and rapid change of IC technologies (Knill 2001, p. 237).

The Management of the IST Program

DG INFSO was formed as a new administrative authority in the late 1990s to manage the IST program in the 5th framework program. DG INFSO’s declared mission was, and still is, to formulate and implement coherent and integrative policies for the broadly defined information society.3 The birth of the adminis-tration went hand in hand with a distinguishing feature of the IST program: the systematic merger of ESPRIT, ACTS, and TAP, which in the 4th framework program had run more or less independently of each another. (The latter two had been managed by the former Directorate General for Telecommunications and Innovation [DG XIII]; ESPRIT, by the Directorate General for Industry [DG VI].)

The organization of DG INFSO reflects the substance of the IST program (see Figure 3.2). Five of DG INFSO’s six directorates are assigned to the program’s key and horizontal actions (“Systems and Services for the Citizen”; “New Working Methods and Electronic Commerce”; “Content, Multimedia Tools, and Markets”; “Essential Information Society Technologies and Infra-structures”; and “Integration and Implementation—Networks and Future Technologies”). In addition to the five research directorates (B–F), DG INFSO has Directorate A as its regulatory pillar. One risk of DG INFSO’s divisional structure is that it can stifle interdepartmental coordination and communication (see, for instance, Mintzberg 1983, pp. 215–252). It implies that these six direc-torates and the units managing the key actions and action lines within the IST program tend to operate as quasi-autonomous entities having little need for mutual coordination. Each has its own budget, personnel, and power to make the decisions on the program’s operations.

3 See the mission statement of DGoINFSO at [http://europa.eu.int/comm./dgs/information_

society/mission_en.htm] (accessed 26 April 2001)

49

Information Society Directorate General

Deputy Director-General Director-General Deputy Director-General

Assistant to the Deputy

Director-General Unit 1 – Analysis,

Policy-planning

and Coordination

Unit 2 –

Interinstitutional

Relations

Assistant to the Deputy

Director-General

Directorate A

Policy and Regulatory Framework

Directorate B

Systems and Services for the Citizen

Directorate C

New Methods of Work and Electronic

Commerce

Directorate D

Multimedia Content and Tools

Directorate E

Essential Technologies and Infrastructure

Directorate F

Horizontal Actions

Unit A 1 – Regulatory framework

Unit A 2 – Implementation of framework

Unit A 3 – Mobile

communications

Unit A 4 – Internet-related

services

Unit A 5 – International

aspects

Unit A 6 – Regional and

societal aspects

Unit B 1 –Applications

relating to health

Unit B 2 – Appli-cations relating to the disabled

Unit B 3 – Appli-cations relating

to administration

Unit B 4 – Appli-cations relating to environment

Unit B 5 – Appli-cations relating

to transport

Unit C 1 – New methods of

work

Unit C 2 – Appli-cations relating

to business

Unit C 3 –Electronic commerce

Unit C 4 – Information

security

Unit D 1 –Electronic publishing

Unit D 2 –Cultural heritage

applications

Unit D 3 –Educa-tion and training

applications

Unit D 4 –Linguistic

applications

Unit D 5 – Management of

information

Unit E 3 –Simulation

technologies

Unit E 4 – Mobile

communications

Unit E 5 – Peripherals and microsystems

Unit E 6 –Microelectronics

Unit E 2 –Technologies for

software

Unit E 1 –Technologies for info-processing

Unit F 2 – Research

networking

Unit F 3 – Tele-communications

networks

Unit F 4 – Inter-national aspects of the program

Unit F 5 – Impro-ving the use of human capital

Unit F 6 – Work program, cross-program themes

Unit F 7 – Operational

program aspects

Figure 3.2 Organization chart of the Directorate-General for Information

Society (Adapted from [www.europa.eu.int/comm/dgs/information_society/

index_en.htm]).

Unit F 1 – Future

technologies

50

The implementation of the IST program is a cyclical process consisting of a number of chronological steps to be carried out by the officials in the five research directorates of DG INFSO (see Figure 3.3). The management process starts with the annual definition of the IST work programs in which the Directo-rate General identifies the priorities of the coming year, refines and updates the topics to be addressed, and specifies a road map for implementing the program. The work program also schedules the calls for proposals and determines the budgetary allocations for each call. A host of external and internal consultations facilitate the definition of the work program. It is particularly during this step of the management process that DG INFSO mediates between large and small companies; constituencies from information technologies, telecommunications, software and content producers; and, of course, the EU Member States. This study on the boundary-spanning activities of DG INFSO was conducted before the EU’s recent enlargement in May 2004. In the process of implementing the IST program, DG INFSO consulted a total of 15 Member States and 15 associated states.

Figure 3.3 Management process Information Society Technologies (IST) Program.

Adapted from Five-Year Assessment Report Relates to the Specific Program: User-Friendly Information Society. Covering the Period 1995-1999, by A. Pompidou, J. Affenzeller, P. Boekholt, R. Cochran, F. Garas, G. Hains, & K. J. Linden 2000, p. 4. See [ftp://ftp.cordis.europa.eu/pub/fp5/docs/fp5_panels_final_report _ist_2000.pdf], pp. 1–42 (downloaded 20 March 2001).

IST Program

Call for Proposals

Evaluation of Proposals

Selection of Proposals

Contract Negotiation

Definition Work Program

Project Monitoring and

Management

Evaluation of Results

Renewal Work Program

51

Once the work program is defined, DG INFSO invites the program constituency to submit research proposals pertinent to a given call. To select the incoming proposals, the administration organizes panels of external experts to evaluate the proposals on their technical merits, weighing a number of general criteria such as scientific excellence, European dimension, and exploitation potential. An implementation plan based on the evaluation reports of the exter-nal experts and on strategic considerations within DG INFSO is drawn up. It ranks the proposals that are eligible for funding. This list is usually cut off at the point where the allocated budget is exhausted. In total, eight calls for proposals have been published during the 5th framework program, with more than 2,150 projects being selected for funding after the evaluation of the calls. The bulk of the program budget, which covered calls 1 through 7, was spent in the first three years (1999–2001).

The contract negotiations with the project participants follow a single contract model that was defined in the Council decision on the 5th framework program. The model is applied irrespective of the funding volume. Because the contracting method is based on cost reimbursement, a major issue in the nego-tiations with the project participants from companies or research organizations is the procedure for the cross-checking of individual spending by DG INFSO. Before contracts can be given out, DG INFSO must consult the Legal Service of the Commission and solicit approval from the College of the Commissioners. As soon as the 20 Commissioners formally agree, the stage is set for the actual undertakings that the selected proposers have in mind.

The chosen projects are monitored in the units of research directorates B through F that relate to the topic of the envisaged research activities. A project officer manages on average anywhere from 10 to 20 projects at a time, remain-ing in contact with about 100 different project partners from academia and companies. For lack of a formal procedure for assessing the progress that projects are making, the different units of DG INFSO have latitude to establish their own practices for monitoring them. The internal evaluation, which may differ from unit to unit and from key action to key action, is accompanied by a program-wide evaluation. The administration involves external experts to engage in the “Integrated Project Portfolio Analysis” (IPPA). The actors involved in the management of the IST program receive from IPPA a program-level overview of the response to the calls and an examination of the portfolio characteristics from a technical, technological, and market perspective. The emergent trends and developments in the project portfolios are analyzed against the comparatively long-term perspective of IST and the priorities of the successive work programs. In total, five IPPA studies are conducted within the 5th framework program.

52

Focal Boundary-spanning Activities

External stakeholders make up two bodies with pronounced roles in the management of the IST program. One is the Information Society Technologies Advisory Group (ISTAG), which consists of experts from industry and acade-mia. ISTAG has the formal role of counseling DG INFSO on the content and direction of the IST program. The administration draws on ISTAG’s advice in the process of deciding on the prioritization of research activities within the IST program and on the spending of the allocated budget. The second body with a prominent managerial role in the IST program is the Information Society Technologies Management Committee (ISTC). It is a Comitology committee that monitors the implementation of the IST program by overseeing the imple-mentation powers conferred upon DG INFSO by the EU Member States. The members of ISTC are delegated by the governments of the Member States, the accession countries, and the associated states to represent their interests. DG INFSO has no influence on the staffing of the committee. ISTC must approve the work program before it can be implemented and has a say in the funding decisions.

The involvement of both bodies in the management of the IST program illustrates two different sets of communicational interactions at the external boundaries of DG INFSO: one with ISTAG as an opportunity to process infor-mation from external sources, the other with ISTC as a vehicle for external representation. The present study focuses analytically on both sets of boundary-spanning activities in order to explore their impact on processes of organiza-tional learning in DG INFSO. The communicational interactions with ISTAG and ISTC are not the administration’s only paths to industry, academia, and Member States. DG INFSO engages in many kinds of boundary work to align itself with its environment. The program’s constituency, potential beneficiaries, and stakeholders in the political environment are consulted formally and informally at various levels so as to ensure that the program meets their expec-tations and to build understanding of the great variety of interests, requirements, and ideas in the field of IST. This practice applies in particular when it comes to defining the annual work programs and, to some extent, to making decisions on the funding of research activities within the IST program.

Judging from the kind of external experts frequently included in the management of the program, however, there appears to be a preference for bringing technical expertise into DG INFSO. This favoritism implies that DG INFSO officials must possess a certain level of expertise themselves in order to benefit from the consultations with the various external specialists. The officials need to understand the technologies and their main drivers and important actors and need to have evaluative criteria at hand to assess the trends and developments bearing on its program management tasks. The large share of

53

category A personnel in DG INFSO’s human resources conveys this apparent feature of the administration (see Table 3.2).

Table 3.2 Human resources of DG INFSOa

Personnel categoryb

Scientific tasks Administrative tasks Total

A 247 38 285

B 3 79 82

C 0 173 173

Total 250 290 540

a Directorate General for Information Society. b Persons in category A are scientifically trained and perform administrative and advisory duties. Category B requires no university degree and accounts for a much smaller share of DG INFSO executives than category A does. Category C consists of clerks performing secretarial tasks.

Adapted from “Effectifs R&DT—Total DG XIII [Directorate General for Telecommunications and Innovation] (hors ex-ESPRIT encore affecté à la DG III [Directorate General for Industry] et hors MONITOR) Situation au 1 January 99,” 12 April 1999 (SJ/vc/990407-1.xls).

Methodology

Empirical research on organizational learning is rare. The ideas about the initial question posed by Argyris and Schön (1978)—“What is an organization that it may learn?”—tend to be abstract, and best practices for the operationalization of organizational learning into indicators are difficult to find. The primary objective of the present study is therefore to refine and advance existing theory in the organizational learning discourse by drawing on empirical research into the impact of boundary-spanning activities on learning processes in supra-national organizations. A systematic account of the research context’s factors that affect organizational learning requires the theoretical concepts to be transformed into an analytical framework that may be altered to fit the findings synthesized from interpretation of the empirical data.

Data Collection

There were essentially two sources of information that were used to become familiar with the contexts of learning in DG INFSO and with the factors that

54

affect learning triggered by the administration’s boundary-spanning activities with ISTAG and ISTC:

1. Qualitative interviews

2. IST-related documents of the EU, including legal acts and program publications such as IST work programs, calls for proposals, program analyses, and ISTAG reports.

The analysis of the IST-related documents provided background informa-tion for conducting the interviews. Familiarity with the legal basis of the IST program and developments in this policy field was necessary to understand the contextual dimensions of learning in DG INFSO. Interviews required some knowledge of the policy-related issues that concern the IST program. Most of these documents were downloaded from the Internet. A total of 43 qualitative interviews were conducted, almost one third of which were with members of ISTAG and ISTC. Roughly two thirds of the interviews were conducted with officials of DG INFSO (see Table 3.3). The major purpose of the data collection was to elaborate the analytical framework. Interviews were therefore conducted and IST-related interviews were analyzed until the point of saturation was reached, that is, until the collected evidence of boundary-spanning activities and learning processes appeared redundant with previously collected evidence. In this method of research, only a subset of the potential evidence is used, so the sample is not representative of the larger set of populations.

Table 3.3 Distribution of interviews

Affiliation of the

interviewees

Number of interviews Number of interviewees

DG INFSOa (N = 540) 30 25

ISTAGb (N = 26) 6 6

ISTCc (N = 85) 7 7

a Directorate General for Information Society. b Information Society Technologies Advisory Group. c Information Society Technologies Management Committee.

The selection of the interview partners from DG INFSO followed a top-down logic. The directors of research directorates B through F and heads of unit—all listed in the organization chart published on the DG INFSO website—were approached for an interview by e-mail. In almost all cases it was possible to arrange for an interview within the envisaged time period. As familiarity with DG INFSO grew, it became possible to interview officials from lower manage-

55

ment levels as well. The interview partners from ISTAG, all of whom counted as active participants in the meetings of the group, were suggested by DG INFSO personnel. Likewise, an important criterion for the selection of indi-vidual Member State delegates as interview partners from ISTC was their active participation in that committee’s meetings. A member of the committee who had been approached in a preparatory stage of the research suggested a number of potential interlocutors from ISTC, who all agreed to be interviewed. All interviews but one were taped and later transcribed.

The interviews were based on interview guides and usually lasted from one to two hours. These guides structured the interviews by keying in on the topics and issues of relevance to the research interest while leaving enough flexibility to adjust to the specific circumstances of the interview situation and the personalities of the interviewees. The conceptualization of the interview guides stemmed from the general idea that processes of learning are embedded in actual practices of the focal organization and that knowledge in organizations is relational, context-specific, and anchored in the intersubjectively shared beliefs and commitments of the organizational members (Gherardi et al. 1998). The interviews covered the three main topics of theoretical interest for the study:

1. Social constitution of boundary-spanning activities

• Organization of linkages with internal dynamics (structural inte-gration, regulation, perceptions, orientations)

• Communication and conflict resolution

• Provision of knowledge

2. Factors and conditions in DG INFSO that affect organizational learning

• Perceptual filters and construction of relevance

• Communication and arrival at common opinion (interest accom-modation, conflict resolution, reliance and use of knowledge from external sources)

3. Contexts for learning from boundary-spanning activities with ISTAG and ISTC

• Impact of ISTAG and ISTC input on daily activities

• Strategic planning for IST program

• Operative implementation of the IST program

• Integration of strategic and operational decision-making

56

Initially, two data collection phases were planned in order to capture the dynamics of organizational learning and to create opportunities for an empirical examination of questions that arise at later stages of the research process. In early stages of the research process, however, it became clear that two phases of data collection were not sufficient to achieve these ends. There was evidence that DG INFSO drew on its boundary-spanning activities to scan its environ-ment, but deepened insights about its internal dynamics were necessary to understand the implications that these activities had for strategic and operational decision-making. The two data collection phases therefore gave rise to four pha-ses, with three of them entailing the interviews with officials of DG INFSO, and one focusing on the members of ISTAG and ISTC. The design of the interview guides was modified with each phase of data collection as the conceptualization of the interview guides gradually shifted from general and abstract to specific and concrete (Table 3.4). At first, they were primarily theory-driven and almost void of any practical notion of boundary-spanning activities and their impact on organizational learning. The analysis of the input from the early interview rounds was indispensable for addressing the relevant issues in the distinctive contexts for learning from the boundary-spanning activities with ISTAG and ISTC. The evidence that the first two rounds of interviews produced about the contextual dimensions of organizational learning facilitated their detailed exploration in the third and fourth rounds.

Table 3.4 Shifting emphasis in the data collection

Interviews Social constitution

of boundary-

spanning activities

Factors and conditions

in DG INFSO that affect

organizational learning

Contexts

of learning

September 2000

DG INFSOa (n = 5)

X X

May 2001

DG INFSO (n = 15)

X X

June–July 2002

ISTAGb & ISTC

c (n = 13)

X X

January 2003

DG INFSO (n = 10)

X X

a Directorate General for Information Society. b Information Society Technologies Advisory Group. c Information Society Technologies Management Committee.

57

To deal with the shifting emphasis in data collection, different groups of interviewees examined the three main topics of theoretical interest to the study. The 20 DG INFSO interviewees who responded to items about the focal boundary-spanning activities consisted of directors, heads of unit, repre-sentatives of specialized units that served as secretariats to ISTAG and ISTC, and staff of directors with specialized tasks pertaining to the relations with ISTC. Their perspectives on the boundary-spanning activities were comple-mented by those of ISTAG and ISTC members who were interviewed. Detailed understanding of the factors and conditions of learning in DG INFSO was gained through interviews of personnel at different levels (including heads of unit and project officers) and across the different directorates within the Directo-rate General. Lastly, evidence of the contexts for learning from boundary-spanning activities was collected mainly through interviews with project officers and complemented by perceptions that the interviewed members of ISTAG and ISTC had of the extent to which their activities influence strategic and opera-tional decision-making in DG INFSO.

Analysis of the Data

The interpretation of the qualitative data from the interviews with DG INFSO officials and members of ISTAG and ISTC was based on the “interpretative strategy of analysis” by Meuser and Nagel (1991, p. 452). The objective of this method is to arrive at an understanding of the themes that are representative of and commonly shared by the interviewees. Meuser and Nagel’s method places an emphasis on deciphering structures of relevance, constructions of reality, interpretations, and frames of reference and is therefore a compass for the analysis of sense-making processes in the focal organization. Systematic generalization of the empirical findings is possible through the strategy’s two-step approach:

I. Analysis of individual interviews

• Transcription

• Paraphrase of the interview text to identify themes

• Labeling of the identified themes

II. Comparison of the main interview findings

• Thematic comparison of interviews that follows the logic and terminology of the interviews rather than the conceptual logic of the analytical frame

58

• Use of conceptual framing to bring different aspects of a phenomenon together, which implies abstraction from the logic and terminology of the interviews

• Interpretation of the empirical findings on the basis of the analytical frame, which implies identification of typical patterns by interlinking the findings

The interpretive strategy of analysis fueled the ongoing development of the analytical framework. Development of an analytical framework that contri-buted to the understanding of the complex social processes inherent in learning triggered by boundary-spanning activities was a major issue during the research process. To prevent rash simplifications, it was necessary to integrate deductive and inductive elements into the research process, which was therefore typified by iterative movements and trial-and-error loops. They progressively refined the images that arose from the interpretation of the empirical data and better-specified the analytical framework. Alternation between the collection and inter-pretation of the data helped to relate the idea of boundary-spanning activities to the practices and beliefs in DG INFSO. It thus identified findings about this phenomenon that made sense for this study about learning processes in supranational organizations. The existence of four data collection phases, each based on individual interview guides that were refined after the first number of approximately three interviews in each phase, reflects the openness and flexibility of the research process.

Two reports were written to prepare the concluding analysis of the empirical research. They drew inferences from the interpretation of the inter-views and the IST-related documents and explored the explanatory power of the conceptual notions developed in the theoretical framework. The presentation of the empirical evidence reflected the shifting emphasis of the data collection phases from abstract to specific. Whereas the first report centered on the sub-stantiation of analytical categories, the second resembled a policy analysis of the IST program. The present comprehensive analysis of the empirical evidence occupies the middle range of the deduction–induction spectrum.

59

4. Learning from Boundary-spanning Activities

with ISTAG

The Role of ISTAG in the Management

of the IST Program

DG INFSO faced tremendous challenges when it began its activities in 1998 to launch the IST program to anticipate the convergence of IC technologies. To cope with the increase in the program’s complexity and with the pace of devel-opment in IST, the newly formed administration had to develop its capacity for strategic action. The management of the program as a “rolling work program” illustrates the daunting task that confronted DG INFSO. The annual definition of the work program enabled DG INFSO to spell out and update the substance of it in response to perceived trends and developments in the field. DG INFSO installed ISTAG, the group of experts from industry and academia, to gain sup-port for the management of a complex and dynamic program. To explain how DG INFSO draws on the knowledge and support of this advisory group, the following examination of boundary-spanning activities between these two parties is written from a viewpoint of organizational learning.

As an external advisory group to DG INFSO, ISTAG is designed to pro-vide the Commission with knowledge deemed relevant for carrying out research policies. The legal basis for ISTAG’s involvement in the management of the IST program is the Commission Decision of 22 October 1988, which permits groups of experts to be set up to advise the Commission on the content and orientation of the key actions in the policy field of RTD. Representatives of the relevant communities are consulted as well in order to strengthen the strategic position of European research policies. ISTAG’s official mandate is to advise on proposals for guidelines spelling out the annually updated work programs, on views about the timetable of calls for proposals, on criteria for evaluating project proposals, and on verifiable objectives for achieving the aims of the IST program’s key actions. It “makes comments in many directions: on the participation, the end results, the technical content of the new calls, the new concept for the future, new instruments. And, of course, it is with a very strong content on technical aspects” (AG4, p. 11).

To assume the role of an advisory group to DG INFSO, ISTAG has consisted of 26 experts from industry and academia since it was established in 2000. About two thirds of the experts are recruited from the corporate environ-ment, with the remaining third coming from academic or other publicly funded

60

scientific organizations. The group’s members originate from what was then 15 Member States of the EU, the accession countries, and a number of asso-ciated states such as Norway or Israel, which have research contracts with the EU. The Commission officially appoints the members of ISTAG on the basis of their personal merits that distinguish them as experts for the IST program. DG INFSO, however, prepares a list of experts to be included in ISTAG. The actual choice of the administration is finalized by the College of Commissioners.

To gather a group of experts in the advisory group, DG INFSO draws on its experience with the field and may consult civil servants in the ministries of the Member States. The interviews with officials of DG INFSO reveal a pre-ference for people with a technical background in IC technologies and with an ability to focus on strategic issues, such as “what kind of technologies seem to be the emerging ones and which should be taken seriously also in a research program of the Commission” (Com01_5, p. 8). The selection of the ISTAG experts depends, furthermore, on their ability to act as respected spokespersons for one or more sectors. The authority of the group’s experts distinguishes them from other external stakeholders whose opinions are acknowledged by DG INFSO:

ISTAG is one voice among all the voices that DG INFSO may listen to. . . . But due to the audience of such an expert group, it may certainly give them a qualified opinion—qualified for them and qualified for all the participants. You know they are not obliged to agree with ISTAG, but they take into account what ISTAG is saying. (AG3, p. 9)

The Rise of Organizational Learning Needs

Having begun the effort to foster the convergence of technologies in the infor-mation society and promote the utility and affordability of the new goods and services to be developed within IST, it became clear to the participants in the IST program that the task was more difficult than expected. The reason was that the complexity of the environment had increased. Whereas the predecessor programs, particularly ESPRIT and ACTS, had focused on distinguishable sectors in the economy, DG INFSO had to operate somewhere between the con-stituencies of IT, telecommunications, software, content producers, and various fields of use such as automotive manufacturing and the entertainment industry. DG INFSO needed a strategy that would make its activities “reflect what has happened in the real world” (Com00_3, p. 3). The need for strategic action by DG INFSO can be related to the market value of the goods and services to be developed within the IST program:

61

62

Our role is not simply to give money to research. It is really to give money in order to support research which will sustain economic growth in the coming years . . . [so as] to be able to create some focus and some concentration in order to have the power to put something on the market. (Com03_8, p. 5)

Building a strategy in a multifaceted field like IST depends on gaining a broad overview of the developments in the underlying IC technologies, including its artifacts, actors, and main drivers in the economy and in policy. One DG INFSO official points to the importance of “continuously keeping the channels wide open and assessing what is going on” (Com01_6, p. 5). Another representative of the administration stresses its receptiveness to developments in the environ-ment as essential for the running of the program: “You are forced, if you want to do this job properly, to follow very closely all technological, policy, and economic changes” (Com01_15, p. 7). DG INFSO draws on its relationships with external stakeholders to assess future trends in the information society:

We must be able to predict a bit more ahead than anyone else. If we are not doing that by adding contacts, by adding [to] the network and getting that working, we are not doing a good job. (Com01_12, p. 4)

The involvement of external stakeholders in the implementation of the IST program reflects the awareness of DG INFSO officials that they benefit from having their own activities questioned. It is “fresh air in order to avoid repeating always the same scheme” (Com03_8, p. 3). However, the willingness to take in ideas sometimes also raises concerns about their implications for the program’s effectiveness. Developments in the information society may be perceived to be

overtaking what is happening within this program, which invalidates in a certain sense what is happening within such a program. Those are disturbing signals, and they are making people wonder again: Is it at all effective what we are doing, the question of why am I here to do this? That is the identity question. (Com01_14, p. 7)

DG INFSO develops a readiness to learn about strategic orientations with which the program can channel the multiple views, concerns, and preferences expres-sed by the external stakeholders who are consulted in the process of drafting the annual work programs in a certain direction. The administration seeks ISTAG’s advice so as to set priorities and offer an element of choice in the implement-tation of the IST program. The advisory group consists of “people who are in a position where they really have a good overview of the direction things are taking” (Com01_6, p. 5). Tapping into the expertise and authority of ISTAG helps the administration establish the need to focus as a work program is drafted. The danger of failing to do so is clear. If the drafting of the work program is only about ensuring “that everybody is happy, you make a work program of

63

500 pages, and every Member State, every research organization has its own research need 10 times the budget or 15 times. So we have to make priorities” (Com00_5, p. 3).

You need to focus because you do not have available all the money in the world to spend . . . And this focusing runs counter to the fact that there are many interests from the various constituencies . . . but [that] only a few aspects are considered to be of strategic importance for Europe. That has been quite well highlighted by the ISTAG. (Com03_5, p. 2)

Yes, we need focus and concentration, and it was clearly a recommendation coming from ISTAG and coming from other organizations. . . . You have to make some choices, and that is what is difficult when you prepare a work program to make these choices. (Com03_8, p. 5)

Acquiring Knowledge from ISTAG

The need to develop strategic orientations for the IST program sets the stage for DG INFSO’s knowledge-seeking behavior toward ISTAG. But the existence of a need to learn is not in itself a guarantee that the administration engages in learning. The ideas and perspectives of the ISTAG experts need to become relevant knowledge if they are to be acknowledged in DG INFSO’s internal con-text. Knowledge acquisition is an active process. This section explores the administration’s efforts to identify those learning areas viewed as relevant and to tailor its learning needs to the provision of knowledge by ISTAG.

Providing a Vision for the IST Program

ISTAG provides DG INFSO with knowledge that fits its need to learn about strategic objectives for the IST program as a whole. The advisory group proposes the vision of ambient intelligence, which describes its assessment of future developments in the information society. As defined by the members of ISTAG, the notion of ambient intelligence implies the convergence of IC technologies and user-friendliness—key issues in the IST program description within the 5th framework program. Moreover, it emphasizes the possibilities of

64

mobile communication, which is a presumed strength of the industrial land-scapes in Europe.4

I think that ambient intelligence is the concept of converging technologies. The concept of bringing together electronics, information technology, and communications in such a way that technology becomes less relevant—or less visible, I should say. I think it becomes [even] more relevant . . . but less visible. In that way it will lead to novel electronic environments without actually being experienced as electronic environments that will be in support of our needs. (AG2, p. 1)

The information is reaching you without having to worry about where you are or which is the interface you are using, [for] everything appears independently of the kind of device or the place where you are. (AG4, p. 3)

The vision of an invisible, yet ubiquitous, technological infrastructure serving the needs of consumers at all times everywhere has been around since the early 1990s. An article by Weiser (1993), a computer scientist at Xerox Laboratories, can be considered an important milestone in the rise of the notion of ambient intelligence because it introduced the term “ubiquitous computing.” Ubiquitous computing is a building block of the notion of ambient intelligence. Weiser defines the term as enhancing “computer use by making many computers available throughout the physical environment, while making them effectively invisible to the user” (p. 75).

Ambient intelligence is a new paradigm. It is a summary of ubiquitous computing. . . and intelligent interfaces, based on the opportunities of mobile communication and mobile computing. (AG6, p. 1)

The problem would then be to create a new kind of relationship between people and computers by constructing new computing artifacts (e.g., the personal digital assistant, or PDA) for use in everyday life. In the years after publication of Weiser’s article, workshops and conferences about the implications of ubiquitous computing were held.5 Digitization spurred the rise of ubiquitous computing. As a result of improvements in chip design and manufacturing, it has become possible to put more and more features and power onto silicon chips.

4 The interviewed officials of DG INFSO referred a lot to the Commission’s role in the

GSM standard-setting in the early 1990s, among others through the ACTS program. The Commission succeeded in bringing the powerful actors in telecommunications together while seizing upon the assumed market opportunities that would flow from the priva-tization of this sector and the internal market (see, e.g., Bender 1999).

5 An example of such a workshop is the Ladenburger Discourse titled "Living in a Smart Environment – Implications of Ubiquitous Computing" in January 2001, see [http://www. inf.ethz.ch/vs/events/LivingSE.html] (accessed 23 April 2001).

65

The miniaturization of chips and a simultaneous rise of performance power have led to ever-smaller mobile phones, intelligent refrigerators, navigation systems, multimedia in cars, and digital TV, to name only a few examples.

DG INFSO personnel perceived the ISTAG vision of ambient intelligence as just that—a vision. It (a) creates a shared overall goal, (b) offers orientation in a complex environment by anticipating future developmental tracks and poten-tial of IC technology, and (c) provides common ground for different sectors and disciplines to work in the same direction. The ISTAG vision of ambient intelligence is widely understood in DG INFSO as a major contribution to the creation of overall goals for the IST program. For instance, the endorsement of the ISTAG vision is

important to set[ting] the scene for research communities that we [DG INFSO] are working with and to mak[ing] them aware of the key focus that the Commission is looking for in the work that it is going to support. That’s part of a very important process—to convey the message to our research constit-uency. (Com03_9, p. 1).

It answers the question of “what the objective of an initiative is” (Com01_1, p. 21) and gives “some more focus to services, wireless services, to network-ing. So therefore you give probably a bit less attention to some more heavy applications in industrial areas” (Com01_2, p. 5). The vision of ambient intel-ligence builds on the developments in previous framework programs but gives them new meaning. Its endorsement denotes continuity while giving a focus to technological innovations that were already underway in European RTD.

We were already doing initiatives like the disappearing computer. We were already doing initiatives on micronetworks. We were already doing initiatives on middleware. There was already work going on body networks and Palms. . . . We were already working on very low power devices. There were already projects running on putting sensing devices into bottles, cups, tables, chairs. So all of that stuff had been proposed and was being funded. But it wasn’t coherent. . . . What that initiative of ISTAG did was to put a label on the package. (Com03_3, p. 5)

The vision of ambient intelligence is seen as “an agreement on the trend” (Com03_6, p. 1). Once agreed, reference to the vision among the various groups in DG INFSO and the diverse communities in industry and academia with their different interests and priorities offers orientation and provides common ground.

The paradigmatic appearance of ambient intelligence and the fact that it triggers discussion across boundaries foster collaboration in IST-related RTD. “It has the possibility to pull together a lot of the existing directions that we are going toward . . . It kind of fulfills the mission that a lot of companies are working on a single technology. It brings them all together” (Com01_6, p. 8).

66

The advisory group introduced the notion of ambient intelligence for the first time in its recommendations for the 2000 work program to suggest guiding principles for the spending of the €3.6 billion budget for IST. Merely one year after its introduction, the ISTAG vision of ambient intelligence became the program vision. The fact that DG INFSO adopted this concept is evidence of ISTAG’s impact. The core sentence of the advisory report was copied into the work program and has become a standard formula. The work-program orienta-tions can be summarized by the following vision statement:

Start creating the ambient intelligence landscape for seamless delivery of services and applications in Europe relying also on test beds and open source software, develop user-friendliness, and develop and converge the networking infrastructure in Europe to world class.6

Tailoring Knowledge Provision to Learning Needs

Taking ISTAG’s official mandate as advisor to the Commission seriously, the group’s members comment on various matters, such as the substance of the calls for proposals and the appropriateness of the funding models in the 6th framework program (2003–2006). But the group’s views on these topics appear to be of limited relevance to DG INFSO, whose knowledge-seeking homes in on ISTAG’s views about the long-term developments in the field of IST. A member of ISTAG observes, for instance, that the group is trying to comment on the discussion of the program’s content and on changes to make when moving from the 5th to the 6th framework program.

We [had] a very good input into the Commission when it came to describing, sketching the vision and its consequences for the program. With . . . similar discussion on the instruments [for funding], it was not unsuccessful, but I would say that there our influence has been less probably . . . Frans de Bruine7 has also [told] us: “Well, be sure that in the ISTAG you focus really on content because that’s why we need you. That’s why we have this set of experts gathered.” (AG2, p. 5)

During the 5th framework program, ISTAG provided DG INFSO with nine reports containing recommendations for the content development of the IST program (see Table 4.1). The starting point for ISTAG’s recommendations is the

6 See the ISTAG report “IST Work Program 2000”, [http://www.cordis.lu/ist/ library.htm]

(accessed 10 March 2000). 7 Frans de Bruine – one of the directors in DG INFSO – followed George Metakides as the

chair of the ISTC meetings in 2000 and participated regularly in the ISTAG meetings.

67

report on the orientations for the 2000 work program. It defined the ISTAG vision of ambient intelligence and discussed implications for the priority-setting in the work program. The majority of ISTAG’s recommendations that follow the initial report of the group explored the thinking behind the vision of ambient intelligence and its ramifications for the content development of the IST program.

Table 4.1 Advisory reports of ISTAG during the 5th

Framework Programa

Year ISTAG reports

1999 Orientations for 2000 work program and beyond

2000 ISTAG recommendations for the 2001 work program

Implementing the vision

2001 Scenarios for ambient intelligence in 2010

Small business in framework program 6

Integrated projects in framework program 6

2002 Strategic orientations and priorities for IST in framework program 6

Trust, dependability, security, and privacy for IST in framework program 6

Software technologies, embedded systems, and distributed systems

a From the website for ISTAG (Information Society Technologies Advisory Group). See [http://www.cordis.lu/ ist/fp5-ISTAG.htm] (accessed 18 September 2003).

Through interaction with ISTAG, DG INFSO gains access to the strategic knowledge it is seeking. ISTAG “is very important because [it has] outlined a concept, a vision” (Com01_14, p. 4). Officials in DG INFSO particularly appreciate ISTAG for its contribution to the development of a common view about where to go with the IST program. Looking at the heterogeneity of the program’s constituency, competing interest constellations, and the difficulties of realizing the convergence of technologies, ISTAG’s value added for DG INFSO consists in speaking with a concerted voice about future trends in the Infor-mation Society.

The large value of ISTAG in 2000 was to build a consensus around the fact that this ambient intelligence became the key paradigm. . . . You need some visionary people that have enough power, convincing power like ISTAG, to convince the Commission and the research community that this was the paradigm for research in the years to come. (Com01_11, p. 3)

68

The key contribution to be given to the IST specific program was to give a common vision for this program. (Com03_8, p. 2)

The ISTAG meetings are used as a forum for the concertation of views and perspectives held by different actors representing different fields of research and sectors in the economy. DG INFSO plays an important role in the concertation of views. By organizing the activities of ISTAG, the administration creates opportunities to tailor the input of the group to its learning needs. Once the members of ISTAG were selected for their technical expertise and authority, DG INFSO established how it would work with them. “There were no prece-dents for how to manage or run ISTAG [before the 5th framework program]. . . . There was no ideal size and no ideal constitution” (Com03_3, p. 1). The administration has discretion in the development of procedures that encourage the ISTAG experts to share their views about the program’s content and direction.

DG INFSO treats the ISTAG experts as partners in a collaborative effort that culminates in the development and exploration of a vision for the IST program. Both parties exploit the scope of their boundary-spanning activities to define the issues to be addressed by ISTAG. The interviews with members of the group and officials of the administration reveal that the process of establish-ing the agenda is interactive. Representatives of DG INFSO address questions or requests to ISTAG, asking for advice or comments on a certain issue. The group’s members also come up with topics that they consider worth investiga-ting. Both sides thereby usually define the starting point for the advice that is considered beneficial in the activities of DG INFSO. The issues to be addressed by the group are put on the agenda after being discussed between both sides.

This is how we define our goal. This is our objective. That is our time frame where we want to do this, a report or something like that. And then these kinds of things are linked with internal things in the Commission: We have to take such and such decisions. . . . For us to influence it, we need that thing [by] then. That is how it works, and it works quite well. (AG5, p. 15)

Within the time frame of the 5th framework program, the broad guidelines for ISTAG’s work are determined in the plenary sessions, which are held on four or five days a year in Brussels. They are “used to discuss the way to go for the ISTAG itself” (Com03_2, p. 2). Specialized working groups complement the plenary sessions. They usually consist of ISTAG members, but the working groups include pertinent external experts with no ISTAG status if their partici-pation is thought to be necessary. But plenary meetings were not sufficient to satisfy the Directorate General’s learning needs. An official explained that DG INFSO “rapidly discovered that we have some issues that need much greater and deeper debate than just a plenary session” (Com03_3, p. 1). It was in this

69

spirit that the idea of having additional working groups was born. The working groups meet ad hoc and produce draft reports and draft presentations, which were then shared in the plenary sessions. “So the working group is only prepara-tory to. . . ISTAG advice” (Com03_3, p. 1). The overall process of cooperation is described by an ISTAG member at some length:

In general we say: “Do you consider that this is useful, this aspect or that it can help?” And they say: “OK, yes, that could be very helpful in this relation.” Or they say: “We would need that you make some paper on this aspect. For example, we want your input on the instruments.” And then we create a work-ing group that analyzes in depth, that presents a draft to the committee. And then we discuss that, and we make a final version of the document that is being sent to the Commission. And so this document in general is quite aligned because . . . [during] the launching of the working group, and even during the running of the working group, there is a person from the Commis-sion [who] helps us basically with the preparation of the draft of the document. So in general, the rapporteur is a person usually coming from the Commission [who] helps us because we don’t have too much time really to dedicate to this preparation of the document. But we try to give the ideas and we try to discuss them. But we want someone preparing the document. (AG4, p. 12)

“On the whole there is a feeling of satisfaction about the efficiency of our work,” said one as ISTAG member, explaining that efficiency means “being heard” and that “this is time well spent” (AG2, p. 5). The members of ISTAG are pleased with the working relationship that has developed between the group and DG INFSO, frequently calling it “excellent” (AG1, p. 10) and “extremely good” (AG5, p. 16).

The positive connotations about the working relationship with DG INFSO were not necessarily expected when ISTAG began operating in 1999. “We were initially skeptic[al] but then positively surprised that the members of ISTAG are taken seriously by the Commission with regards to content. If they wouldn’t have done so, they wouldn’t have used ambient intelligence, no doubt” (AG6, p. 5). The ISTAG members appreciate the participation of the administration’s senior management in the plenary sessions, including the director-general, his deputy, the directors of the distinctive directorates, and a number of heads of unit, because it gives credit to the group. An additional reason for the satis-faction among the ISTAG members is the practical support the group receives from DG INFSO’s F6 unit, “Work Program and Cross-program Themes.” It acts as a secretariat for ISTAG to support the group’s activities in the plenary sessions and the ad hoc working groups. It prepares the minutes of the plenary sessions. The F6 unit also supports the activities of the ISTAG experts in the working groups by providing them with necessary background information and “rapporteurs” from outside the Commission to help them prepare their

70

documents. Figure 4.1 depicts the supportive role of the F6 unit in the organization of ISTAG activities.

Figure 4.1 Support of ISTAG activities by the F6 unit.

The F6 Unit as ISTAG’s Advocate

The support that the F6 unit affords ISTAG stems from the unit’s mandate to coordinate the drafting of coherent work programs. To perform this task, F6 draws on ISTAG’s recommendations to reduce complexity and stake out devel-opment tracks for IST. ISTAG’s advisory reports are seen in F6 “as key recommendations, but we have not to replicate exactly what is written by ISTAG” (Com03_8, p. 9). Another member of F6 mentions that the group’s input “reinforced our position here as a unit in order to give orientations to the other units and toward specific areas of activities in line with ambient intelligence” (Com01_3, p. 13). Using ISTAG recommendations, F6 links the

F6 support * Provision of Information and Background Data

* Provision of “Rapporteur” for ISTAG Working Groups * Preparation of Plenary Sessions

* Drafting the Minutes of Plenary Sessions

Plenary Sessions

(4–5 a Year)

Ad hoc Working Group

* ISTAG Members

*Representatives of DG INFSO

* ISTAG Members

* Additional External Experts

Ad hoc Working Group

Ad hoc Working Group

71

activities of the group with DG INFSO’s internal decision-making processes that govern the drafting of the annual work programs. In this capacity the F6 unit is the central interface between ISTAG, which represents the “conceptual side” (Com01_6, p. 1), and the officials in the units of DG INFSO’s research directo-rates, who provide the content of the work program.

There was no such thing as a dialogue between the people [who are active in internal strategic design] and ISTAG because all the interfaces went until now by one unit, the F6 unit, so there were quite little direct interactions with ISTAG members. (Com03_1, p. 4)

The [F6] unit prepares the work program, so there is a relationship, . . . a link with the operational units for the preparation of the work program. (Com03_2, p. 8)

The F6 unit ensures that the recommendations of ISTAG are channeled into DG INFSO, and it affects how they are taken into account internally. Looking back, an official of DG INFSO finds it obvious that the unit should speak for ISTAG to the various groups of the administration that are involved in drafting work programs. “F6, of course, [which was] responsible for promoting both what ISTAG [was] doing and how to make use of it, was also responsible for the drafting of the work program” (Com03_3, p. 4). The acknowledgement of ISTAG views on the long-term development of the IST program inside DG INFSO complements the fact that they are channeled into the administration by F6. A member of F6 emphasizes that the unit is accountable for both the transmission and acknowledgment of ISTAG’s advice to DG INFSO:

We forward everything to all directors. OK, so [we] keep them informed on everything which comes from ISTAG to the program, comes through this unit. So we are to some extent the mailbox of ISTAG and the other way around by the way. So we receive information from the directors, and we send it to the ISTAG members as well. . . . This should lead to initiatives taken by specific units. . . . But then we have to explain to ISTAG what happened with the information we have sent and what is the main outcome. So, of course, we need to understand exactly the ways on which these inputs are taken on board. (Com00_2, p. 7)

To summarize the empirical evidence indicating the role of the F6 unit in the boundary-spanning activities between DG INFSO and ISTAG, it is instructive to view the unit as an advocate of ISTAG (see Figure 4.2). F6 greatly contributes to the fact that the recommendations of the advisory group bear directly on action in DG INFSO.

72

Figure 4.2 The F6 unit as an advocate of ISTAG.

Distributing the Knowledge Acquired from ISTAG

To assess whether DG INFSO learns from its boundary-spanning activities with ISTAG, it is necessary to shed light on the factors that affect the dissemination of the acquired knowledge. The analysis of the impact that boundary-spanning activities have on the distribution of knowledge will benefit from a look at the way ISTAG’s activities are linked with the internal dynamics of DG INFSO. The nature of the linkages affects both how and how much the acquired know-ledge passes among the groups in the administration.

The Linkage between ISTAG Activities and

Internal Dynamics of DG INFSO

The legal basis for the administration’s drafting of an annual work program is the European Council’s decisions on the 5th framework program and the IST program. However, to adopt a final version of the annual work program within the 5th framework program, the directors of DG INFSO must take a formal decision in consensus with ISTC—the committee representing the Member

F6 supports ISTAG activities

F6 draws on ISTAG advice

F6 is the inter-face between ISTAG and DG INFSO.

F6 channels ISTAG advice

into DG INFSO

F6 ensures acknowledgement of ISTAG advice

ISTAG has impact on action in DG INFSO

73

States, the accession countries, and the associated states. Before that step occurs, DG INFSO has room to organize the drafting process. The opportunity to imple-ment the IST program along the lines of an annual work program enables the administration to anticipate changes in technology, markets, and socioeconomic contexts, including trends in industrial competitiveness. Because the develop-ment of the program content encompasses a massive amount of internal and external consultation, new knowledge is incorporated from different sources and at different levels. The result is perceived as a “delicate balancing act [in DG INFSO] between. . . the objectives of the framework program and the specific program. . . on the one hand and what the outside world, the constitu-ency, wants to see there on the other [hand]” (Com03_7, p. 1).

To understand the factors that affect the dissemination of ISTAG’s input to DG INFSO, it is worthwhile to characterize the internal decision-making process with which ISTAG’s activities are linked (see Figure 4.3). A major issue as the work program is drafted is that of increasing DG INFSO’s awareness of trends and developments in the field. The drafting of the work program illustrates the open attitude that officials in DG INFSO have toward new information. “It is a situation where you have a lot of input into one thing. You are not going to miss. . . too many things in that process” (Com01_6, p. 6). Receptiveness to the field and its actors is common in DG INFSO. “Trying to talk with as many organizations and as many people as possible, in order to get a feeling of where this field is going to” (Com01_2, p. 2) is considered a normal thing to do. “You have to be able to go to the outside world,” as another official commented (Com01_13, p. 13). New information circulates frequently via e-mail, in corridor meetings, or at the coffeemaker, for example. “If you are work-ing [in DG INFSO], you have to be an informed person. There is no way to stay on an island of not knowing inside. . . a sea of knowledge” (Com03_5, p. 3).

74

Figure 4.3 The main DG INFSO actors involved in drafting the annual work program.

The first step in drafting the work program is consultation with the pro-gram’s constituency. The specialized units in DG INFSO’s research directorates (Directorates B, C, D, and E) set up meetings to inquire about the needs and preferences of the research communities they deal with, and the topics and audience of those meetings reflect the specialization of the units. The objective of these consultation meetings is. . . to draw conclusions as well about possible actions to anticipate future needs (Com01_3, p. 4).

Directors of DG INFSO

F6: Work Program and Cross-program Themes * Coordinates Editorial

Board * Coordinates ISTAG

Activities

Directorate E Essential Information Technologies a. Infrastructures

Unit E1 – E6 Organize Focused Consultation Meetings

Directorate D IST: Content, Multimedia Tools, and Markets

Unit D1 – D5 Organize Focused Consultation Meetings

Directorate C IST: New Working Methods and Electronic Commerce

Directorate B IST: Systems and Services for the Citizen

Unit B1 – B5 Organize Focused Consultation Meetings

Unit C1 – C5 Organize Focused Consultation Meetings

Horizontal Dir. F IST: Integration a. Implementation

75

We use those consultations. [They are] not just cosmetic. . . . We follow what the people are saying, what the experts are saying, and we try to use [these encounters] as much as possible to discuss. . . a new work program. (Com01_4, p. 4)

The units then produce consultation reports based on the meetings with the research communities. That task can be more difficult than expected, however. The “danger [of consultation meetings is] that you might end up with a report that [has] a little bit of everything and. . . no definite line of producing” (Com01_6, p. 6). Having the reports establish priorities and a focus is therefore as important as trying to “have a good representation of the field” (Com01_2, p. 4). Referring to past consultations with the pressure groups that give input, one official explained that no unit was “obliged to take all their opinions on board, but it was very good to know what their opinions were because it was a type of sanity check. . . . But at the same time we would have to make our own decisions” (Com03_7, p. 6).

The consultation reports of the units are synthesized within the directo-rates in order to come up with the entire directorate’s specific contribution to the drafting of the work program. To define the interests that the directorates wish to see reflected in the editing, their coordinators of the work program remain in close contact with their respective directors and with one representative from each contributing unit. Hence, communication and negotiation are essential for arriving at common directorate-wide input to the drafting of the work program. As one work-program coordinator underlined,

[t]here is a lot of, I would call it, win-win capabilities that you need to apply here in order to get your colleagues with you to support this and to say, . . . “Yes, this is our work program.” And, of course, they send it out to other people. They consult[ed] their projects, and they got some paragraphs from here and there which they wanted to add. It was not all their input. (Com03_5, p. 8)

Each directorate’s contributions to the work program are pulled together in the editorial board, also called the drafting team, which is in charge of writing a first draft of the work program. The editorial board consists of the work-program coordinators, who try to represent the interests of their directorates, and repre-sentatives of the F6 unit, which is responsible for coordinating the activities of the editorial board. As at the intradirectorate level, the great challenge to co-operation within the editorial board is to align the different and sometimes conflicting views of its members. “I would say that this editorial board worked very well, [although] it is a difficult process because you get different [and] sometimes conflicting opinions, in particular from the units. They have to report positions of their directorates, positions of their units” (Com03_2, p. 11).

76

Another official, however, is less positive about coming to terms with the differences in the editorial board:

I really think that this work is hell, this consolidation process of things that do not always fit together but have to fit together. And everybody has to agree. . . . It is not done by a director but . . . by one person who is not necessarily authorized. (Com03_4, p. 9)

In this vein the role of the F6 unit in the editing process is subject to some scru-tiny in DG INFSO.

It was a role which in the beginning made everybody unhappy because they were editing a lot or cutting down or bringing in things that you had submitted but leaving half of it or cutting it down. And, then, you had to insist that these things were kept because otherwise it would not make sense and all that. . . . We had a lot of fights in the editing board group; yes, lots of fights also between the directors because the editing board was represented by people from each directorate. The things could not be solved in this editorial board, so the decisions had to be taken by directors. (Com03_5, p. 9)

As this description indicates, the F6 unit reports to the directors of DG INFSO, who gather in the directors’ meeting each week. The unit proposes a first draft of the updated work program to the editorial board and the directors but lacks formal authority to decide on the content. However, “everything else is an adjustment of the very first strategy that is taken” (Com03_7, p. 5). F6 consoli-dates the various contributions that feed into the work program, “holding the pen” in the editing process:

So they [pull] this together with my director and some support from the operational sectors of the five directorates.8 But essentially someone has to hold the pen, so it comes down to these few people . . . in the process of writing or distilling information, in passing it on that certain facts and figures have to be put in, certain facts have to be left out. It’s in the mind of the author, for the purpose he is writing, what should go forward. Some informa-tion [is] passed [on], some [is] left behind. (Com01_7, p. 10)

Because F6 proposes a first draft of the updated work program that emerges from the consultation cycles inside DG INFSO, it plays a key role in the decision-making process bearing on the program’s content. A large proportion of the administration’s officials therefore tend to believe that the major decision

8 The interview partner refers to the director of the horizontal directorate F "IST: Integration

and Implementation – Networks and Future Technologies". Directorate F of DG INFSO complements the activities of the directorates B, C, D and E that are named after the four key actions of the IST program.

77

on the content development of the IST program during the drafting of the work program is taken “in F6 and ISTAG and whoever they talk to” (Com03_9, p. 7).

Facilitators of and Barriers to the Distribution

of the Acquired Knowledge

The factors that affect the distribution of ISTAG’s recommendations within DG INFSO when the F6 unit begins drafting the annual work program are inherent in this process. The drafting of the program is described as a “common effort [in DG INFSO] because you could not recognize the original idea any-more. It was a lot of, let’s say, reformulation and drafting effort that went into this[, and] you cannot say that the text is clearly coming out of one person’s mind” (Com03_5, p. 8). The process “[operates] at different levels, [so] ideas tend to get filtered and turned, and the more pivotal ones tend to rise to the top” (Com03_7, p. 2), that is, they acquire more weight than others do. The decision-making process for the drafting of the work program is therefore not entirely transparent for the officials who contribute to its development. Not all actors in the drafting of the annual work program participate equally in deciding on the incorporation of the input. It is a “process that tries to get consensus from everybody but within the restrictions you have” (Com03_4, p. 9), such as a prescribed number of action lines that match with the key actions of the IST program. The drafting of the work program as a common effort by various groups across the hierarchies of DG INFSO thus greatly helps disseminate the advisory group’s inputs.

A second, related facilitator of this knowledge-sharing is the DG INFSO discussion about the implications of adopting the vision of ambient intelligence for the program’s development. Discussing the vision’s impact is part and parcel of the vision’s diffusion within the administration. In the process of talking about ISTAG’s views, officials try to forge links between ISTAG’s advice and their years of cumulative knowledge and experience. Discussing the vision of ambient intelligence contributes to its contextualization. Given the status of ISTAG’s vision in DG INFSO’s internal discussions, however, there is clearly disagreement about its relevance for the content development of the IST program. The recommendations by ISTAG, for instance, are sometimes equated with the input coming from other pressure groups:

78

[It] is just background information for us . . . and these things all together form a framework of knowledge which enables the people to make intelligent decisions. (Com03_5, p. 15).

The vision is one of the inputs we get; it is not the thing we will follow and [say] “This is where we go.” . . . It is one of the inputs we have besides other inputs like reports on ongoing technologies, like other roadmaps of technologies that some others were writing. It is more an aspect that we have also in mind. (Com03_4, p. 1)

A third factor facilitating the distribution of ISTAG input within DG INFSO is the top-down flow of information about issues discussed by the meetings the two organizations have with each other. Directors and members of middle management who attend them decide how to share those pieces of information with their subordinates in the directorates, who may act selectively on what they circulate.

You need to have a good boss that is forwarding the information to you. . . I am quite happy because my boss is forwarding information that is of general use to everybody. . . . It’s more than you can really handle. It is just too much often. But I prefer . . . receiv[ing] . . . a lot of information out of which I can choose [what] to take. (Com03_5, p. 4)

Although the drafting of the work program along the hierarchies in DG INFSO’s various directorates fosters wide distribution (and interpretation) of ISTAG’s advice on the long-term development of the IST program, the sharing of ISTAG input across the directorates encounters barriers. There is no evidence that DG INFSO has procedures or opportunities that enable a horizontal cross-fertilization of knowledge within it. An additional factor that complicates and discourages the sharing of ISTAG views across DG INFSO’s boundaries is the specialization into distinctive fields of inquiry to be addressed in the adminis-tration’s research units and directorates. The following quotations from inter-views with officials convey the barriers to the horizontal sharing of the ISTAG views within DG INFSO:

There are no systematic efforts in this DG to have overall presentations of developments in particular fields. . . [in order] to understand what the neighbor is doing. We don’t have that. We are just concentrating on our own activities, most of the time also because of a lack of time and resources. . . . The average project officer or head of unit or whatever does not have a full view of what is going on elsewhere. (Com03_6, p. 10)

79

In my opinion perhaps there should be more interaction. . . . We have not enough meetings in the directorate. We have enough in the unit, but not on a bigger scale. . . . It is very important because in the directorate we are doing research. . . I know what is happening in my unit, but I do not know what is happening to these people in other units, I don’t know exactly what is their strategy, and so on. OK, I can go to their website, but it is not so easy to follow things like that. (Com01_4, pp. 12–13)

Interpreting the Knowledge Acquired from ISTAG

The wide circulation and varied interpretations that the knowledge acquired from ISTAG enjoys within DG INFSO do not necessarily mean that the differ-ent groups in the administration share the meaning that the F6 unit ascribes to ISTAG recommendations. Although the ISTAG vision of ambient intelligence serves as a major guide for the drafting of the annual work programs, its interpretation within DG INFSO illustrates that the administration is far from being monolithic. The understanding of how ISTAG’s views are applied within DG INFSO will benefit from an analysis of the way in which the interpretation stage and the politics of organizational knowledge are intertwined.

Making Sense of the Acquired Knowledge

within the Context of DG INFSO

DG INFSO officials are frequently called upon to make sense of what they per-ceive as remarkable, disturbing, or in other ways noteworthy signals in the administration’s environment. The issue is not only to become aware of relevant trends in the environment but also to find suitable responses. “This thing about sense-making is in a certain way constantly the question that comes up. . . . It is not just signaling. The question is also what you do with it as an organization” (Com01_14, p. 15). However, it is difficult to act selectively on the sometimes contradictory information that enters DG INFSO, a problem that also makes it difficult to define appropriate responses.

De facto there seem to be less and less precise criteria [for] being able to select information. You could say it is a gut feeling based on the cumulative experience of a core network of people. (Com01_12, p. 5)

Being almost constantly aware of what is happening in the field yet having only narrow opportunity for these perceptions to be reflected at levels of aggregation higher than the research unit sows a feeling of “permanent instability”

80

(Com01_13, p. 16). The vision of ambient intelligence as a guiding force for drafting the work program does not resolve this quandary.

In this context the F6 unit lends weight to ISTAG and plays an important role in ascribing a guiding function to ISTAG views. By organizing the activities of ISTAG and serving as the interface between it and DG INFSO, F6 creates opportunities to influence what DG INFSO as a whole can expect from its advisory group. F6 thereby helps tailor ISTAG’s advice to what it perceives as the learning needs of the Directorate General. But the empirical evidence shows that F6 does not establish the meaning of ISTAG’s views once and for all. The understanding of the advisory group’s views on the long-term develop-ment of the IST program within DG INFSO shows that the administration cannot be treated as a unitary actor. ISTAG’s views of ambient intelligence, for example, also have ascribed meanings other than that of being a guiding function for the program as a whole. A major issue in the discussions about the vision of ambient intelligence is to realize

from where it is coming because this clearly came from Philips. So there is some company culture also behind it. There are some corporate interests also behind it, and you have to always see [that] when you discuss how far this should influence yourself. You should know what interests are behind [this concept] and what industries. (Com03_4, p. 1)

The vision of ambient intelligence is criticized for its bias toward home electronics and entertainment, for it is of little practical use in work environ-ments. “It doesn’t cover everything in the IST program. In retrospect, we will probably see that the vision of ambient intelligence is more like information technologies in everyday life. [It will be found] especially in homes but not in work environments” (Com03_6, p. 2). Another official points out that the ISTAG vision does not measure up to its promise; it does not give guidance in structuring the IST field. The vision of ambient intelligence is considered by some people to be a fuzzy term with limited relevance for the great bandwidth of research activities to be carried out within the IST program.

I strongly believe that there is no IST vision. . . This vision doesn’t apply to the full information society applications. It. . . originated [in] consumer elec-tronics manufacturers’ view and the specific situation of the home user. So, when you try to expand it and to speak of ambient intelligence as an overall structuring vision, either it totally breaks down—that is, it becomes such a fuzzy term that everybody can just say: “Oh, I am doing something that con-tributes to ambient intelligence”—or it generates relatively absurd results. . . I think I can even prove that it doesn’t exist[, for] I can prove that many of the individual visions that are implemented are contradictory with each other and have not been reconciled or arbitrated in the sense that if a vision. . . structur[es], then we will make choices to the one thing and not another. (Com03_1, p. 1)

81

It is likely that questioning the vision of ambient intelligence as a cohesive ele-ment of the IST program will create a gap between the official rationale of the program and DG INFSO practice.

The Impact of Organizational Politics

on Knowledge Interpretation

Although the vision of ambient intelligence is the IST program’s official ration-ale guiding content development as the work program is drafted, and although it is pivotal in funding decisions, the voiced skepticism about it’s capacity to integrate the program and structure the IST field calls for examination. The espoused way of doing things clearly coexists in DG INSFO with critical perceptions of the vision of ambient intelligence. To understand the absorption of ISTAG’s vision, it will be helpful to analyze the interpretation stage as being intertwined with organizational politics. The empirical evidence reveals that the resource, process, and meaning dimensions of organizational politics cause the vision of ambient intelligence to be absorbed in DG INFSO primarily as a communication issue of little practical relevance.

The F6 unit draws on the recommendations of ISTAG to win authority in the editing process of drafting the work program. Conversely, the unit therefore has a stake in establishing ISTAG’s views as a major guide for the content development of the IST program; it stands to gain weight among the multiple actors in this process. In particular, the intention of entrenching the vision of ambient intelligence as the lodestar for drafting the work program is advanced by F6’s immediate access to the directors of DG INFSO and the ISTAG experts. F6 uses the plenary sessions of ISTAG for this purpose and builds coalitions between the administration’s directors and the ISTAG experts in order to produce agreement on a rationale that binds the different parts of the program together. The involvement of the ISTAG experts in the decisions on the program’s strategy—a prerogative of the administration’s directors—plays an important role in the adoption of ISTAG’s views by DG INFSO.

Looking back on the initial unawareness of ambient intelligence as a vision when it arose in late 1999, one official sums up this strategy: “The key issue was to sell the idea to the directors that there was something in it for them” (Com03_3, p. 3). By the end of 2000, just one year after the vision of ambient intelligence had been introduced, it “had become a mainstream philosophy, and [in] that sense you [get] things more easily done if you link to that concept” (Com01_8, p. 11). Another official echoes this assessment, stating: “If you start from that particular focus, then it is much easier to get your view through because you can always argue that this is what we are after, this is the overall view of the program” (Com01_2, p. 8). The directors now take note of ISTAG’s

82

vision because it entails the promise of furnishing a common vision for the program that the ISTAG experts agree upon. The ensuing recommendations of the group “could cross the organization, could cross the different interests of the directors” (Com03_8, p. 2).

Invoking the vision of ambient intelligence results in desirable action, particularly on contributions to the drafting of the work program. Despite the considerable criticism within DG INFSO about the aptness of the ISTAG vision as a cohesive framework for the IST program, it remains the official rationale that guides the substantive development of IST. In early 2003, when DG INFSO started to implement the follow-up IST program within the 6th framework program, pragmatic use of the vision of ambient intelligence was common practice among officials in the administration. “When you try to write something and you want to get it through, you tend also to make a relation to the ambient-intelligence vision” (Com03_4, p. 2). Another official mentions at the same stage of the program that the vision of ambient intelligence “now has made quite an impact on the thinking. . . . It is a guiding vision for the whole of the program and everybody is recognizing it” (Com03_5, p. 18).

The Organizational Memory and the

Knowledge Acquired from ISTAG

To understand how the absorption of ISTAG’s views affects the behavior of DG INFSO, one must examine this process in light of the administration’s memory. The pragmatic use of ISTAG’s vision of ambient intelligence within DG INFSO indicates the existence of established beliefs and practices in mana-ging and developing the program. The administration’s officials tend to relate ISTAG views to the knowledge they have built up with colleagues in their immediate work environment. The final section of this chapter presents a discus-sion of how that repository of organized knowledge saves the administration from questioning its basic assumptions in the process of absorbing ISTAG’s views.

Challenges to Established Structures in DG INFSO

The adoption of ISTAG’s vision of ambient intelligence as the guide for the content development of the IST program when the annual work program is being drafted affects the description of the different contributions to that pro-cess. When the discussion is about the “work program or about next activities or what is necessary in technology, then, of course, this concept is part of [it]”

83

(Com03_4, p. 2). Using the vision of ambient intelligence as a core concept in the annual work programs also influences the communication with the pro-gram’s constituency. A project officer notes that he and his colleagues “are involved daily in discussing the Commission’s vision with our customers, with the [members of the] research community in Europe who want to know where our focus is, . . where they would be more likely to be successful in bidding for funding for research, . . . [and] what topics to choose” (Com03_9, p. 2).

The internal discussion about the program-related implications of adopt-ing the vision of ambient intelligence is related to the external communication with the constituency of the IST program. The internal consultation that accom-panies the drafting of the work program has been described as the “internal selling job” (Com03_7, p. 4) prior to the external selling job

because . . . the normal staff in INFSO [has] to sell this product, which is the work program in a way. They are dealing with all their constituents and all the outside meetings, all the information days, and so on. So first you’ve got to have a buy-in by them. . . . People need to feel that they have actually contributed to the work program to be able then to sell it as being something. . . . You know they have their part-ownership. (Com03_7, p. 4)

Be that as it may, the vision of ambient intelligence in the internal and external discussions has effected only limited change in DG INFSO’s underlying beliefs about goals and objectives of the IST program. Reference to the vision is pri-marily a communication issue with little relevance for how the program is conducted.

Where you find visions is at another level of granularity, that is, in the objectives certainly of units and possibly of directorates. And then what they do is they try to, in the worst case, paint, in the best case, describe their vision in relation to this big label. So that has actually generated at some point some frustration in ISTAG in the sense that they were saying: “Well, you endorse our vision, but we don’t recognize it in the actual implementation.” (Com03_1, p. 1)

The lack of effect that the vision of ambient intelligence has on the behavior of DG INFSO is inherent in the operational aspects of the program’s implement-tation. Running the IST program entails the evaluation and selection of research proposals and the contracting and monitoring of research activities. It is therefore

not just the technology they [the research communities] are developing and what is happening in the technological world, but also how they perceive the difficulty of making a proposal and making the contracting and things like that. That’s the other side of the coin if you want. . . . There is the techno-

84

logical side, but there is also the operational side of implementing the program. (Com01_10, p. 2)

Project officers in DG INFSO, in charge of operating the IST program within the prescribed confines in Comitology and the 5th framework program, insist that the ISTAG vision is less of an issue in their realm.

We are aware of trends in the evolution of technologies and ambient intelli-gence[.] . . . I don’t think that it fundamentally changes our work. . . . It probably changes our priorities and focus. (Com03_6, p. 2)

I wouldn’t put it as a very strong guiding force because what the implemen-tation of the program is about . . . is . . . details and about what we are exactly going to fund and what we are not going to fund. . . . I think if we had another vision, I don’t necessarily think that we would be doing something totally different today in our work programs. In other words, I think it is very good to have a vision, but the way we work doesn’t necessarily fit into the mode of having a vision. (Com01_6, pp. 7–8)

The management of the so-called cross-program actions draws attention to the factors in DG INFSO that restrict the impact that ISTAG’s views have on the operational aspects of implementing the IST program. The administration taps the advisory group for ideas for developing themes that cut across the program’s key actions and action lines. Cross-program actions demonstrate that DG INFSO has almost no experience with measures that do not fit its organizational struc-ture. The difficulties with conducting cross-program actions reveal the establish-ed beliefs and practices that underlie the internal fragmentation of the administration.

DG INFSO’s commitment to overcoming these problems is clear, how-ever. With cross-program themes accounting for 12 percent of the budget allotted to the IST program, considerable resources are devoted to the support of its integration, the underlying convergence of technologies, and multidisci-plinary research. The €435 million allocated to cross-program actions almost equals the size of a directorate’s budget within DG INFSO. The official website of these activities states that a total of 19 distinct themes have been developed in the four work programs of IST during the 5th framework program, resulting in the launching of 244 projects.9 The F6 unit suggests themes to be called as part of the cross-program actions and organizes the evaluation of the incoming proposals under a given call. In an effort to conquer the “predominance of a directorate-, unit-structure which is contradictory with an implementation of things that cut across units” (Com03_1, p. 10), F6 uses the recommendations of

9 See [http://www.cordis.lu.ist/cpt/home.html] (accessed 9 May 2003)

85

ISTAG to act on the initial intentions of cross-program actions. A member of the unit explains:

One of the things that ISTAG did was to point out that we shouldn’t constrain research to the way we were organized. We should be more open in the way we accepted research ideas and then find ways in-house how to manage them. (Com03_3, p. 6)

Project officers from different units are gathered in a team, with one of them assuming the role of “baton holder,” or team leader, to monitor research projects called under cross-program actions. In total, 75 project officers are involved in such cross-unit teams.

Carrying out cross-program actions requires a certain openness and flexi-bility among the project officers and heads of unit involved. “From time to time they [the officials involved in the teams] meet together and . . . find synergies between what they were doing” (Com03_3, p. 6). The management of cross-program actions through teams that span DG-INFSO may eventually increase understanding of approaches and methods for monitoring projects and may foster cross-fertilization among the units. It may also help strengthen linkages between units and may prompt questioning of the territorial thinking in DG INFSO.

But during the 5th framework program, the difficulties of implementing cross-program actions seemed to outweigh the prospective opportunities. The management of cross-program actions is considered to be of only moderate success in DG INFSO. “There was a very strong will to have these cross program actions, but this decision was not translated in organizational terms” (Com03_1, p. 10). The management of cross-program actions is at odds with the persistence of “territorial thinking” (Com03_4, p. 17) and “turf fights” (Com03_7, p. 8) in DG INFSO. The departmental boundaries in the adminis-tration’s structure are a serious barrier to engagement in cross-cutting activities. The lack of a matrix structure is a frequent complaint in DG INFSO when the issue of cross-program actions is raised: As one project officer commented: “There should be a matrix-type of organization. . . . But the whole organization is not built for that, with all its bureaucratic decision lines and whatever” (Com03_4, p. 14). The present structure would foster the existence of “very separate territories which do not necessarily reflect what is going on outside” (Com03_6, p. 2). Moreover, heads of unit may “not [be] willing to look beyond their little box” (Com03_3, p. 13) and to support their project officers by granting them the discretion to act beyond the boundaries of the unit. The existence of strict demarcation lines between units has a political dimension that partially accounts for the limited success of cross-program actions.

86

These cross-program activities are based on the matrix organization. And the matrix organization works only if people have the power—vertical power in the units, for instance—[and] cannot use their power against these cross-program activities. (Com01_3, p. 11)

Inertia in DG INFSO

The adoption of the ISTAG vision of ambient intelligence as a guide for devel-oping the content of the IST program and strengthening cross-program actions in an organizational climate of territorial thinking and working draws attention to the difficulties of changing the organizational memory of DG INFSO. The fact that the vision was adopted before it was introduced into the units that must implement it has impeded its use as a starting point for developing a coherent frame of reference in discussions of activities across the units and directorates of the administration. During the 5th framework program, DG INFSO therefore continued to operate in a fragmented way. The administration’s memory acted as a template for interpreting the vision of ambient intelligence and acting upon it when the annual work program was being drafted and when cross-program actions were conducted. The absorption of the ISTAG vision and ensuing recommendations was, and still is, affected by the internal fragmentation of DG INFSO. Incorporating this vision, however, did not have the intended effect of softening boundaries inside DG INFSO.

When the newly formed DG INFSO took up its activities in 1998, it was internally as fragmented as the IST program, itself the product of the merger of ESPRIT, ACTS, and TAP. The convergence of technologies may have been sound reasoning for that consolidation, but once it had been decided upon by the Council of Ministers and the European Parliament, reality had to catch up. Bridges had to be built, sectors had to be crossed, and multidisciplinary research had to be fostered in the relevant research communities by an administration that lacked the background to do so. When the implementation of the IST program began, DG INFSO had to develop a background that would make it a driving force bringing the technologies together. Building such an administration “was a very ambitious step” (Com00_4, p. 11).

The formidable tasks and implications of the Directorate General’s birth were one of the major concerns in the Five-Year Assessment Report related to the IST program by Pompidou et al. (2000). They pointed out that “the effort and difficulties of merging three quite different cultures, with different work practices, into a single unified and efficient administrative mechanism were clearly underestimated by those concerned” (p. 20). An official of DG INFSO has a similar perspective on the administration’s situation in 2000:

87

The way in which the three programs. . . worked in the past was extremely different. So the way we work, think, et cetera is extremely different. . . . The way we address the same problems are very different. If you speak with people who are involved in communication technologies and others in software, the same word doesn’t mean the same thing. So you’ll speak actually about the same thing, but you’ll reach. . . completely different conclusions. (Com00_2, p. 12)

Like the structures and rules of DG INFSO, its personnel embodies organiza-tional memory. These people practice it through their ways of doing things and dealing with certain problems or issues, taking some approaches while avoiding others that are not within the practiced and established frames of reference. In 2003, four years after DG INFSO was formed, officials pointed out the continued existence of different working practices that impede communication and coordination across departmental boundaries in the administration:

How reviews are done, how projects are monitored, there are different cultures going. It depends if you come from ESPRIT, from ACTS, from Telematics. . . . I mean they got a bit more streamlined, but you still see a lot of different ways of how projects are monitored, depending on where the people came from. (Com03_4, p. 15)

To change habits is per se extremely difficult, and in addition to that you have always the not-invented-here syndrome. (Com00_2, p. 3)

The very organization of DG INFSO helps keep the different program cultures alive. The units and directorates are structured according to the content of the IST program. They address the distinctive action lines and key actions of IST, and it is these specialized fields of attention that distinguish them from their neighbors in the same or other directorates. “The basic filter is the specific areas in which the units work” (Com01_12, p. 15), as a head of unit explains. The officials of DG INFSO tend to consolidate their views on and experiences with the actors and developments in the IST field mainly at unit level. The demar-cation lines between units and directorates help them keep on doing what they were doing in predecessor programs. In other words, the structural arrangement of DG INFSO preserves its organizational memory.

On content, the integration has not worked at all, that is, [in terms of the] definition of objectives, what you are trying to do in a technology field, et cetera. And this is also reflected if you look at the composition of staffing units. You will see that in practice. For example, the ACTS units moved as a whole to become new units in the program. . . . There was not much mixing of people. (Com03_1, p. 14)

88

Despite the major changes that led to the program-related and administrative integration, “maybe 80 percent of the people are still in the same job” (Com00_5, p. 5). The majority of the units keep their personnel and, hence, their inherited memory irrespective of the merger. As an official of DG INFSO notes, the experience with the actors and developments in the field is “mainly something which is our legacy, the knowledge we have built up and how it relates to communities and the economy and the policies we want to adopt” (Com01_10, p. 6). The continuity in the staffing of the units helps them to uphold their identity over the years. The same personnel tend to approach the same constituency with the same topics irrespective of the changes from one framework program to the other.

I can still tell colleagues that I am working for the software unit. They know what it is because it has been there for a long time, since the beginning of ESPRIT. It also has an identity to the outside, the so-called constituency. And it remains as such. That is, people who used to talk to the former unit know that we are still around. (Com03_6, p. 9)

Summary of the Main Findings

This case about the boundary-spanning activities between DG INFSO and ISTAG, consisting of experts from industry and academia, illustrates how the development of a vision triggers processes of organizational learning. The ISTAG recommendations that revolve around the notion of ambient intelligence are perceived as a vision by DG INFSO personnel because those recom-mendations create an overall goal, offer orientation, and provide a basis for the different groups in the administration to work in the same direction. The attention that ISTAG’s advice receives indicates that the group has a strong impact on DG INFSO. Lacking the ability to implement the intended strategy of the IST program, which is to foster the convergence of IC technologies in an environment that is known for its increasing complexity and dynamism, DG INFSO is willing to adopt the ISTAG vision of ambient intelligence.

The administration limits its ISTAG-related knowledge-seeking behavior to knowledge about long-term developments in the IST field. It uses the meetings with the ISTAG experts for joint exploration of the multiple and some-times competing views and perspectives about the content and direction of the IST program. Thanks to the strong linkages between the ISTAG activities and the decision-making processes within DG INFSO, the broad guidelines that are agreed upon in the ISTAG meetings with senior management are disseminated widely within the administration. However, DG INFSO personnel does not equally share the ISTAG vision of ambient intelligence as a cohesive vision for the program. The extent to which the strategic intent of the vision of ambient

89

intelligence informs organizational action is limited because there are few, if any, opportunities for persuasion, negotiation, and advocacy regarding the vision’s impact on the program across the administration’s units. Organizational politics plays an important pragmatic role for the vision of ambient intelligence within DG INFSO. But because territorial thinking and working prevail within the administration, the vision has little relevance for how the program is carried out.

90

5. Learning from Boundary-spanning Activities

with ISTC

As noted in chapter 4, the implementation powers of DG INFSO are embedded in an institutional environment that is shaped by the EU Member States of the European Union. The analysis of the boundary-spanning activities between the administration and ISTC as a trigger of learning highlights the influence of the normative obligations that take on a rule-like status in the learning behavior of DG INFSO. Looking at ISTC as the embodiment of the Directorate General’s institutional embeddedness contributes to the understanding of how learning processes relate to the efforts of supranational organizations to gain legitimacy among the governments of the Member States.

The Role of ISTC in Managing the IST Program

The IST program is part of the European policies on RTD. The implementation of the European research programs is regulated by Comitology,10 the Amsterdam Treaty (1997), and the framework program of the European Com-munities. The ISTC, by virtue of its composition and its oversight function, serves as a link between DG INFSO and the national governments of the EU Member States in matters of implementing the IST program.

The legal basis for implementing the program sets the frame for the agency of DG INFSO. The administration receives a mandate and the resources to carry out the IST program and to contribute to its development for new initiatives in this area. Simultaneously, there are constraints on the adminis-tration’s latitude for action because the Member States want to ensure that DG INFSO acts in their interests. DG INFSO must gain approval among the representatives of the Member States for a number of steps that concern the implementation of the IST program. As stipulated in Article 148 (2) of the Amsterdam Treaty, the Council’s decision on the IST program specifies the conditions under which a majority vote of the ISTC is required for the approval of initiatives to be taken by DG INFSO:

10 See the "Procedures for the Exercise of Implementing Powers conferred on the

Commission" in the Official Journal of the European Communities, 28.6.1999 (L 184/23) – 1999/468/EC.

91

• Preparation and update of the work program, including the content of the calls for proposals

• Approval of RTD actions proposed for funding in cases where the estimated amount is €1.5 million or more

• Preparation of the terms of reference for external assessment

• Adjustment of the indicative breakdown of the IST program budget11

On matters not included in the aforementioned list, the members of ISTC also have the opportunity to express their views, though no formal vote in the decision-making process. This rule relates in particular to the provision of information about the overall progress of the IST program and the other research programs of the 5th framework program that are managed by the Directorate General for Research of the European Commission. As described by a national delegate, the scope of ISTC’s activities through DG INFSO is broad:

We are involved in the definition of the program. We are involved in the definition of the call, when to launch a call, what kind of call, for what action lines, [and] what activities. “Yes, our people are ready to do this.” “No, they are not ready, it is too early.” . . . And then in the review of the returned proposals to say: “Yes, it is OK.” “It is not OK.” “Why did you return this proposal?” “It seems not to be fair.” And we have hard discussions with the Commission. (PC4, p. 8)

In 1999 the ISTC adopted its own rules of procedure for its meetings, complying with Article 7 (1) of the Comitology rules, which pertains to each oversight committee consisting of Member State delegates:

• Upon request for the committee’s opinion, the chairperson draws up an agenda and attaches the draft measures.

• The agenda, copies of the draft measures requiring an opinion of the committee, and a notice convening the meeting must reach the delegates not later than 10 days before the date of the meeting.

• If a considerable number of amendments are made to the draft measures, the procedure for voting on them may be deferred to the end of the meeting.

• A majority vote of the committee is defined in Article 148 (2) of the Amsterdam Treaty.

• The committee may decide to hear experts on specific points, but they are not allowed to participate in the committee’s deliberations.

11 The Council decision on the IST program is published in the Official Journal of the

European Communities, 12.3.1999 (L 64/20) – 1999/168/EC.

92

• The delegates must state their opinions or positions on draft measures in the meetings that are to be recorded in the summary record and hence to be taken into account.

• The committee is supported by a secretariat provided by the Commission.

However, to understand the ISTC’s part in the management of the IST program, it is necessary to look beyond formalities. Granted, DG INFSO’s autonomy in implementing the IST program is curtailed by rules that mandate the involvement of the ISTC in certain instances, but those rules do not define the nature of the committee’s participation in the management of the IST program. DG INFSO must engage in discussions with the ISTC members to facilitate committee’s work.

In formal terms it would be sufficient to submit a document to the committee where approval is required. Yet chances are that they would not give a positive majority vote. So that is a little bit of a problem. We don’t do that of course. You’ll have to prepare them, involve them in the discussions. (Com00_1, p. 4)

In the eyes of one national delegate, the ISTC operates at more than one level:

The cynical view is the Member States are there as far as the Commission is concerned to do one thing. And that is to say “yes” at appropriate occasions. . . . So on that basis you would say that the role of the committee is in fact very limited because all it has to do is say: “Yes, yes, and we’d like you to do [it].” . . . So it is on the surface not much of a role, and I think underneath there is quite a lot of activity. (PC2: 13–14)

Prescribed Learning Needs

The consultation of ISTC follows the regulatory procedure specified in the Comitology rules, which pressures DG INFSO to listen to the national delegates on the committee. If ISTC does not approve the measures envisaged by the administration, there is a procedure to resolve the ensuing conflicts. Failure to reach agreement with ISTC means that DG INFSO must submit the measures to the Council and the Parliament, which may result in operational delays of up to three months—a long time in an annual work program. However, DG INFSO and ISTC succeed in preempting conflicts. Because of time pressure and other factors, the regulatory procedure was never applied during the implementation of IST in the 5th framework program.

93

You never ask them [the members of ISTC] for an opinion unless you are quite sure it is favorable. . . . If they ever refuse on an opinion, it is a disaster. There are procedures there that will lead to very, very lengthy delays before anything gets agreed. . . . The psychology of dealing with them: A lot of preparatory work is done. We keep them very well briefed in all stages. We provide them with information, and we have working parties . . . exploring with them on their feelings et cetera, et cetera. So when it comes to the formal business of what they need to do, they are very well up-dated and we are pretty sure that something will go through. (Com00_4, p. 3)

DG INFSO’s need to ensure the support of the Member States for the imple-mentation of the IST program particularly relates to the annual updating of the work program. Only if the Member State representatives adopt the work program can the Directorate General engage in the other steps of conducting it. “One of the decisions we [DG INFSO] can make is a work program. But the Member States must agree to that work program before we can implement it” (Com00_3, p. 3). The members of ISTC must agree with the description of the topics, the priority-setting, and the definition of the budget allocation in the work program before the calls for proposals can be published. Hence, the subsequent evaluation of the submitted proposals and the monitoring and evaluation of the selected research activities are also contingent on the ISTC’s consent. In a sense, then, DG INFSO “can only act as good and efficient as we are supported by the Member States” (Com01_1, p. 12).

The institutional embeddedness of DG INFSO implies that it acts as a politicized bureaucracy sensitized to the concerns and preferences of the govern-ments of the Member States. The necessity to become aware of the issues that matter to the governments in the implementation of the IST program is rooted in the regulatory framework of DG INFSO. That legal basis essentially prescribes what the administration needs to learn from the activities that span the boundary between DG INFSO and ISTC. DG INFSO must respond effectively to that need in order to reduce the risk that ISTC will obstruct the administration’s implementation of the IST program. The following section presents an analysis of the impact that these prescribed learning needs have on the administration’s knowledge-seeking behavior and on the absorption of the acquired knowledge.

Acquiring Knowledge from ISTC

The knowledge-seeking behavior required of DG INFSO because of its institu-tional embeddedness reflects the administration’s identity as a supranational organization intended to serve the interests of the EU Member States. DG INFSO must inform itself about the different approaches to RTD that are fostered in the Member States. The existence of prescribed learning needs raises

94

questions about the extent to which DG INFSO develops agency in the creation of openness to the national delegates to ISTC.

Gaining Access to Policy-oriented Knowledge

Becoming aware of the views that the Member State delegates have on the IST program helps DG INFSO “identify how to please each of them, and [it] is the only way to get a positive vote from everybody” (Com01_3, p. 10). The national delegates are treated as knowledgeable agents representing the interests of their governments as the work program is defined and funding decisions are taken (see Figure 5.1). The inputs of the national delegates address “more policy-oriented matters” rather than the “technique-oriented, technical-issues-oriented input from ISTAG” (Com01_3, p. 4). They “bring along their emphasis,” having a “very good feel for the issues that concern their countries the most” (Com01_6, p. 4). Having ISTC is therefore “actually the best way for us to maintain the links between the national interest on policies and the program activities” (Com01_3, p. 2).

Figure 5.1 Dialogue with the Information Society Technologies Management Committee

(ISTC) as a platform for knowledge acquisition.

The policy focus of DG INFSO in its knowledge-seeking behavior toward ISTC comes especially to the fore as the work program is drafted. The national delegates are expected to provide the administration with knowledge about the concerns and preferences of their governments in the definition of the work program. Increasing the awareness of Member State policies that are similar to IST is part of the effort to develop complementarities between the activities to be funded within IST and like initiatives in the Member States.

DG INFSO

* becomes aware of Member State Concerns and preferences and

* responds to delegates’ inquiries.

ISTC

*communicates concerns and preferences of Member State governments.

95

What we expect from them [ISTC delegates] is to keep us informed on the development of new activities in the Member States because we need to know exactly what is happening in the Member States and to develop on this basis complementarities in the work program and in the strategy of the work program. (Com00_2, p. 3)

It becomes a bit more mundane when you are talking with Member States because they have a certain vision of how things could be from their countries’ point of view. And then you would see the work program as being a compromise to what their own vision for technological development in their country is. (Com03_7, p. 4)

The members of ISTC and officials in DG INFSO are all intent on having the IST program correspond with similar initiatives at the national level. A delegate points out that “when it is possible, they [DG INFSO] try to take into account the opinions, particularly when the opinions are expressed by a number of Member States” (PC5, p. 16). By keeping “a careful record of what is coming in” (Com00_4, p. 9), the administration shows the delegates that it takes seriously the representation of their governments’ views on the development of the IST program. The intention is to have the ISTC members see the changes in the work program that are made in response to their feedback. As a delegate notes, however,

looking at all those papers, it is very difficult to say that they have not taken into account all those matters. Because the problem here, of course, is that the IST program is very broad in content, and therefore they can very easily say that they have taken into account all the requests and viewpoints. . . . The weakness, of course, is . . . if you want to really create competitive industries in Europe. It’s very difficult [for] the program [to] support these matters if we have very many areas which we want to support using these available funds. (PC6, p. 8)

The national delegates to ISTC provide DG INFSO with knowledge about the interests of their governments when funding decisions are taken. This timing implies that most of them have a major interest in supporting IST program measures that foster industrial competitiveness and innovation in their individual countries, which is not necessarily the same as the broader picture of Europe that DG INFSO is concerned with.

We all want our own national companies to do well. We want to get as much money as we can out of the Commission. . . . We all get on quite well together in the group, but there is also this element of competition. (PC2, p. 1)

96

Committee members want to promote or support their own countries only. So it’s a very national way to handle these matters. A very large part of the committee discussion is a very useless discussion to be exact. (PC6, p. 10)

There is quite some lobbying for the interests of your country[’s] participants. And actually this kind of representation of your country could be misleading the activity because in that case the national percentage or the national participation would be against the professional character of the given call. (PC1, p. 8)

As a result, “some committee members take a very positive interest in the way their countries’ actors are represented” (Com01_13, p. 18) and tend “to look for something which is called the juste retour. . . . In practice, Germany always gives more than it gets back. . . . It [the distribution of the funding] is skewed in favor of the smaller countries” (Com03_7, p. 3).

Limits to the Knowledge-seeking Behavior

of DG INFSO

Although DG INFSO involves ISTC in the determination of who gets what in the program, the administration does not discuss with the delegates their views on trends in technology, competitiveness, and fitting measures for RTD. As one administration official portrays the relationship, the members of ISTC do not attend the meetings with DG INFSO “à titre personelle” (Com00_1, p. 4); they fulfill an official mandate, and they are equipped with a number of rights for this purpose. The knowledge that accrues to DG INFSO in its boundary-spanning activities with ISTC is deemed relevant if it relates to the committee’s moni-toring of the program’s management.

The empirical evidence reveals that the limits to the administration’s knowledge-seeking behavior create some frustration among the delegates. The ISTC members perceive themselves to be reduced by DG INFSO to adminis-trators and representatives of their governments to the detriment of their “inter-esting backgrounds in science, technology, and strategy . . . and experiences like having run a national program” in the field (PC2, p. 13). As one delegate observes, “it is true that the committee deals primarily with administrative tasks, but it also approves a work program, doesn’t it? So in the ISTC there is definitely expertise available” (PC3, p. 16). Regarding the preparation of the IST program to be included in the 6th framework program, for instance, another delegate notes that DG INFSO “didn’t analyze the industrial situation. All these discussions, they were at the technology level. . . . They [DG INFSO staff] didn’t have a very clear understanding of the industrial situation in the different countries” (PC6, p. 7). In light of these considerations, the ISTC members tend

97

to question the strong technology focus in the vision of ambient intelligence. They see a need to incorporate more socioeconomic thinking in the program’s vision in order to “really affect the needs of the industry” and “define in an appropriate manner their needs” (PC1, p. 6).

The way the ISTC is seen by DG INFSO officials managing the program limits the administration’s knowledge-seeking behavior. The perceptions of the committee correspond with the learning areas to which DG INFSO gains access when these two bodies interact. The national delegates provide the adminis-tration with the knowledge necessary for coming up with acceptable proposals, and DG INFSO gets what it expects from ISTC. The ISTC’s role in defining the annual work programs, and the relationship between that role and the knowledge that the committee provides, could be described as follows:

ISTC . . . as a group [is] in charge of making sure [that] what is proposed in the content of the work program is compatible with the national priorities. . . . So these representatives know the priorities of their ministries, and they want to make sure that this is reflected by the work program, as it is defined in such a way that the European funding reinforces the efforts which are made by the Member States. (Com00_2, p. 2)

As for the definition of the work program, ISTC is seen in DG INFSO as the interface with the Member States. The committee delegates give the adminis-tration input “on what initiatives there are going in their own countries or sometimes on initiatives the Commission has not yet proposed and Member States would want to propose” (Com01_9, p. 5). Once the work program is approved, ISTC oversees the implementation of the research activities that are selected for funding. It is responsible for ensuring that DG INFSO applies the program’s rules and procedures correctly. To one committee member, the ISTC’s evaluation of the incoming proposals is therefore regarded as a “safety belt” (PC4, p. 8). “If the things are correct, you can trust the committee and the things are correct” (PC4, p. 12). This view on ISTC as a control board, particu-larly when funding decisions are prepared or taken, is widely shared in DG INFSO:

ISTC is more of a formal thing to approve things. It is not so much to find new ways. . . . It is more to control, to make sure that we are following what we should in the implementation of the program. (Com01_4, p. 4)

The task of ISTC consists in approving the draft measures of the Commission. . . . Regarding the implementation, I would rather see them as a control board. (Com01_1, pp. 10–11)

The dominant view in DG INFSO that ISTC represents the interests of the governments of the Member States encapsulates the perceptions of ISTC as an

98

interface and a control board. When a vote of the committee is required, the delegates, according to one official, “will ultimately vote the way they are formally supposed to do so by their ministers” (Com00_1, p. 4). Another official points out that “ISTC as an entity is probably not that valuable. . . . The indivi-duals in many cases are part of the administrative fabric in their own countries” (Com01_13, p. 7). The tendency among officials of DG INFSO to treat ISTC as the sum of individual delegates whose interests need to be accommodated is related to the proclivity of ISTC members to express themselves in terms of domestic interests. The interviews with the national delegates reveal that they are very well aware of different occasions when, for example, the “Portuguese wasn’t satisfied” or “France blocked a decision” (PC1, p. 9). Bringing knowledge about the developments in the IST field to bear on the management of the program appears to be detrimental to the delegates’ role as representatives of their governments: “There I do not sit as an expert. I sit there as a repre-sentative of my government” (PC5, p. 12).

The issue of leadership in ISTC is an additional factor that limits DG INFSO’s knowledge-seeking behavior. A DG INFSO director chairs the committee meetings. In that capacity he or she influences how the prescribed way of doing things is acted upon and, hence, determines what counts as rele-vant knowledge in this context and what does not. Because the chair leads by means of the various points on the agenda of the meetings, that person plays a major role in clarifying the sometimes conflicting views of the delegates and the way to a compromise. “He has to get the cooperation of a group of people in front of him,” so “there is both a formal and an interpersonal side” (Com00_4, p. 6).

It is the chairman who discusses with all the Member States [whether] is there any Member State against a proposal or not, . . . and so we have a discussion [about] why, what can we do, etc. And that is it. The chairman says: . . . “Do you accept what is proposed?” If everybody says “Yes,” he says: “Now we pass to the vote.” And if there are no further concerns about the proposals, it is OK. So you have an informal discussion before each important decision. (PC5, p. 11)

The turnover of the committee’s chairpersons during the 5th framework program shows that the personality of the chairperson affects the nature of the dialogue with the committee. The first chair (1998–2000) had a tendency to “sweep controversial issues under the carpet[;] . . . you wouldn’t get answers to your questions,” as one delegate observes (PS3, p. 17). By contrast, the same delegate notes that the successor (2000–2002) was “more explicit and open,” allowing more opposition in the committee meetings (PS3, p. 17).

The dominant tendency among officials and delegates to use the ISTC meetings as an opportunity for domestic interest representation results in ISTC

99

input that is more diverse than the strategic recommendations of ISTAG. “The Member States are unanimous once in a blue moon” (PC2, p. 10). The com-mittee does not forge a common opinion on issues that relate to the IST program. A delegate explains that “a common opinion means in practice that committee members express their own opinion, . . . and when summing up that opinion it is the committee’s common opinion” (PC6, p. 10). The lack of a “procedure to create really a common opinion” in ISTC (PC6, p. 10) increases the significance of DG INFSO, especially that of the committee’s chair, in influencing the terms of reference when a compromise needs to be reached. The administration’s role in leading the ISTC meetings, combined with the emphasis on domestic interest representation, reduces the scope of the committee’s input to the formal provisions of their working relationship.

Channeling of Policy-oriented Knowledge

into DG INFSO

The tendency of the DG INFSO officials and national delegates on ISTC to act within the prescribed boundaries of their formalized working relationship and therefore to engage primarily in rule-guided behavior affects the way in which the inputs from the delegates are channeled into the administration. The impact of the prescribed limits in this regard is indirect. The formal committee meetings set up, but do not actually facilitate, the incorporation of the concerns and pre-ferences represented by the government delegates. The formalized working relationship between DG INFSO and ISTC affords the actors opportunities to incorporate and acknowledge the delegates’ inputs in decisions that require the committee’s approval.

Every time there is a meeting in Brussels with delegations, there is a lot of time in between meetings where a lot of discussions are going on. So first of all, you are picking up information all the time from Member States informally, from the corridors. . . . We could not have a situation where we just went to a meeting once a month and just discussed in an open meeting the issues of the day and walk away. We would not be informed. We do not learn very much in a three-hour meeting, believe it or not. We don’t learn enough; we need to have constant dialogue. (Com01_6, p. 5)

The formal powers, you need them in order to have the informal influences over what goes into the work program and how the evaluations are done and so on and so forth. I think the main contributions that we make are the informal ones because the formal ones are essentially trivial. It’s just saying “Yes.” (PC2, p. 14)

100

The interviews reveal that informal discussions and negotiations between both sides complement the voting procedure and the other formal arrangements. Informal committee meetings, corridor meetings, e-mail, phone calls and, as one delegate explains, “informal wine-tasting and dinner” with officials (PC2, p. 14) are created ad hoc and with no formal rules in place to find an agreement. Conflicting views and interests are reconciled on a bi- or trilateral basis in an informal atmosphere rather than in the committee meetings. Officials and dele-gates stress that a “continuous dialogue” with the Directorate General acting as a “consensus seeker” (Com00_3, p. 2) distinguishes the achievement of consensus when a work program gets drafted. “During the course of that, . . . views change and we begin to understand what is important [and] what is less so through the discussion” (Com00_4, p. 5). The importance of informal encounters equally applies to the way agreement is reached about funding decisions. The interviewed delegates are more explicit about this part of the communicational interactions than their counterparts in DG INFSO are. ISTC members speak frequently of bargaining and negotiating to steer the decision-making about the selection of proposals:

Some people of the Commission say: “What will you say in the committee? Will you say ‘No’ on this project? Will you say ‘Yes’? What do you want? What is the problem? Why is there a problem?” And we have a negotiation. . . . And when we are in front of the committee, yes, we agree. But all the problems have been treated before, outside of the committee, in bilateral or trilateral meetings or between different partners. It is the classical way of work. It is a corridor meeting. What we say [is that] we have a corridor meeting. We have an Internet meeting. We have a lunch meeting. We have a lot of things. (PC4, p. 11)

The delegates emphasize the importance of personal relations for the informal side of the communicational interactions with DG INFSO. They “cannot change the rules,” as a delegate notes, but “the personal relation can make the communication easier” (PC1, p. 10).

One of the things you discover is that it is much easier to do things informally than it is to write letters to the Commission and so on because formal requests get formal answers. And it’s much easier and much better to try and persuade people that your idea is their idea. . . . In that way one develops a sort of a colleagueship, a collegial approach, a group of colleagues getting together and trying to decide what is the right thing to do, what is the rational thing to do. . . . It is fairly open because we usually say . . . you need to try and involve as many people as possible, get as many Member States as possible in these things. (PC2, p. 15)

101

I went a lot to project officers, heads of unit, directors just to talk in general about IST, not to further some kind of tangible interest, but to build up understanding and to learn. And I believe that this was beneficial because the people knew me, and I knew them. And I believe that these informal channels are very, very important, you get a lot of things done. (PC3, p. 12)

Time is important in creating and nurturing personal relations. It is essential to figure out who could be a potential ally in DG INFSO and among the committee members from the other Member States. Personal relations are a resource in the hands of ISTC members, and those who have a long history in committee participation tend to have the advantage of being heard by DG INFSO. Inter-views with delegates illustrate both ends of the spectrum. The first person cited below has long-term experience with the Commission and is satisfied with how the personal relations evolved. The second person, whose history of relation-ship-building is briefer, sees a need for improvement in this respect.

I don’t know what is the case among other Member State representatives, but some people from the Commission are now friends to us, we have lunch together, we invite them, we have a good contact with people. But it takes two or three years. (PC4, p. 10)

Networking and these informal networks between different people in Europe, I think that this is our weakness. . . . We have very much to learn in this area. . . . We have to know our key counterparts in many countries and know also Commission people, what they are doing. (PC6, p. 13)

The importance of personal relations and the informal nature of channeling new knowledge into DG INFSO create informational asymmetries among the delegates in ISTC. Knowledge about the issues that concern the governments of the Member States accrues to the administration because it acts as “a broker for many network nodes, so it is an intermediary, you might say. And it provides a forum for people to speak, so it is a political body, too” (Com03_5, p. 15). DG INFSO tends to know more about the concerns and preferences of the Member State governments than the sum of the individual delegates in ISTC. The incorporation of this kind of input enables the Directorate General to generate agreement between the Member States when their approval is required and, hence, to take the lead in decision-making about the program’s direction and budget allocations. But the strong reliance on informal channels for the incorporation of the delegates’ views hampers the administration’s ability to learn collectively from ISTC. If conflicting interests between Member States and DG INFSO are addressed mainly in an informal atmosphere, the issues at stake tend to remain in the minds of the individual delegates and officials involved and are not shared by a wider audience in the administration. Exami-nation of the distribution and interpretation of the policy-oriented knowledge

102

that accrues to DG INFSO from its boundary-spanning activities with ISTC sheds light on the extent to which the administration learns.

Distribution and Interpretation of the Knowledge

Acquired from ISTC

ISTC’s capacity to block DG INFSO’s implementation of the IST program, and the resulting need for DG INFSO’s decisions to reflect the program-related interests of Member States, involve mechanisms to be kept in mind in any consideration of distribution as a means of reducing uncertainty in DG INFSO. Analysis reveals that the internal dynamics in the administration are buffered from the knowledge acquired in the process of building awareness of those interests because the linkages between the ISTC’s mandate and DG INFSO’s internal dynamics are not conducive to wide knowledge dissemination. Because it is important to make sense of knowledge in the internal context of organiza-tions, the next section will also draw attention to the relevance that the acquired knowledge has to officials of DG INFSO.

Barriers to the Wide Distribution of the

Acquired Knowledge in DG INFSO

The knowledge to which DG INFSO gains access through its boundary-spanning activities with ISTC is not widely shared in the administration. The strong reliance on informal encounters to win the delegates’ approval of the annually updated work programs and the related funding decisions indicates the lack of explicit efforts, procedures, and rules to expedite the wide distribution of the acquired knowledge among the realms of DG INFSO. The result is that the policy-oriented knowledge provided by ISTC members does not feed into internal decision-making processes that bring various groups of the adminis-tration together. ISTC activities and DG INFSO’s internal dynamics are missing a crucial link that would make the wide dissemination of the acquired know-ledge possible.

This lack comes into play particularly in the gaining of approval for the annual work programs. Many officials in DG INFSO are not involved in the processes of finding a compromise with the delegates. The officials working on project management, the project officers, do not identify compatibility between the content and direction of the IST program and similar initiatives in the Member States. They are not involved in the informal communication that prepares the positive vote of the committee for the proposed work programs.

103

Agreement with the delegates comes through people at senior and middle-management levels of DG INFSO, including directors and heads of unit. A project officer therefore notes that “the level of priority about policy design is dependent a lot on your position with regard to policy design” (Com03_1, p. 8).

There are units like F6 or F7 or groups working with directors, heads of units who are concentrating on policy issues. As a project officer, I am dealing to a large extent with operating the projects. . . . The time which we spend on policy as project officers, on that kind of issue, is limited, so I don’t see the individual Member State delegates. (Com03_9, p. 3)

Normally, I am not involved in ISTC meetings. Directors go there, advisors. I don’t see them in my normal work. (Com03_6, p. 8)

The operational steps after the definition of the work program—the evaluation and selection of proposals for funding—do occasionally open up possibilities for project officers to talk directly with the members of ISTC. Project officers recognize ISTC by at this stage of the program’s implementation because the committee has an oversight role.

The only contact at the PO [project officers’] level you have with ISTC. The real direct contact, is . . . after you have project selection and you have been responsible for a project area. Then there is this process that the projects are selected. The selection choices are presented to the ISTC. (Com03_4, p. 12)

But the direct contact with the national delegates is not used as an opportunity to tap new knowledge. In this realm ISTC is perceived mostly as a control board, and the issue at stake is whether the selection of proposals for funding is conducted as prescribed. The communication with the delegates about the operational tasks of DG INFSO centers on the regulations of the program.

So they [project officers] have to give a justification. They have to prove . . . it was fair [that] they [the proposers] were ruled out because they were not participating in a good team or whatever. So these things are raised again before a contract can be given. (Com03_5, pp. 16–17)

We are not supposed to influence the opinion of experts, but we are supposed to be sure that the process implements with the rules including, for example, considering the objectives of the work program as they are written in the work program. (Com03_1, p. 8)

104

It is important that the evaluation and the selection that you have done [are] done in the correct way, and then it is very easy to defend it. . . . I mean they [ISTC] don’t have a problem with that because they also see [whether] it is done correctly—and not only done correctly but also documented that it is done correctly. (Com03_4, p. 12)

The project officers respond to the inquiries of the delegates through clarifica-tion, justification, and explanation. DG INFSO’s responses to the delegates’ inquiries about the regulatory aspects that govern the operation of the IST program are essentially outward-oriented communication. ISTC’s dominant per-ceptions in the program’s management are important to this kind of communi-cation with the committee. As representatives of the Member States, the delegates are perceived as an interface with and a control board of the govern-ments, as monitors of DG INFSO’s implementation of the IST program. The delegates’ inquiries are not seen as challenges to the established modus operandi of project management but rather as indicators of where the reasoning behind operational decision-making needs to be better explained to those who act at “arm’s length from the Council” (Com00_1, p. 4). Because the administration looks at its boundary-spanning activities with ISTC as a way to gain legitimacy among the Member State governments, it provides the delegates with information that helps them oversee the administration’s use of its imple-mentation powers.

The ISTC secretariat is an additional organ that provides the national delegates with information about the program. It is located in the F7 unit (“Operational aspects of the program”) of DG INFSO and helps the committee carry out its formal responsibilities. F7 looks mainly at the administrative and logistic aspects of program implementation and is the operational contact point for the DG INFSO directorates dealing with research, ISTC, the proposers, and other services of the Commission such as the Directorate General for Research. One of the secretariat’s responsibilities is to provide sufficient information to the committee. An official, for instance, points out that “if the ISTC members have questions, difficulties, problems, this is their first address” (Com00_1, p. 9). The provision of information is important for smoothing out the operational steps that require a vote by the committee, as the following quotation from an inter-view with a national delegate demonstrates. The bone of contention in this case is the funding of selected proposals:

At the beginning of the program we wanted to know exactly the financing of the partners and the projects. The Commission said, “No, we have problems.” We said, “We don’t vote on the proposals and we plug the system.” And the Commission gives us all the details now. We receive a CD ROM after each call with all the data bases. Not all the programs do the same. (PC4, p. 10)

105

Table 5.1 presents the elements that keep DG INFSO from wide internal dis-semination of knowledge about IST-related issues that concern the Member State governments. It shows that the perceptions of ISTC correspond to dis-tinctive modes of communication and points to the underlying buffering mecha-nisms in the administration. The conjunction of these factors reduces the risk that ISTC may hamper the administration’s implementation of the IST program. DG INFSO creates agency from the committee’s oversight role in the management of the IST program by (a) demonstrating the acknowledgement of the governments’ concerns when defining the work program, and (b) showing that the decisions on the funding of selected proposals are made legitimately.

Table 5.1 Uncertainty reduction at two levels and its constituent elements.

Levels of

uncertainty

Perception of

ISTCa

Communication

with ISTC

Buffering mechanism

Approval for

IST work

program

Interface with the Member State govern-ments

Informal com-munication between senior and middle management and delegates

Division of labor between senior management and middle and lower management for reaching consensus with delegates

Monitoring of

IST imple-

mentation

Control board in the operations of the program

Outward-oriented communication at project management level and in ISTC secretariat

Bureaucratization of evaluation and selection procedure

a Information Society Technologies Management Committee.

The analysis of the distribution stage in processes of organizational learn-ing triggered by the boundary-spanning activities between DG INFSO and ISTC shows that the creation of barriers to the wide dissemination of knowledge is related to the absorption of uncertainty. The rise of impediments in DG INFSO to broadly based understandings about governmental concerns and preferences stems from the administration’s efforts to reduce the risk that ISTC will block the implementation of the IST program. If the knowledge gained from the boundary-spanning activities with ISTC is narrowly disseminated in DG INFSO, one may question the extent to which the interpretation stage in this learning process is relevant.

106

Relevance of the Acquired Knowledge in DG INFSO

Organizational learning triggered by the boundary-spanning activities between DG INFSO and ISTC is narrow, for the administration creates barriers to the wide distribution of the policy-oriented knowledge to which it gains access. A major reason why this kind of knowledge is not widely shared among DG INFSO officials lies in the factors that limit its relevance to project manage-ment (see Figure 5.2). A main one is the administration’s division of labor in obtaining approval for the work program among the delegates. Many DG INFSO officials have only a vague idea of what policy-oriented issues concern the governments of the Member States and how those issues are taken into account. But the officials remain largely unenlightened by ISTC members, for these national delegates are not perceived as having “strong opinions about directions for research and things like that. They relay comments and criticisms from organizations in their respective states” (Com03_6, p. 9), and “most of the time they can’t agree because they all have different interests” (Com03_4, p. 11). The delegates’ views on the program are not considered pertinent when project management officials contribute to the drafting of the work program:

This is a committee that has become more and more of a financial monitoring committee that has an overseeing role in our implementation of the program, but we get very little feedback from this committee now about directions for research. (Com01_8, p. 3)

It wouldn’t make sense [to project officers] if you say, “This would please the UK” or “This would please whatever.” . . . [I]f they are supporting a vision like the ambient intelligent vision, then we relate to the vision. And, of course, what you do is you try to find what you think or what the unit thinks is the best within the strategies or policies or visions. And then you do that, and you see what ISTC is thinking about it. And basically what you see often is that . . . when you talk to them, . . . they, of course, think about their national interests, think about their own industry. (Com03_4, p. 11)

107

Figure 5.2 Factors limiting the relevance of ISTC input.

Particularly when it comes to clarifying the funding decisions, the dele-gates’ views on the program may reach the attention of project officers before contracts are given out to the program’s participants. The ISTC members enter into discussions about the interpretation of the criteria that are applied during the evaluation of the incoming proposals on a given call. Scientific excellence, the European and social dimension, exploitation potential and societal impact, and the management of the given research activity are the general criteria to be applied in the evaluation of the proposals. But the technocratic nature of the decision-making process that leads to the funding of a selected number of pro-posals makes it difficult for the delegates to engage in serious discussions with officials about the aptness of the decisions that are taken. DG INFSO draws on external expertise to evaluate the incoming proposals by means of a peer review system. Their advice provides a basis for an implementation plan written by the administration’s officials and formally proposed to ISTC for discussion. The implementation plan ranks selected proposals in a list that is usually cut off at the point where the allocated budget would be exhausted. The selection of proposals may be subject to some scrutiny because the reasoning behind the selection of proposals for funding is not always clear to the delegates, especially if it does not reflect a given country’s share of proposals to a given call.

Perception that delegates lack technical knowledge

Technocratic basis of decisions on project funding

Dominance of prescribed limits at project management level

Perception of ISTC as control board at project management level

Limited relevance of ISTCinput at project management level

108

One official described the usual procedure during the evaluation of incoming proposals:

So we organize panels of experts, independent experts drawn from the appropriate fields. . . . This technical evaluation leads to a selection of proposals, which we do. . . . We then draw [up] an implementation plan, which is essentially the advice of the technical advice board, but also take on board some more strategic considerations that come essentially from our side. (Com00_4, p. 4)

One ISTC member asserts that the role of the committee in the evaluation and selection phase is limited “because the Commission with the panel [consisting of external experts] is setting up the order of the projects. . . . [I]f you want to change the order, this is not really the duty of the committee” (PC1, p. 8). Citing externally consulted experts “is a standard way for our justification,” notes an official (Com01_4, p. 7), recourse that applies to both the definition of the work program and the evaluation of incoming proposals. The impact that committee members have on the discussions in both instances is limited because they tend to lack the technical knowledge on which the officials in DG INFSO draw heavily whenever they take decisions. The comprehensiveness of the IST pro-gram and the size of its budget make it difficult for the delegates to discuss with the officials of DG INFSO.

The more widely open a program is, the less likely it is that the delegates are experts. They tend to be bureaucrats. . . . I would know much better my own country because I would know all the players. So I would understand exactly the problems in a much better way. If you make a very, very large and very wide program, it is very difficult for one person to be able to discuss with the Commission. . . . The politician will not be able to discuss because he will lose any discussion with the Commission. The Commission will be an expert on what they are saying. Politicians will just say vague things. (PC5, pp. 14–15)

Although ISTC has a mandate to monitor the operations of the program, its impact on the way the program is carried out at the project management level is quite limited in scope. Once the projects are selected and launched, there is barely any room to consider external inputs, including the delegates’ views. Explaining just how much the behavior of officials is constrained by prescribed limits at the project management level, one project officer states:

If you talk about the pure, let’s say, the daily project management work you have, then you don’t have much flexibility[. There are] certain rules, regulations, processes, signatures, and sometimes—as absurd as it might seem—you have to follow those. (Com03_4, p. 6)

109

Another head of unit stresses that “all the bureaucracy and the paperwork . . . is part of the obligations that you take on when you initiate something in this system” (Com01_13, p. 14). Because of the procedural stipulations that have to be taken into account, officials lack the time and energy to go beyond their immediate responsibilities when projects are selected and monitored. However, the regulatory system of the program’s management does not unfold the same way with every project that is underway. Each one entails “individual consortia with individual problems and everything” (Com01_13, p. 14). Being able to deal with those problems necessitates “communication about the operations, which is most communication” at the project management level (Com03_6, p. 10). The members of ISTC do not engage in resolving problems that concern the applica-tion of rules. Such involvement would be at odds with the dominant under-standing of ISTC as a control board for project management. Despite the committee’s supervisory role, then, “the Member State delegates in the management committee [ISTC] know very little about the projects, a lot less than we do,” as a representative of DG INFSO emphasizes (Com01_8, p. 4).

Organizational Memory and the Knowledge

Acquired from ISTC

In terms of organizational learning, the boundary-spanning activities between DG INFSO and ISTC reveal that the need to gain legitimacy among the Member States coincides with a learning process that supplies DG INFSO with new knowledge about governmental concerns and preferences but simultaneously buffers the administration’s internal dynamics from it. Stability in the imple-mentation of the IST program seems to be a function of adaptation to the interests represented by the national delegates to ISTC. This section elucidates the relationship between the stabilizing forces in DG INFSO and its institutional environment, which exerts considerable pressure on the administration to engage in learning. It is shown that DG INFSO draws on its repository of organized knowledge in a way that both enables the administration to create its own space for decision–making vis-à-vis the Member States and still satisfies their expectations.

Stability in the Implementation of the IST Program

The empirical evidence of learning triggered by the boundary-spanning activities between DG INFSO and ISTC demonstrates that a significant part of the administration’s repository of organized knowledge takes the form of com-pliance with the program’s legal basis. It determines the administration’s

110

knowledge-seeking behavior in the implementation and development of the IST program. Accordingly, DG INFSO identifies the issues that concern the govern-ments of the Member States and require agreement with the committee. The delegates are involved to facilitate quick decisions about the definition of the annual work programs, the funding of proposals, and the smooth running of the other steps in implementing the IST program. But reaching agreement along these lines does not entail development of commonly shared policy objectives for the IST program, for they are developed in cooperation with ISTAG and other external experts. The ISTC chair tends to “push the committee toward its administrative role” (PC3, p. 16) rather than use the meetings with the delegates as a sounding board for the development of a strategic perspective on the Information Society in Europe. Whereas DG INFSO uses its knowledge about the issues that concern the governments of the Member States to strengthen its role in the determination of the content and direction of the IST program, the delegates on the committee find it difficult to play an active role in this regard.

DG INFSO builds alliances with powerful players in industry and acade-mia, including the members of ISTAG, to “help the emergence of a so-called common view” (Com00_4, p. 11) in its “main aim to push the industrial and economic structures toward a new paradigm” (Com01_12, p. 1). DG INFSO and the nongovernmental experts that are brought together have the same per-spective on the future Information Society. The administration therefore tries to convince the Member State delegates that its perspective matches their own rather than building a synthesis of the delegates’ views. DG INFSO needs “to convince the researchers and the policy-makers in the Member States that there is an interest to develop such a vision [of ambient intelligence] and to address certain aspects in research” (Com03_8, p. 12). Assessments of the ISTC’s status and potential for influencing the Directorate General’s decision-making differ, however. The delegates regret what they see as their passive role of simply affirming the results of the Directorate General’s decision-making, whereas DG INFSO officials tend to point to the formal mandate of the committee:

I think the best way . . . to gather everybody behind a common goal is that everybody participates in defining the goal. And I think now that the Commission is running the show really with little influence of the Member States. And I think that is not the best way. (PC5, p. 13)

At the end the ISTC is the body that approves the work program, and if they would have been against the ambient-intelligence vision, they could have taken it out of the program. (Com03_2, p. 6)

111

Taking Regulations for Granted

The enactment of regulatory changes modified the administration’s memory. Alteration of the regulatory framework between the 4th and the 5th framework programs had implications for the knowledge-seeking behavior of DG INFSO vis-à-vis Member State delegates. The 4th framework program entailed more formal opportunities for the Member State delegates to bring their views to bear on the management of the IST program’s predecessors, including ESPRIT, ACTS, and TAP. A broader range of activities—the definition of the calls for proposals and the selection of proposals apart from the volume of funding—required a positive vote of the national delegates. An official of DG INFSO puts the changed rules of the game in cautious terms, explaining that “historically they had perhaps more influence than they do now because historically this area of research was three programs. . . . There was much more scope for Member States’ individual interests to be taken into account” (Com00_4, p. 3). By contrast, the Member State delegates tend to look at the regulatory changes more critically than the officials of DG INFSO do because it implies encroachment on their prerogatives.

The Member States have lost most of their decision-making power in the 5th framework program because the rules have changed. . . . That means that many, most of the proposals do not require an opinion. . . . We are informed, of course, but we don’t have a say. (PC5, p. 12)

At the end someone has to decide on the implementation plan, and this is the DG and not the committee. This is [distressing] especially [to] those who participated earlier in ESPRIT and ACTS. At that time the committee had the right to decide. This was the major change between F[ramework] P[rogram] 4 and F[ramework] P[rogram] 5. (PC1, p. 11)

In other words, the pressure on the Commission to respond to the views of the Member State delegates was higher in the 4th framework program than it has been since. In addition to the perceived curtailment of the delegates’ role, the 5th framework program introduced an annual work program that included the decisions on the topics to be called and on the evaluation and selection of proposals. Once approved, that program, by virtue of its very existence, con-siderably decreased the need to understand the issues of concern to the governments of the Member States. An official explains that the “committee used to approve each call for proposals [in the 4th framework program], but because they do it now implicitly by approving the work program, [that approval] is not necessary anymore” (Com00_1, p. 7). The decreased scope of the committee’s influence on the management of the IST program makes it

112

very hard to actually stop them [DG INFSO]. In fact, it is very hard to even block them. . . . [T]he real power is with the Commission on all these things. I mean mainly because if the power was with the Member States, I think it would be very hard to achieve quick decisions on things. (PC2, p. 11)

The tendency of DG INFSO officials to take the regulatory changes for granted indicates that regulation of the program’s implementation establishes the administration’s memory. The repository of organized knowledge enables DG INFSO to create its own decision-making space out of the Member State delegates’ role of overseeing the management of IST. It helps the administration gain access to policy-oriented knowledge that guides compromises with the individual ISTC members and helps confine the committee’s involvement to the prescribed areas. The knowledge-seeking behavior of DG INFSO fits what the administration’s operational core requires, which is to ensure reliable imple-mentation of the program. Stabilizing the operations of the program is a function of pleasing the delegates with decisions that reflect the interests of their governments.

Summary of the Main Findings

Through the governmental representatives who make up ISTC, the boundary-spanning activities that take place between it and DG INFSO bring societal forces into the analysis of learning processes. ISTC is the embodiment of the administration’s institutional environment. The delegates of this committee have an oversight role in the implementation of the IST program and challenge DG INFSO to engage in prescribed learning needs. ISTC may obstruct the DG INFSO’s implementation of the IST program if the administration fails to win the committee’s approval of the annually defined work programs and the funding of research proposals. To forge such agreement among the ISTC dele-gates, the administration must inform itself of governmental concerns and preferences. The present analysis makes the point that DG INFSO handles these challenges in its capacity as politicized bureaucracy. Knowledge about govern-ments’ concerns and preferences is created in the administration and used to achieve approval, but DG INFSO’s internal dynamics are buffered from it in order to stabilize the operations of the IST program.

The administration’s learning behavior is politicized in the sense that the boundary-spanning activities with ISTC center on the incorporation of govern-mental concerns and preferences in decision-making processes where money is involved. Moreover, DG INFSO is keen on satisfying the delegates by making decisions that reflect their governments’ interests. The dominant perception of ISTC as the interface with the Member State governments reflects these sensitivities. Lastly, the senior and middle management of DG INFSO often

113

negotiates informally with the national delegates to gain their approval. As a result, the incorporated knowledge is distributed narrowly within DG INFSO.

Once the ISTC approves the annual work program, the committee moni-tors the operational steps taken by DG INFSO to implement the IST program. In particular, the ISTC sees to it that the administration complies with the program’s legal basis. Compliance by DG INFSO saves the administration from having its conduct questioned by the ISTC delegates, who are formally em-powered to do so. A number of bureaucratic elements in the learning behavior of DG INFSO curtail the relevance of ISTC input at the project management level. They include the outward-oriented communication with the delegates to clarify the technocratic grounds on which the funding decisions are taken, the tendency to emphasize prescribed limits in operating the program, and the prevailing view at the project management level that the ISTC is a control board. The funding decisions are a major concern of the national delegates to ISTC, but their influence on the decision-making process that leads to the funding of research activities is negligible in practice.

114

6. Discussion of the Empirical Evidence in Light

of Theory

The design of this study is based on an analytical model developed to under-stand how the European Commission’s Directorate General for Information Society (DG INFSO), an example of a supranational organization, copes with changes and developments in its environment by learning. The model integrates the notion of boundary-spanning activities into a process model of learning to address a general claim in theories of organizational learning that the efforts of organizations to align themselves with external changes may trigger learning. The concept of boundary-spanning activities emphasizes the interpersonal rela-tions at an organization’s external boundaries as a trigger of organizational learning. Because the boundary-spanning activities of organizations tend to be linked with the internal dynamics of those organizations, the activities may contribute to either a broad or narrow incorporation of knowledge acquired from external stakeholders. The integration of the boundary-spanning activities con-cept with theories of organizational learning provides a framework for examin-ing the conditions and circumstances in the social context of organizations that affect their ability to cope with changes in the environments in which they are embedded. The embedded-process model accounts for how, and under which conditions, knowledge about external changes is transformed into an organiza-tional property. It indicates that the knowledge acquired through boundary-spanning activities needs to be shared and made sense of against the backdrop of established beliefs and practices if organizational learning is to occur. The model rests on two assumptions:

1. Processes of organizational learning differ depending on the quality of boundary-spanning activities between organizational members and external stakeholders.

2. Organizational learning is shaped by, and possibly shapes, situational factors in the social context of organizations, reciprocality that leads to potential changes in organizational behavior.

The aim of this chapter is to explore the plausibility of these assumptions about the relationship between boundary-spanning activities and organizational learning. The main empirical findings of the two case studies presented in chapters 4 and 5 (see Table 6.1) are compared in the light of theory in order to refine this investigation’s theoretical underpinnings, especially the argument that boundary-spanning activities illuminate the conditions governing processes of learning about an organization’s environments.

115

Table 6.1 Summary of the main findings about the impact of boundary-spanning activities on

organizational learning

Learning processes DG INFSOa–ISTAG

b DG INFSO–ISTC

c

Rise of

organizational

learning needs

Involvement of experts from industry and academia is used to increase internal pressure to learn.

Learning needs are prescribed in the legal basis of program implementation.

Knowledge

acquisition

Only the views of technical experts about long-term developments in IST field are included. DG INFSO deter-mines how acquired know-ledge is channeled into it.

Only the identification of governments’ concerns and preferences is involved. DG INFSO shapes the circumstances under which acquired knowledge is channeled into it.

Distribution of

knowledge

Acquired knowledge is widely disseminated during the content development of the IST program.

Acquired knowledge is narrowly disseminated in order to achieve political agreement.

Knowledge

interpretation

Acquired knowledge is used as a resource. It affects decision-making on strategic planning and understanding of desirable action.

DG INFSO’s internal dynamics are buffered from acquired knowledge because it lacks relevance for strategic and operational decision-making.

Organizational

memory Acquired knowledge destabilizes established beliefs and practices.

Stability of activity system dominates knowledge acquisition.

a Directorate General for Information Society. b Information Society Technologies Advisory Group. c Information Society Technologies Management Committee.

Variations in Processes of Organizational Learning

Organizations may respond to changes and developments in their environments through learning. But organizational learning is not contingent on dynamic environments. There is an element of choice in the willingness of organizations to tap into the knowledge of external stakeholders and incorporate it in a way that it affects organizational behavior. The empirical evidence from both case studies reveals that the concept of boundary-spanning activities is a practical analytical tool to observe the ways in which DG INFSO engages in learning about its environment. It shows that both the incorporation of externally acquired knowledge and the creation of barriers to learning are social accom-plishments. The first case—the boundary-spanning activities between

116

DG INFSO and ISTAG, the committee of technical experts from industry and academia—demonstrates that the creation of learning needs is a crucial pre-cursor to the processing of acquired knowledge. To cope with the complexities and the dynamism of the IST field, DG INFSO takes in the strategic ideas of the technical experts. The second case—the boundary-spanning activities between DG INFSO and ISTC, the committee consisting of national representatives of the EU Member States—supports this finding because it focuses on the impact of learning needs, which are prescribed in the societal institutions that govern the administration’s activities. DG INFSO listens to the ISTC’s Member State delegates in order to gain legitimacy for its implementation powers.

But the ISTC case differs from the ISTAG case in that DG INFSO’s internal decision-making processes are buffered from acquired knowledge. The variations in processes of organizational learning triggered by boundary-spanning activities result from the different circumstances under which DG INFSO takes in externally sourced knowledge and relates it internally to established beliefs and practices. To assess how these variations are shaped by and may be shaping situational factors, it is helpful to compare the two sets of boundary-spanning activities in terms of the distinct stages of learning. The systematic comparison sheds light on DG INFSO’s facilitating and inhibiting factors that affect organizational learning.

The Rise of Organizational Learning Needs

The starting point for learning processes in organizations is the existence of a need to learn. The investigation into the role of boundary-spanning activities in organizational learning shows that the involvement of external experts from industry and academia is used by groups in DG INFSO to increase the internal pressure to learn about developments in the environment. The lack of strategic thinking in, and the fragmented nature of, the IST program is defined as a problem that continually needs to be addressed. The perception of ISTAG’s role in managing the IST program reflects this learning need. Created by DG INFSO to gain access to the perspectives that technical experts have on the long-term development of the various communities that populate the IST field, ISTAG is seen as a strategic advisor helping DG INFSO cope with the convergence of technologies by means of an integrated IST program. This type of knowledge is deemed relevant for priority-setting as the annually updated work programs are drafted.

The boundary-spanning activities between DG INFSO and ISTC produce no evidence of a rise of learning needs in the administration. Certain learning needs are prescribed before both parties engage in their boundary-spanning activities. ISTC’s involvement in the management of the IST program illustrates

117

the embeddedness of DG INFSO in its institutional environment. ISTC is a regulatory and regulated body installed to attain stated goals. It monitors the implementation powers that are conferred on DG INFSO by the Council of Ministers according to the rules and procedures laid down in Comitology, the 5th framework program, and the IST program. The national delegates in ISTC are involved to approve the final version of the work programs and to gain support for the attendant funding decisions. Compliance with the legal basis implies that DG INFSO becomes aware of the delegates’ concerns and prefer-ences in the prescribed instances. The administration needs to listen to the ISTC delegates if it is to use its implementation powers, and the delegates want to ensure that DG INFSO acts according to the interests of their governments.

The empirical evidence reveals that the rise of learning needs inside DG INFSO, or lack thereof, corresponds to the ways in which the functions of DG INFSO–ISTAG boundary-spanning activities differ from those with ISTC. The ISTAG case illustrates an emphasis on the information-processing function, whereas the ISTC case illustrates external representation. ISTAG is designed to provide the administration with intelligence about changes and developments in the IST field. The officials of DG INFSO who attend the sessions with ISTAG take the opportunity to tailor ISTAG advice to the learning needs of the administration and to make it available to those officials who do not participate in the meetings. The acquired knowledge is processed internally to affect how the program is conducted. Gaining access to new knowledge is not the focus of the boundary-spanning activities with ISTC but rather a means to win the committee’s approval. ISTC is set up to monitor the performance of DG INFSO in the management of the IST program. The administration gives account of its management tasks to the ISTC delegates to assist them in their monitoring function. By doing so, DG INFSO creates opportunities to influence their assessment of how well it meets their expectations.

The comparison of the two cases of boundary-spanning activities along each of the stages in learning processes (acquisition, distribution, and inter-pretation of knowledge; and organizational memory) sheds additional light on the impact that functional differentiation has on organizational learning. A look at the flow of the advice from ISTAG to DG INFSO demonstrates that the knowledge acquired from interacting with external stakeholders cannot be treated as a neutral commodity. The way the knowledge is handled within DG INFSO is affected by where it comes from and how its social implications are interpreted. The analysis of the flow of information from DG INFSO to ISTC in terms of organizational learning shows how DG INFSO obliges the delegates with decisions that reflect the interests of their governments. DG INFSO projects an image of a responsive administration that complies with the rules and procedures set forth by the Member States.

118

Knowledge Acquisition

DG INFSO acquires knowledge from external stakeholders through its boundary-spanning activities with ISTAG and ISTC. ISTAG provides the administration with the views that experts have on the long-term developments in the IST field. ISTC helps DG INFSO develop its understanding of the con-cerns and preferences that the governments of the EU Member States have regarding the IST program. There is evidence that other informational resources, too, are gained through the boundary-spanning activities that ISTAG and ISTC each have with DG INFSO. For instance, some ISTAG experts comment on the germaneness of the instruments for RTD that are used in the IST program. And from experience with similar initiatives in the Member States, a number of ISTC delegates are skeptical of this program’s socioeconomic ramifications. DG INFSO’s awareness is restricted to the learning areas of strategic ideas for IST and to the interests of the governments when it comes to the program’s definition and budget. In the boundary-spanning activities, what are the factors that contribute to the filtering that takes place as knowledge is acquired? How are ISTAG’s strategic ideas and ISTC’s concerns and preferences that are repre-sented by the national delegates considered relevant to DG INFSO’s activities?

Comparing the two cases of boundary-spanning activities indicates that regulatory aspects have a major part in determining the relevance of knowledge from external sources. The flexibility of ISTAG’s involvement enables DG INFSO to shape the boundary-spanning activities according to its learning needs and to decide how it uses the group’s advice for the purposes of program management. The relevance of ISTC’s input for the IST program is established in Comitology, as is its use in the management of the program. Rule-guided behavior dominates the interaction of those involved in the boundary-spanning activities between DG INFSO and ISTC and blocks communication of inputs that do not fit the system. Because the thinking of the actors involved is formatted in procedures and regulations, new perspectives and ideas tend to be equated with unwelcome disruption of routines.

The regulatory aspects of the boundary-spanning activities that DG INFSO has with ISTAG and with ISTC enable the administration to achieve certain goals and objectives attached to the involvement of each. The differences between the objectives underlying the involvement of ISTAG and ISTC in the management of the IST program affect the administration’s knowledge-seeking behavior. Consisting of experts from industry and academia, ISTAG is designed to provide technical knowledge to DG INFSO, whereas the provision of know-ledge through the boundary-spanning activities with ISTC is a means to a pre-defined end. ISTAG assists the administration with the definition of the content and direction of the annual IST work programs. DG INFSO’s acknowledgement of ISTAG’s advice is voluntary. Because ISTC monitors the implementation of

119

the IST program, DG INFSO’s acknowledgement of the issues that concern the governments of the EU Member States is mandatory. The administration must agree with the national delegates on the final draft of the annual work program and must consult them when funding decisions are taken. ISTC may obstruct DG INFSO’s implementation powers if the administration does not succeed in pleasing the delegates with decisions that reflect the interests of their govern-ments.

The objectives of the boundary-spanning activities are related to the underlying orientations toward ISTAG and ISTC in the management of the IST program. ISTAG is seen in DG INFSO as a strategic advisor to the adminis-tration, whereas ISTC is widely perceived as an interface with the Member States and as a control board. Viewing the two bodies through these lenses helps DG INFSO pursue certain interests in its knowledge-seeking behavior. Lacking the internal resources to develop overall goals by itself, DG INFSO turns to ISTAG for strategic ideas on the program as a whole. The difficulty is not a lack of knowledge about prospective tracks of development in the IST field but rather the young and fragmented administration’s incapacity to pull these bits and pieces of knowledge together. Beyond the expertise and the ideas that DG INFSO gains access to through its boundary-spanning activities with ISTAG, there is also the authority of the group, which bears on the need to create a rationale for decision-making about budgetary allocations.

DG INFSO’s recognition of the national delegates in ISTC as know-ledgeable agents is confined to the few areas where the committee may hinder the implementation of the IST program, and defining the underlying strategic ideas of the annual work programs is not one of them. However, ISTC does play a pronounced nonexpert role in the management of the IST program, helping DG INFSO create its own decision-making latitude and competence vis-à-vis the Member States as it draws heavily on the technical knowledge from ISTAG and other experts. DG INFSO’s knowledge-seeking behavior toward ISTC is therefore aimed at keeping this committee to its formal function and at creating a certain degree of autonomy. DG INFSO is not interested in either developing a common view with the delegates about IST and European competitiveness or openly discussing with them their views on the program in the committee meetings.

Critical views of DG INFSO officials challenge the dominant perception of ISTAG as a strategic advisor. The expertise and independence of the ISTAG experts is questioned, as is the relevance that a guiding vision like ambient intel-ligence has for the wide range of the IST program’s key actions. The findings of this study therefore indicate that DG INFSO is not a monolithic culture in which everyone accepts the strategic value of the ambient intelligence vision for the IST program as a whole. There are also critical accounts of ISTC’s role, but they are mainly voiced by the national delegates rather than by officials in

120

DG INFSO. A number of national delegates would like to go beyond their formalized role in the management of IST, acting as advisors to DG INFSO with a say in spelling out the objectives and socioeconomic ramifications of the IST program. But because DG INFSO does not see a need to discuss these issues with the delegates, the opportunities for the ISTC delegates to do so are minimal.

The limiting factors in both sets of boundary-spanning activities (DG INFSO–ISTAG and DG INFSO–ISTC) affect how openness to both bodies is achieved and, hence, how the knowledge gained enters the internal dynamics of the administration. Table 6.2 summarizes how DG INFSO copes with the differences in degree of regulation, objectives, and underlying orientations toward ISTAG and ISTC in the communicational interactions with them.

Table 6.2 Filtering and channeling relevant knowledge into DG INFSOa

Learning filters ISTAGb

ISTCc

Learning area Long-term recommendations for program development

Concerns and preferences of governments

Limiting factors Dominance of discretionary content in flexible arrangement

Serves as expert group

Perceived as strategic advisor to DG INFSO

Dominance of prescribed limits in rule-guided behavior

Serves as monitoring body

Perceived as interface with the Member States and control board

Quality of

communicational

interactions

Organization of group activities and explicit linking with internal decision-making processes by the F6 unitd

Problem-oriented definition of goals and objectives for group activities with participation of DG INFSO

Discussion about implications of group's outputs for program management with participation of DG INFSO

DG INFSO effects on enactment of formalized procedures

Informal discussions complementary to formalized committee meetings

Emphasis on interest representation and accommodation

Communication with committee: dialogue, clarification, and bargaining

a Directorate General for Information Society. b Information Society Technologies Advisory Group. c Information Society Technologies Management Committee. d DG INFSO’s unit on “Work Program and Cross-program Themes.”

121

The empirical evidence that DG INFSO has achieved openness to know-ledge from external sources indicates that the administration focuses on learning areas deemed relevant to its activities in the implementation of the IST program. The active element in the boundary-spanning activities with ISTAG and ISTC also becomes apparent in the way the knowledge is channeled into the adminis-tration. The lack of predefined ways to incorporate external expertise as the IST program develops gives DG INFSO the necessary latitude for shaping them. And although the consideration of the concerns and preferences represented by the ISTC delegates is formalized by Comitology and therefore expected by the Member States, DG INFSO develops agency in the way it deals with those limits. DG INFSO’s active role in shaping the boundary-spanning activities with ISTAG and the enactment of the formalized procedures pertaining to the com-mittee’s monitoring role are therefore crucial in understanding how the know-ledge regarded as relevant to the management of the IST program enters the administration’s internal dynamics.

The involvement of external expertise from industry and academia is a traditional building block of the European research programs. In that sense the reliance on ISTAG for development of the IST program within the 5th frame-work program is not surprising. What is remarkable according to the inter-viewees is the adoption of the ISTAG vision of ambient intelligence as the lodestar of the IST program. DG INFSO’s attention to ISTAG reflects the administration’s intent on ensuring a coherent view of the program’s strategic development by external stakeholders who are recognized for their expertise and authority. DG INFSO defines the lines according to which ISTAG’s recom-mendations take shape and make their way into the administration when its directors and representatives of middle management participate in ISTAG’s plenary sessions. Both parties agree on the goals and objectives of the group’s activities and discuss the implications of ISTAG’s recommendations for the IST program’s development. The F6 unit (“Work Program and Cross-program Themes”), the body responsible for editing the various contributions to the annually revised and specified annual work programs of IST, helps align those recommendations with the learning needs of DG INFSO. F6 organizes the ISTAG meetings, makes sure that their output is relevant to DG INFSO’s internal decision-making processes, and feeds it into them, particularly when following ISTAG advice as it elaborates the work program under the auspices of the directors. Because the F6 unit is an advocate of ISTAG, it contributes strongly to DG INFSO’s acknowledgement of the group’s advice.

DG INFSO does not define the lines along which the concerns and preferences of the national delegates enter the administration; that is done by the Member States in the Council of the European Union. But DG INFSO does shape the circumstances under which it listens to the delegates and takes their views on the program management into account when implementing the

122

program according to the prescribed rules and procedures. DG INFSO gives credit to the committee’s monitoring role in the management of the IST program by listening to the delegates when their approval for the annual work program is required. The administration responds to their concerns, inquiries, and questions during the drafting of the work program, when it prepares records that contain these types of knowledge. These records indicate the stage at which the dele-gates’ concerns are taken into account in the revised work program. Drawing on these records may help reconcile the varying interests of the national govern-ments, but it does not eliminate the need for further communication.

DG INFSO relies strongly on informal discussions with the delegates to prepare the voting procedure in the formal committee meetings and to gain their support for funding decisions. The formal committee meetings help the dele-gates voice difficult issues and problems, but the officials of DG INFSO and the delegates do not use those encounters as a platform for the resolution of such issues. The meetings are the venue for the formal incorporation of the delegates' concerns that have already been articulated informally and taken into considera-tion. The administration enters into dialogue with the delegates beforehand to understand what concerns them, clarify what would prevent them from giving a positive vote, and bargain with them bi- or trilaterally on the program’s budget. The emphasis in the communicational interactions with ISTC is on interest representation and accommodation. The strong reliance on informal discussions in this regard creates informational asymmetries among the members of ISTC, helping DG INFSO secure agreement between the various positions and thereby take the lead in the operational program decisions.

Distribution

DG INFSO’s boundary-spanning activities with ISTAG and ISTC illustrate the significance of the distribution stage in processes of organizational learning. The evidence that different types of knowledge are acquired does not necessarily imply that they are widely disseminated in the administration. The two sets of boundary-spanning activities can be compared along lines suggested by theory (e.g., Huber 1991). The distribution of ISTAG’s recommendations is evidence of the occurrence and breadth of organizational learning, whereas the ISTC case illustrates that learning need not occur only through mechanisms of broadly based learning. DG INFSO draws on its boundary-spanning activities to create knowledge about the interests of the governments in the EU Member States, but it hampers the wide dissemination of the acquired knowledge. To assess the factors that facilitate and impede the distribution of the acquired knowledge, it is helpful to analyze the linkages of the administration’s boundary-spanning activities with intraorganizational processes. The linkages are important because

123

they affect how and how much the different groups in DG INFSO become aware of the acquired knowledge.

Both ISTAG and ISTC are involved in decisions that concern the imple-mentation of the IST program, but the decision-making processes differ in key respects when examined from an organizational learning point of view. ISTAG’s recommendations become part of DG INFSO’s common effort culmi-nating in the annual IST work programs. Because the ISTAG vision of ambient intelligence is used as a major filter by F6, it is disseminated widely among the groups that represent the different key actions in DG INFSO. The thrust of the communication with ISTAG experts is to process their recommendations internally. The distribution of the delegates’ views on the management of IST is not the issue when DG INFSO seeks approval of the annual work programs and when it responds to the delegates’ inquiries about the selection of research proposals for funding. The communication with the delegates is outward-oriented in the management decisions where ISTC is involved. Officials in DG INFSO inform themselves of the governments’ interests in the program in order to show that the work program supports similar initiatives in the Member States and that the funding decisions comply with the rational and legal basis of the program. DG INFSO thereby creates opportunities for the national delegates to assess the administration’s performance.

This study reveals that the functional differentiation of boundary-spanning activities produces a variety of mechanisms to disseminate the knowledge acquired through those activities. The factors contributing to the distribution of ISTAG’s recommendations show that the administration’s boundary-spanning activities with ISTAG have a processing function, whereas the external repre-sentation effected through the boundary-spanning activities with ISTC has built-in obstacles to the wide distribution of the governments’ concerns communi-cated by the delegates. The facilitators of and barriers to the distribution of the knowledge gained by means of boundary-spanning activities are inherent in the decision-making processes involving ISTAG and ISTC (see Table 6.3).

124

Table 6.3 Facilitators and barriers to the wide distribution of knowledge acquired through

boundary-spanning activities between DF INFSOa and two bodies of external stakeholders

Distribution

determinants ISTAG

b ISTC

c

Linkage with intra-

organizational

processes

Development of IST content as the annual work program is drafted

Gain of approval for final draft of annual work program and funding decisions

Broadly based

organizational

learning

Factors that facilitate wide distribution of ISTAG input:

• Drafting of the annual work program as a common effort, with the F6 unitd in charge

• Discussion about implications of ISTAG input for the drafting of the annual work program at lower management levels

• Top-down flow of information about ISTAG meetings from directors and middle manage-ment

Factors that impede wide distribution of ISTC input:

• Outward-oriented communication with delegates

• Perception of delegates as governmental representatives with little understanding of program content

a Directorate General for Information Society. b Information Society Technologies Advisory Group. c Informa-tion Society Technologies Management Committee. d DG INFSO’s unit on “Work Program and Cross-program Themes.”

The major driving force behind the wide distribution of the ISTAG vision of ambient intelligence is the drafting of the annual work program as a common effort by the F6 unit. In the process of contributing to the content development of the IST program, the administration’s project officers and their heads of unit, along with the directors of the directorates and their immediate staffs, are involved with F6. The unit uses the ISTAG ideas to pull the various contribu-tions together when it produces a first draft of the work program. It is therefore almost impossible to overlook the strategic views of ISTAG. A second facili-tating factor is that the drafting of the work program is accompanied by internal discussions about the implications that the ISTAG vision of ambient intelligence has for the content of the IST program. The internal discussions facilitate the dissemination of ISTAG’s views when the various units in the research directorates of DG INFSO prepare the inputs that feed into the annually updated and revised work program. Lastly, lower management levels become aware of ISTAG’s views as a result of whatever information is passed on to them by their leaders who participate in the meetings of the group.

125

DG INFSO has an elaborate system of rules and procedures for devel-oping contributions to the work program across the directorates in similar ways. Based on the consultations with the constituency, the development of contribu-tions proceeds bottom-up with guidance from top-down. The organization of this process along the hierarchies in the directorates of DG INFSO dealing with research encourages the wide dissemination of ISTAG’s input, with varied inter-pretations given to it. The officials in the units of the directorates ensure the acknowledgement of their contributions to the editing of the work program by communicating their understandings of the ISTAG views to the research area in which they are specialized. The attendant discussions between project officers and heads of unit and between heads of unit and directors help to develop specific understandings of the ISTAG vision of ambient intelligence and the resulting recommendations. There is little evidence, though, that the implica-tions of ISTAG’s views for the program are mutually understood across the structural and cultural barriers of DG INFSO. The procedures necessary for such cross-fertilization are lacking.

Organizational learning triggered by the boundary-spanning activities between DG INFSO and ISTC is hampered at the stage of DG INFSO’s internal distribution of the policy-oriented knowledge represented by the national dele-gates constituting ISTC. The knowledge that DG INFSO acquires from ISTC in the latter’s monitoring role tends to remain in the minds of those who talk directly with the delegates about the work program and the funding decisions. There is no evidence of dedicated efforts, procedures, and rules in DG INFSO that facilitate the wide dissemination of the delegates’ concerns and preferences to those officials who do not directly talk with them. The outward-oriented communication between DG INFSO and ISTC is a main barrier to the distribu-tion of ISTC input. The administration acquires knowledge from the delegates—if necessary—so as to provide them with information specifically adapted to them, including the records that demonstrate acknowledgement of their govern-ments’ interests in the definition of the annual work programs and that clarify matters that bear on funding decisions. DG INFSO provides information about the program management to the delegates in order to legitimate the implementation powers conferred upon it by the Member States.

The perception of the delegates’ role in the management of the IST pro-gram is another factor preventing ISTC input from being distributed across the administration’s units and groups. The delegates are seen as representatives of their governments with little understanding of the content of the IST program. The view that dominates communication with the IST committee is that the delegates engage primarily in the representation of their governments’ individ-ual interests, whereas DG INFSO is in charge of adapting to this set of fragmented interests.

126

Interpretation

The essence of the interpretation phase in processes of organizational learning is to give meaning to information and achieve mutual understandings across an organization’s groups and units. Information acquired through boundary-span-ning activities needs to be contextualized if it is to become knowledge relevant for an organization and comprehensible to its groups. Knowledge acquired from external sources becomes contextualized if it feeds into the strategic and opera-tional decision-making of an organization and becomes related to the routines and practices that guide these actions.

The theoretical framework of this study suggests that the contextualization of knowledge acquired through boundary-spanning activities is intertwined with organizational politics. The empirical evidence supports this assumption. It illustrates how the different dimensions of power—including the resource, process, and meaning dimensions—facilitate and hinder the sense-making of the knowledge that accrues to DG INFSO through its boundary-spanning activities among the groups in the administration. The two cases of boundary-spanning activities demonstrate in different ways that the initial definition of meaning by those who are directly involved becomes relative in the process of interpretation. Table 6.4 presents the interpretational differences in terms of (a) the decision-making processes to which the acquired knowledge is related and (b) the impact of organizational politics on the achievement of mutual understandings of the knowledge in question.

127

Table 6.4 Interpretation of knowledge acquired through boundary-spanning activities in

DG INFSOa

Interpretation

determinants

ISTAGb ISTC

c

Relationship

between acquired

knowledge and

decision-making

processes

ISTAG views are explicitly incorporated as guidance for drafting the annual work program.

ISTAG views are used to define, label, and describe what is being done in the IST program.

Predefined, represented concerns and preferences are incorporated to ensure program management task of DG INFSO.

Strategic planning and operational tasks are buffered from delegates’ views.

Impact of organiza-

tional politics on

sense-making of

acquired knowledge

ISTAG views are used as a resource for editing of work program by F6.d

Decision-making process about strategic planning of the IST program is opened to ISTAG experts.

Reference to ISTAG views means desirable action

Meaning of ISTC in DG INFSO inhibits sense-making of delegates’ views.

Technocratic decision-making process inhibits sharing of delegates’ inquiries about program management.

a Directorate General for Information Society. b Information Society Technologies Advisory Group. c Information Society Technologies Management Committee. d DG INFSO’s F6 unit, “Work Program and Cross-program Themes.”

F6 uses the ISTAG advice as a resource to strengthen its role in this delicate balancing act between different foci and competing interests. F6 relates the activities of ISTAG to what is perceived as a learning need in DG INFSO: the development of goals and objectives for the IST program as a whole. The unit is crucial in ensuring the acknowledgement of the advisory group’s recom-mendations within DG INFSO. The understanding that ISTAG is the embodiment of technical expertise and authority helps F6 establish the vision of ambient intelligence as the frame of reference for drafting the annual work program.

The direct interaction between the ISTAG experts and the directors of DG INFSO is an additional factor facilitating the establishment of the group’s vision of ambient intelligence as a major source of guidance for developing the program’s content. Both parties are brought together in the plenary sessions that the F6 unit organizes for ISTAG. The brainstorm-like atmosphere of these

128

meetings contributes to the rise of coalitions between the directors of DG INFSO and the ISTAG experts, with the development of strategic ideas for IST as the common denominator. Although the directors are formally in charge of strategic planning, in practice they lack the ability to define common ground for the program. The visionary input of ISTAG helps them initiate the dis-cussion and go beyond their particular positions on the future of IST. The involvement of ISTAG in the decision-making process about the strategic planning of IST therefore contributes to the awareness in DG INFSO that the vision of ambient intelligence is a helpful framework in which to bring the different foci and interests of the administration’s groups together.

It is desirable in DG INFSO to refer to the vision of ambient intelligence as the annual work program is drafted. When contributing to that program, proj-ect officers construct relationships between the vision and their perceptions of the developments and trends in the segments in which they specialize. The administration’s project officers do so mainly for pragmatic reasons, namely, to shape the annual work program according to what they perceive as the needs of the constituencies with which they work. A closer look at the factors that affect sense-making at lower management levels reveals that reference to the vision of ambient intelligence is primarily a communication issue. Its adoption as the program vision has made it the espoused way of doing things and has little impact on perception in DG INFSO. Those who must implement the vision of ambient intelligence tend to view the developments in the field through different lenses. Sense-making by project officers is affected by their specialized fields, the experience with these fields and their actors, and the direct interaction with colleagues and peers in the immediate work environment. The existence of characteristic working methods in these territories engrains the administration’s internal fragmentation. As a result, different understandings of the vision of ambient intelligence and its implications for the IST program coexist in DG INFSO with little likelihood of cross-fertilization.

The interpretation of the knowledge gained from the boundary-spanning activities with ISTC emphasizes the importance of power as a dimension of meaning. The monitoring role of ISTC in the management of the IST program means interest accommodation and clarification in DG INFSO. The adminis-tration draws on these understandings when it communicates with ISTC in a way that enables the delegates to articulate their views on the program’s management in the prescribed manner yet does not obligate DG INFSO to incorporate them. The decision-making on the program’s goals and objectives are buffered from the ISTC input because DG INFSO relies on informal encoun-ters with the individual delegates as it prepares the vote on approval of the program. Moreover, the operations of the program are buffered because DG INFSO implements the program within the prescribed limits of Comitology.

129

The buffering enables DG INFSO to create its own space for making decisions on matters that ISTC oversees.

The meanings ascribed to ISTC inhibit the distribution of the delegates’ views on the management of the program. The empirical evidence demonstrates that the mechanisms affecting the dissemination of the knowledge acquired through the boundary-spanning activities with the committee are closely related to those informing that knowledge’s interpretation in DG INFSO. The strong reliance on an informal atmosphere for reaching agreement with the delegates on the annual work programs drastically reduces the likelihood that awareness of their concerns and preferences will go beyond the senior and middle manage-ment officials directly involved in those encounters. One of the few instances in which officials from lower management levels may talk directly to the delegates occurs once the work program is approved and the evaluation and selection procedure for a given call for proposals is underway. But direct interaction does not facilitate the sharing of acquired knowledge in DG INSFO. The inquiries of the delegates are not viewed as a resource that needs to be shared and made sense of within the administration. The dialogue with the delegates focuses on the technocratic decision-making process, an emphasis that obstructs the dissemination of their inquiries within the administration. The communication with ISTC is about explaining and ensuring the transparency of the application of the rules and procedures that guide the evaluation of incoming proposals, whereas the consultation of technical experts (like ISTAG) has to do with the ranking of the proposals on the funding list.

Organizational Memory

The theoretical framework of this study suggests that organizations draw on their memory when they learn. Organizational memory consists of the know-ledge that is embedded in the activity system of organizations, including their structures, rules, procedures, and routines. Knowledge gains are stored for future use if they are related to organizations’ knowledge about how to do things. Without reference to the concept of organizational memory, information cannot be identified as relevant knowledge, and the mechanisms that contribute to its distribution and sense-making in organizations cannot be defined. As the reposi-tory of organized knowledge, the memory of organizations helps them find and use knowledge that fits their requirements. Because organizational memory acts as a template for interpretation and action and because it is taken for granted, it tends to stabilize learning processes. The stability of an organization’s activity system is at stake when knowledge is acquired from external sources. Externally acquired knowledge is not necessarily compatible with the established beliefs

130

and practices in organizations and may greatly challenge their absorptive capacity.

The empirical evidence highlights the stabilizing elements in DG INFSO that affect learning triggered by its boundary-spanning activities with ISTAG and ISTC. The incorporation of the ISTAG advice, which is seen as relevant knowledge by key members of the administration, is a serious internal challenge to the established beliefs and practices. The analysis of the boundary-spanning activities between DG INFSO and ISTC points in the opposite direction. The communication with the national delegates centers on the stability of the way in which DG INFSO carries out its operational tasks. The administration limits its openness to ISTC to the learning areas that are compatible with this require-ment.

The adoption of the ISTAG vision of ambient intelligence as groundwork for drafting the annual work programs illustrates that overall goals and object-ives for the implementation of the IST program did not originally exist in DG INFSO. The administration incorporates the distinctive program cultures of ESPRIT, ACTS, and TAP, the predecessors of the IST program. In light of these different approaches to Information Society, the orchestration of collaborative action in DG INFSO is more difficult than initially expected. Particularly at the beginning of the 5th framework program in the late 1990s, the administration drew strongly on the perceptions and modus operandi that had been established during the 4th framework program.

The adoption of the ISTAG vision as the lodestar of the IST program calls the inherited memory of DG INFSO into question. The administration is challenged to use the vision of ambient intelligence as an incentive to develop overall goals and objectives for the IST program. Such a process of strategy-building is limited to those who attend the plenary sessions with ISTAG (senior management, the members of the F6 unit, and the ISTAG experts). Little note is taken of the strategic ideas and understandings of the IST field that exist at lower management levels. Moreover, there are few opportunities that allow for persuasion, negotiation, and advocacy regarding the implications of the ISTAG vision for the program’s development across the administration’s internal boun-daries. The top-down endorsement of the vision of ambient intelligence affects how a large proportion of the officials in DG INFSO describe their activities, but it does not fundamentally change the stated goals and interests they are trying to pursue with the program.

Challenges to the inherited memory of DG INFSO come also from a different angle. Beyond the drafting of the annual work program, DG INFSO also turns to a new instrument as a way to incorporate ISTAG recommendations: “Cross-program actions.” These actions were introduced in the 5th framework program to foster multidisciplinary research and are to be managed by the units whose specialized fields are addressed. The F6 unit taps ISTAG for ideas to

131

propose cross-program themes to the directors. The implementation of cross-program actions along the lines that are suggested by ISTAG challenges these actors to develop openness to the approaches of other units and to contribute to understanding and cross-fertilization between units.

The stabilizing impact of the administration’s structural arrangement stifles the implementation of cross-program actions. Because these kinds of activities bring together multiple fields of research, they need to be managed differently than the bulk of the projects in IST, which address one specialized field. But the management of cross-program actions has turned out to be more difficult than initially anticipated. Strong departmental boundaries between units and directorates tend to stabilize established beliefs and practices, enabling heads of unit to preserve the way that their respective units monitor and evaluate research projects. Project officers in cross-program actions find it difficult to go beyond their unit’s point of view.

The boundary-spanning activities between DG INFSO and ISTC indicate that the legal basis of the administration’s implementation powers shapes its memory. The rules and procedures of Comitology, the 5th framework program, and the IST program are formalized knowledge underlying the administration’s agency in the management of the IST program. DG INFSO’s knowledge-seek-ing behavior toward the national delegates reflects the monitoring role of ISTC. It is limited to those areas where ISTC may obstruct the implementation powers that the Member States confer upon the administration. DG INFSO listens to the delegates to gain legitimacy among the Member States for its implementation of the program. It identifies and considers the views that the national delegates voice during their informal encounters with DG INFSO officials in the stage of seeking approval for the annual work programs and explaining the legal rationale of the funding decisions. DG INFSO thereby buffers its internal dynamics from the concerns and preferences of the delegates and preserves the stability of its operational tasks.

Concluding Comments

With this comparative analysis of the impacts that boundary-spanning activities have on organizational learning, what are the implications for the plausibility of the analytical model that integrates these notions? First, the nature of the com-municational interactions with external stakeholders is indeed significant for processes of organizational learning. The functional differentiation of boundary-spanning activities is useful for assessing qualitative differences in the organization-environment nexus and their influence on learning processes. A core finding of the comparative analysis is that the information-processing and external representation functions of boundary-spanning activities correspond

132

with identifiable qualities of learning processes. The achievement of openness to knowledge from external sources is different in the two sets of boundary-spanning activities. These differences impinge on the processes whereby the acquired knowledge is handled and used inside DG INFSO. Knowledge acquired for use within DG INFSO goes through mechanisms that facilitate its wide dissemination with varied interpretations given to it. Conversely, know-ledge acquired to gain legitimacy from the national governments is kept from being widely distributed and made sense of within DG INFSO. Table 6.5 shows how the inward flow of knowledge in the boundary-spanning activities with ISTAG and the outward flow of information in the boundary-spanning activities with ISTC affect the scope of learning and points to different contexts for learning. The organization makes explicit and implicit choices in its efforts to cope with changes and developments in its environment. These choices are different in the two sets of boundary-spanning activities that are studied.

Table 6.5 Embedded learning from boundary-spanning activities with ISTAG and ISTC

Embedded learning DG INFSOa–ISTAG

b DG INFSO–ISTC

c

Scope of

organizational

learning

Emphasis on internal processing of knowledge through the development of a program vision, a focus that fosters convergence of ICT and softens internal fragmentation in DG INFSO

Emphasis on external representation, a focus that contributes to undisturbed program management by pleasing Member State delegates

Context for learning Filtering between relevant and less relevant inputs for drafting the annual work program; participation in decision-making about program strategy; communication and negotiation between different actors drafting the annual work program

Identification of concerns and preferences of Member State governments; bureaucratization of program management; influence on external assessment

a Directorate General for Information Society. b Information Society Technologies Advisory Group. c Information Society Technologies Management Committee.

The evidence presented in this study substantiates the argument that the choices organizations make about learning from boundary-spanning activities

133

are situated. But the empirical evidence also indicates that explaining the rise of learning in terms of factors in the social context of organizations does not fit well with the idea that learning takes place in stages. The contextual nature of organizational learning implies that a portrayal of organizational learning as separate, sequentially evolving stages cannot be more than an analytical tool. The empirical reality of learning processes is far more complex and requires additional conceptual tools to make sense of it. The present case of learning in DG INFSO indicates that similar factors in the administration’s social context differentially affect the different processes of learning at the same time. Looking at learning in terms of its social constitution means that the differentiation between processes of learning captures only some aspects of empirical reality and neglects others whose consequence cannot be assessed if one relies only on a stage-process model of learning. This study suggests that examination of the overlaps between learning processes in formal organizations would advance research on the subject as much as would analysis of the properties that distinguish them.

Distinctive Qualities of Organizational Learning

Triggered by Boundary-spanning Activities

The boundary-spanning activities between DG INFSO and ISTAG triggered learning processes in the administration that strengthen its ability to act strategically in the implementation of the IST program during the 5th framework program. The involvement of ISTAG was used by the administration’s groups to increase the internal pressure to develop an encompassing program vision with overall goals for IST. The social constitution of the boundary-spanning activities, including the objectives, underlying orientations, and regulatory aspects, were conducive to this learning need. ISTAG’s status as a strategic advisor and the flexible arrangement of the group’s involvement in managing the IST program helped DG INFSO limit its knowledge-seeking behavior to the ideas of the ISTAG experts about long-term developments in the IST field and helped it to draw on their authority.

The difficulties of orchestrating collaborative action continue to give the administration a stake in relating ISTAG’s advice to established beliefs and practices internally. These difficulties surface particularly when the annual IST work program is drafted. The integration of ISTAG’s advice in the attendant decision-making process, which encompasses a great many groups in the directorates of DG INFSO, facilitates its dissemination and interpretation. The F6 unit contributes greatly to DG INFSO’s adoption of ISTAG’s advice. It links DG INFSO’s internal views about the program’s development with ISTAG’s recommendations. Drawing on ISTAG’s advice under the auspices of the

134

administration’s directors helps F6 gain authority in the act of delicately balan-cing between different foci and approaches. The effort that DG INFSO makes to understand ISTAG’s strategic views illustrates that the administration cannot be treated as a unitary actor. ISTAG’s views are related to DG INFSO’s various established beliefs about the program’s goals. A strategic message communi-cated to the program’s constituencies has limited effects on the established ways of monitoring and evaluating the implementation of the IST program.

The quality of learning is distinctive in the second set of boundary-spanning activities—those between DG INFSO and ISTC. They trigger the administration’s acquisition of knowledge that is used to respond to the concerns and preferences of the governments in the Member States. The interaction that the administration has with the delegates reflects its effort to act within the prescribed limits stipulated in the legal framework governing the IST program’s implementation. Failing to comply would risk ISTC obstruction of DG INFSO’s operational tasks. DG INFSO gains legitimacy for the implementation powers conferred upon it when the national delegates approve the annual IST work programs and the decision-making that culminates in the selection of research proposals for funding. Speaking with the ISTC national delegates to identify the concerns and preferences of the EU Member States is a means to this end. But the existence of prescribed learning needs is not to be equated with the rise of learning needs in DG INFSO. The reluctance of the administration to have all its groups and units share in the distribution and interpretation of the knowledge gained through boundary-spanning activities with ISTC indicates a lack of a willingness to learn that is significant for the other steps in the learning process.

The understanding of the decisions to which ISTC contributes is a major barrier to the distribution and interpretation of the concerns and preferences re-presented by the delegates. ISTC’s involvement in the management of the IST program means interest accommodation to many officials in DG INFSO. It is seen as a crucial aspect of the program’s implementation and is dealt with by officials from higher management levels. DG INFSO’s dominant perception of ISTC as an interface with the national governments and as a control board with little understanding of technological developments and the program’s constitu-encies makes the committee relevant only for the formal obligations of moni-toring the implementation of the IST program. The acknowledgement of ISTC’s input is therefore deemed necessary for achieving political agreement but not for running the program. Sharing is not an issue in the forging of political agree-ment among the delegates of ISTC. Accordingly, DG INFSO relies on informal discussions with the individual delegates from the Member States to inform itself of their views, negotiate with them so that decisions reflect their govern-ments’ interests, and show that the program’s implementation complies with the rules of Comitology. The informal nature of the dialogue with the delegates implies that the acquired knowledge tends to remain in the minds of the few

135

officials who interact directly with the delegates. DG INFSO has no incentives or procedures for wide dissemination of the concerns and preferences expressed by the Member State representatives.

Overlapping Stages in Processes of

Organizational Learning

Basing the analysis of organizational learning on a process model that differenti-ates between the rise of learning needs; the acquisition, distribution, and interpretation of knowledge; and organizational memory may suggest that learning occurs in a neat progression of stages that evolve sequentially. But organizational learning processes appear muddled when the organization’s social factors that sustain them are taken into account. The empirical evidence shows that similar kinds of factors—regulatory aspects and perceptual filters in particular—affect the stages of organizational learning simultaneously. They affect the acquisition, distribution, and interpretation of the knowledge to which DG INFSO gains access through its boundary-spanning activities. Regulatory aspects and perceptual filters influence different stages of organizational learn-ing in DG INFSO simultaneously because they establish the administration’s organizational memory, which serves as a template for action and interpretation. It is a resource on which the administration draws when it engages in the different working relationships with ISTAG and ISTC.

ISTAG was set up by DG INFSO to have external experts help develop the IST program. The flexible arrangement of the boundary-spanning activities with ISTAG enables DG INFSO to tap the ISTAG experts for strategic ideas that fit its learning need to strengthen the strategic scope of the IST program. The elaborate system of rules and procedures that drives the drafting of the work program as a common effort is important for the wide dissemination of the acquired knowledge in the administration’s internal context. The involvement of the ISTAG experts in the decision-making about the content and direction of the IST program by the directors of DG INFSO fuels the interpretation of the ISTAG views as a key contribution in the overall process.

Because ISTC is the regulatory (and regulated) body that monitors DG INFSO’s management of the IST program, it can block implementation of the program if the administration fails to consider the interests of the EU’s national governments and to show compliance with rules and procedures governing the program’s implementation. The administration uses the mandated involvement of the ISTC delegates of the Member States to identify develop-ments in national government interests that may be relevant in convincing the ISTC delegates to approve the annual IST work program. But the informal nature of the dialogue is a barrier to wide sharing of the acquired knowledge.

136

The legal rationale of the program’s implementation restricts interpretation of the delegates’ views on the program, whether they stem from the governments of the EU Member States or from experience with national initiatives similar to the IST program.

Perceptual filters complement the impact of regulatory aspects on DG INFSO’s learning processes. The administration sees ISTAG as a strategic advisor on its development of the IST program. Linking the advisory group’s recommendations to that work is considered desirable action as the work pro-gram is drafted. The view of ISTC as an interface with the Member States and as a control board is coupled with the perception that its members act as repre-sentatives of their governments and have little understanding of the program’s content. The administration sees the adaptation to the different interests they represent and the clarification of their inquiries about the operations of the program as the major issues in the working relationship with ISTC.

Although the regulatory aspects and perceptual filters overlap, comple-ment each other, and combine during the stages of organizational learning, it is still analytically instructive to differentiate the rise of learning needs; the acquisition, distribution, and interpretation of knowledge; and organizational memory. The salience of the individual issues differs from one process to the next. They relate to the challenges to the status quo in organizations’ structures and power relations that are inherent in organizational learning. Boundary-spanning activities trigger learning processes differently, depending on whether the willingness to learn arises within an organization or the learning needs are those of external societal forces. In learning processes triggered by boundary-spanning activities (see Figure 6.1), openness to external stakeholders is central during the knowledge-acquisition stage. It revolves around the question of what counts as relevant knowledge about the environment. A crucial issue during the distribution stage is the nature of the linkage between the boundary-spanning activities and the internal dynamics of the focal organization, for it affects the occurrence and breadth of organizational learning. In the interpretation phase it is critical to acknowledge that the initially established meaning of the relevance of the acquired knowledge shifts because giving sense to acquired knowledge is not necessarily the same as making sense of it inside organizations. The final issue is the stability of the established beliefs and practices in the organizational memory stage of organizational learning.

137

Figure 6.1 Overlaps and distinctions in processes of organizational learning.

Boundary-spanning Activities

Willingness to Learn “Internal/External”

Knowledge Acquisition

"Openness”

Distribution

“Linkages”

Interpretation

“Sense-making”

Organizational Memory

“Stability”

138

6. Conclusions

The Study in Retrospect

This study traces processes of organizational learning in one Directorate General of the European Commission. The main interest is in learning about the Commission as a corporate actor through analysis of learning processes that are facilitated and constrained by its linkages with external stakeholders in the national governments of the Member States, communities of experts, and business. The study challenges the neofunctionalist assumption that a certain degree of autonomy and independence can be attributed to supranational institutions, such as the Commission or the European Court of Justice, because of their problem-solving capacity in a world that is increasingly beset by policy-problems that transcend national borders (Haas 1964, 1990; see Grande 1997; Messner 1998; Stone Sweet & Sandholtz 1998). The Commission is viewed in this study as a supranational organization with few resources. The assumption is that its capacity for agency depends on involving external stakeholders in the definition and implementation of policy initiatives in the European Union through the many advisory, preparatory, and regulatory committees populating the Commission’s external boundaries. The notion of boundary-spanning activities is introduced to capture the communicational interactions between Commission officials and external stakeholders as a means of ascertaining demands, expectations, and expertise in its environment. Organizational learning theories provide a framework for understanding how the externally acquired information is transformed into an organizational property beneficial to the focal organization. Organizational learning from boundary-spanning activities is defined as a social accomplishment that rests on the distribution and interpreta-tion of externally acquired knowledge and its storage in organizational memory.

The development of a theoretical model to study the impact of boundary-spanning activities on learning in DG INFSO, which serves as a case of a supra-national organization, is accompanied by the development of a research methodology. The paucity of empirical research in the organizational learning discourse necessitates a research design that allows iterative movements and feedback loops between the basic ideas from theory and the images that are formed from evidence. A case-study approach is chosen to progressively refine the study’s theoretical underpinnings in response to the findings from four data collection phases. Despite the lack of empirical research about learning pro-cesses, the analytical value of the theoretical notions is considerable. Boundary-

139

spanning activities are indeed traceable in empirical reality, and their impact on the individual stages of organizational learning differs depending on their social constitution. The stress on the idea that learning processes are embedded in the social context that sustains them is a helpful heuristic for revealing the situa-tional factors that matter for learning from boundary-spanning activities. The systematic comparison between the two different sets of boundary-spanning activities in this study emphasizes that both the facilitation of learning processes and the creation of barriers to learning are social accomplishments that affect the behavior of the focal organization.

Processes of Learning in the European Commission

This study shows how a Directorate General of the European Commission creates its own decision-making latitude vis-à-vis the Member States through the accumulation, use, and storage of expert technical advice that shapes its behav-ior in the institutional environment framing decisions on the implementation and development of a Community policy. The major findings of the study are evidence of the institutionalist treatment of EU decision-making, according to which “time constraints, scarcities of information, and the need to delegate decisions to experts may promote unanticipated consequences and lead to considerable gaps in member-state control” (Pierson 1996, p. 137). The Com-mission appears as the engine of the European integration in this study, playing a significant part in developing the internal market. Placing the Commission’s learning behavior at the center is a key to understanding the way in which it puts European interests first and sets the policy-agenda with little interference from national governments.

Given the European Union’s neoliberal policy agenda, its supranational institu-tions focus primarily on establishing coordinational, pan-European market-creating or market-compatible solutions to policy problems (Héritier 1997). This primacy of the market has dominated the European polity since the inception of the EU/EC (Lequesne 2000). With the integration of the national markets for labor, goods, services, and capital and with the removal of tariffs and quotas, differences in national regulatory standards (for the environment, health and safety, technical compatibility, and other areas) have been increasingly per-ceived as obstacles to transnational economic exchange (Stone Sweet & Sandholtz 1998). As the Commission succeeds in building linkages between a widening range of policy issues and the principle of the internal market, oppor-tunities for increased agency at the supranational level are created (Schaper-Rinkel 2003).

140

The Learning Behavior of the

European Commission

To understand what the present study can say about the learning behavior of the European Commission as a whole, the underlying cognitive and social processes triggered by boundary-spanning activities are best viewed as a reflection of that body’s capacity for agency (Wiesenthal 1995). Boundary-spanning activities are a means of establishing an organization as an actor in its relationship with exter-nal stakeholders because they affect how boundaries are drawn (Hernes 2003, Jevnaker 2003). Linking external developments with internal activities, people, processes, resources, and intentions, boundary-drawing is intrinsic to the very process of organizing (Hernes & Paulsen 2003). The two different cases of boundary-spanning activities examined in this study in order to understand their implications for learning processes in the focal organization highlight the Commission’s capacity to act as a “supranational technocracy” (Bach 1999, p. 55) and as a “politicized bureaucracy” (Christiansen 1997, p. 76). These two classifications help identify two types of learning behavior characterizing the Commission. Each type corresponds to a specific kind of knowledge absorbed by the Commission and indicates the purposes for which the Commission engages in learning.

First, the accumulation and use of the knowledge to which the Com-mission gains access through its boundary-spanning activities with external experts indicates its organizational agency as a supranational technocracy. The Commission’s power among the institutions of the European Union depends on resources that help solve problems: “expert knowledge, political insight, and bargaining experience” (Kohler-Koch 1997, p. 48). Fischer (1990) defines technocracy as a “system of governance in which technically trained experts rule by virtue of their specialized knowledge and position in dominant political and economic institutions” (p. 17). Technical experts rather than politicians shape the deliberative framework within which policy decisions are taken. Regarding the European Commission as a technocracy calls attention to its tendency to accumulate policy-related and administrative knowledge and to rely on technical methodologies and related arguments in its work (Malek & Hilkermeier 2001).

The EU administration relies on dense networks of experts for “objective” and technical advice (Peterson & Sharp 1998, p. 186). Compared to other public administrations, the Commission has a relatively high-caliber staff; a large share of Commission officials is scientifically trained and knowledgeable in legal and administrative matters (Bach 1999, Cram 1994, Lequesne 2000, Peterson & Sharp 1998). The pairing of competent specialists with bureaucrats who trust them is encouraged because of similarities in values and styles (March & Olsen 1989, p. 32). The Commission recruits competent specialists from the private sector, academia, and civil society rather than from Member State governments.

141

Delegates in Comitology committees are not among the Commission’s major sources of expertise on policy matters. (For a contrasting opinion, however, see Falke 2000.) The Commission’s preference for absorbing external expertise from nongovernmental actors, combined with the formal right to aggregate policy demands and put them on the agenda of the European Union, facilitates its entrepreneurship in the European polity. Its capacity to act as a think tank and its power of initiative are intimately linked in the Commission’s pursuit of collective solutions at the European level (Laffan 1997, Metcalfe 1996).

The second type of learning behavior characteristic of the Commission relates to that body’s origins as an institution created by national governments to facilitate their interests and bargaining (Lequesne 2000). By definition, supra-national organizations such as the Commission act within the constraints of Member State preferences (Pollack 1998). The boundary-spanning activities between the Commission and the delegates from the national governments in the Member States reflect its agency as politicized bureaucracy. Christiansen (1997) defines this term as a combination of (a) the Commission’s aspirations to rule-orientation and rule-coherence, hierarchical organization, and functional spe-cialization and (b) the politicization of the major fields of Commission activity (i.e., proposing legislation, implementing policy, and enforcing compliance). Working in a politicized bureaucracy challenges Commission officials to come to grips with “the charged politicking in an apparent bureaucratic organization, the paper bureaucracy which all condone but few take seriously, and the paradox of widespread informal networking in a formalistic bureaucracy” (Hooghe 2005, p. 878).

The boundary-spanning activities with national delegates indicate the Commission’s identity as politicized bureaucracy because those activities are a vehicle for the absorption of and alignment with governmental interests and preferences in the aforementioned major activities of the Commission. The formulation and implementation of policies is the Commission’s core com-petence, yet it is also simultaneously subject to extensive deliberation and negotiation. The learning behavior of supranational organizations as politicized bureaucracies is characterized by the antagonism between (a) their duty to develop and apply common rules and (b) the political forces in the environment that take exceptional interest in their internal proceedings. The tension that arises from the duality of providing public administration and having to respond to the specific requests from national governments may be handled through buffering (Thompson 1967). Knowledge that relates to national and political considerations will be of limited relevance to the Commission’s internal staff if its activities are more bureaucratic than politicized, that is, if they predominantly involve high specialization, professional backgrounds, and the strict execution of legal rules.

A degree of caution is due at this point, for the findings presented here cannot be more than tentative. Research on the learning behavior of supra-

142

national or international organizations is sparse. Moreover, limits to an overall characterization of “the Commission” make it difficult to extrapolate from the present study to other cases of supranational organizations. The Commission is distinctive “in terms of its sectoralization and fragmentation, its multinational staffing and cultural heterogeneity, as well as in terms of the absence of uniform procedures and practices” (Malek & Hilkermeier 2001, p. 2). With an adminis-trative body composed of 23 Directorates General, a wide range of services and task forces, and an executive college of 25 commissioners, it is a complex organization. The various Directorates General have very different working practices and have quite individual relationships with the important actors of the separate policy sectors they address (Cram 1994). The diversity of adminis-trative cultures across the Directorates General of the Commission stems from their historical backgrounds and the substance of the policies that are adminis-tered (Cini 1997). Given the enormous differences between policy sectors, such as those between the Common Agricultural Policy and the Common Foreign and Security Policy, commonalities are difficult to discern (Peterson & Bomberg 1999). Much additional research is needed on the absorptive capacity of supranational organizations and, hence, on their internal dynamics in order to continue building solid knowledge about the learning processes in this type of organization.

Policy Sector Matters for Learning from

Boundary-spanning Activities

The distinction between the learning behavior of the European Commission as a supranational technocracy and its learning behavior as politicized bureaucracy is a main characteristic of its agency, namely, to engage continuously in the balancing act between autonomy and dependence (Lequesne 2000). The Commission strives for deeper integration at the transnational level and is con-strained by its own intergovernmental origins. The Commission increases its autonomy if it is able to build up alliances with transnational constituencies such as subnational organizations, interest groups, or individuals within Member States. Conversely, the institutional decision-making rules that govern the delegation of powers to the Commission are a key constraint on its autonomy because they imply that Member State interests have to be safeguarded (Pollack 1998; see also Feld & Jordan 1994, Senarclens 2000). Both types of learning are always constitutive of organizational learning in the Commission, but depending on the policy sector in question, the technocratic type of learning is more prevalent than the politicized learning type.

Boundary-spanning activities trigger the absorption of technical expertise if the supranational organization can choose the wording as the drafter of texts.

143

With the authority to phrase the content and direction of a given policy, the organization can specify policy details and policy options to be considered (Peterson & Bomberg 1999, p. 21). The Commission, working groups of the Council, and committees of the European Parliament are the major actors in policy-shaping decisions. But the Commission, by virtue of its right to propose, is pivotal in this phase of the EU policy process. In its efforts to forge a consensus between a wide range of decision-makers and in response to broad policy injunctions, the Commission consults formally and informally with governmental representatives, interested groups, and lobbyists to facilitate exchanges of ideas and agendas (Peterson & Bomberg 1999, p. 22). The bureaucratic element in the Commission’s activity enables the administration to present itself as impersonal, technocratic, and neutral, allowing it to exert control over information and expertise (Barnett & Finnemore 1999, p. 708). In short, that characteristic of the Commission gives it autonomy from Member State control over certain policy-shaping decisions.

Trying to make progress in politically sensitive areas, the Commission frequently resorts to the “master discourse”: competitiveness in the EU (Radaelli 2003, p. 21). It gives meaning and direction to European policies, making them conducive to and compatible with the internal market and the Economic and Monetary Union (EMU), including environmental, technology, and social policy. The Commission draws extensively on external expertise in these areas (Fischer 2003). In the European technology policies, for instance, the Com-mission was able to capitalize on the fluidity and inherent instability of informa-tion and communication technologies by building alliances with important actors from industry (Cram 1994, Schaper-Rinkel 2003). Its round tables with the major players in information technology during the early 1980s resulted in the rise of the ESPRIT program, which became a major building block of the Community’s RTD policies. Environmental policies are an additional example of technocratic learning in the Commission. Because the issues in question are considered highly complex and technical, external experts are consulted to provide solutions for such matters as reducing industrial emissions into the air. Although experts act only in an advisory capacity, the Commission grants them many opportunities to shape the content of environmental policy (Brown 2000, Héritier 1997). Conversely, the rise of a social policy in an integrated Europe suffers from consistent opposition by employers, trade unions, and industry to any expansion of Union competence in this domain and from noncompliance with commonly adopted standards (Cram 1994, Hartlapp 2005). More recently, though, the somewhat neglected idea of a Social Europe has been gaining momentum on the European policy agenda, spurring the Commission’s Directorate General for Employment, Social Affairs, and Equal Opportunities (DG V) to make intelligent use of its committees with external stakeholders.

144

Theoretical Implications of the Impact of Boundary-spanning

Activities on Organizational Learning

Because most of the research on organizational learning has centered on firms in western societies, there is a call to increase the variety of the types of organi-zations represented in future work (Berthoin Antal 1998, LaPalombara 2001, Ventriss & Luke 1988). The common assumption is that the validity and robust-ness of theories and concepts in the organizational learning discourse would benefit from such a move “out of the comfort zone” (Berthoin Antal, Dierkes, Child & Nonaka 2001b, p. 933). Research concerned with learning processes in supranational organizations is clearly a move out of the comfort zone because it looks at organizations as open systems that are enmeshed with their environments rather than as stable entities with defined boundaries. The focus on the organization-environment nexus as a critical factor in accounting for how and under which conditions external changes and developments can inform and stimulate organizational learning draws attention to the question of how organizations achieve openness and work with different kinds of knowledge that are not necessarily coherent. It goes hand in hand with a question confronting many organizations—how to address dynamism, multiplicity, and variety in their environments (Easterby-Smith at al. 2000).

The notion of boundary-spanning activities is integrated with organiza-tional learning theories to contribute to a differentiated understanding of the impact that the organization-environment nexus has on the accomplishment of learning. The present study presents evidence that interaction with external stakeholders is an important means of assembling knowledge deemed relevant for organizational activities. It shows that differences in the ways an organi-zation achieves openness are accompanied by an emergence of distinctive learning qualities. Whether knowledge is gained intentionally or unintentionally depends on the social constitution of boundary-spanning activities, including their goals, perceptual filters, and regulation, and crucially affects the way members of the organization become aware of the acquired knowledge and make sense of it. Learning from boundary-spanning activities is significant for the internal dynamics and status quo of organizations. It may transform established ways of seeing and doing things. Issues of power and politics and regulatory aspects bear directly on the extent to which an organization refines existing practices or invents new ones through learning.

145

The Formation of Learning through

Boundary-spanning Activities

Focusing on organizational learning from boundary-spanning activities high-lights the ability of organizations to cope with changes and developments in their environments. But to spell out what it means for an organization to act in a predictably unstable environment, the discussion about forms of learning in theory must be examined. This study’s presented evidence of distinctive qual-ities of learning permits the assertion that the Commission may increase its ability to affect the development of European programs and policies strategically through what Argyris and Schön (1978) call single-loop and double-loop learning. Single-loop learning is used to accentuate shifts in organizational strategies and assumptions within a constant framework of norms for perfor-mance. Double-loop learning, on the other hand, is a process in which organiza-tional norms themselves are modified to enable path-creating developments in organizational thought and action. Whereas single-loop learning hones current skills and products and improves processes, double-loop learning challenges ways of seeing and doing things.

A slightly different categorization is the one by March (1991), which distinguishes between the “exploitation of old certainties” and the “exploration of new possibilities” (p. 71). March relates both forms of learning to those adaptive processes in organizations in which choices have to be made between using current information to improve present returns and gaining new informa-tion about alternatives to improve future returns. According to March, there is a trade-off between exploration and exploitation, and it has certain characteristic features in the social context of organizations. The empirical evidence of organizational learning in the European Commission indicates a number of them and thus specifies the settings in which the choices between exploration and exploitation are made.

An organization may explore new possibilities if its decision-makers engage in boundary-spanning activities with external stakeholders to gain access to knowledge that has apparently little to do with what the organization does but rather affords a larger picture of what it could be doing in the future. They give sense to the acquired knowledge by “guiding,” which is a highly controlled and animated way of constructing and promoting understandings of external changes and developments to be coped with by the organization (Maitlis 2005, p. 32). This process of sense-giving is a fundamental aspect of leadership activity be-cause it sets out a new framework for organizational activity with different criteria for performance measurement. The exploration of new possibilities in-side an organization is facilitated by the existence of a formalized management system, which serves the wide distribution and use of the acquired knowledge (Shrivastava 1983).

146

But in the more mundane world of most organizations, the system only apparently fosters an objective and impersonal absorption of acquired know-ledge. Organizational members in control of the management system can do much to shape the sense-making of the shared knowledge inside the organi-zation. Doing so gives rise to contested interpretations and accounts of the acquired knowledge that are far from being broadly shared. This process fuels exploration. It indicates the importance that an organization’s structure of pre-viously related knowledge has for the occurrence of exploration. Referring to an organization’s “absorptive capacity,” Cohen and Levinthal (1990) argue that innovative capabilities cannot be explained without looking at the ability of organizations to recognize, assimilate, and use new, external information. The existence of diverse, but related, knowledge structures enhances the capacity of an organization to make “novel linkages and associations beyond what any one individual can achieve” (Cohen & Levinthal 1990, p. 133).

Learning to cope with changes in the environment through exploitation appears to be fundamentally different from exploration. Exploitation may result if an organization uses its boundary-spanning activities primarily as a way to meet the expectations of regulatory bodies in its environment (Child 1997). If operating within an environment that imposes certain conditions on good performance, an organization is rather limited in its scope of action because the criteria for performance measurement lie outside the organization and cannot be changed fundamentally. Regulatory bodies in the organization’s environment are therefore in a position to influence the goals that are set for organizational learning. Granted, an organization that must respond to feedback from the environment is provided with learning opportunities. But because of the institu-tional obligations associated with those opportunities, learning takes shape as exploitation.

Each perceived increase in competence in established activity increases the likelihood of that activity’s rewards, such as public support and legitimacy (March 1991, March & Olsen 1989). If regulatory bodies or other powerful groups are empowered to limit an organization’s learning behavior to their expectations, the organization may respond by protecting its core activities from external influences. In other words, attempts to gain legitimacy through exploitation can lead to “buffering” in organizations (Brunsson 1989, Neergaard 1992). Buffering takes shape in skillful negotiation with external stakeholders and the maintenance of rather loose, informal links between that negotiation and internal decision-making processes. Minimal interest in sharing and making sense of the acquired knowledge, and lack of previously related knowledge inside an organization, contribute to buffering and thus reward positive local feedback and path dependence more than encourage a search for new ideas and relations (March 1991).

147

The distinction between the two forms of learning as described above is clearly analytical. The different processes of learning observed in this study come about in a single organization within the same period and are interrelated. The evidence seems to contradict March’s (1991) idea that organizations choose between the exploration of new opportunities and the exploitation of old uncertainties. To provide leverage for the institutionalist view that it is legiti-macy rather than superior performance that explains organizational survival, March argues that the trade-off between exploration and exploitation is biased toward the latter. In a seminal article on the myopia of organizational learning, Levinthal & March (1993) reiterate the trade-off argument by pointing to the limits of improving organizational performance through learning. They maintain that organizational learning privileges the short run and the near neighborhood and ignores the long run and the larger picture. However, bringing the environment into the analysis of organizational learning and looking at the ways in which perceptions of external changes interact with features of the setting in which learning occurs implies that learning is a collective activity of both preservation and innovation (Cook & Yanow 1993). Different forms of learning may exist simultaneously in organizations, and the intertwining between them is part and parcel of what constitutes organizational learning, namely, the “movement between familiar and emergent activities and between established and emergent social relations” (Blackler & MacDonald 2000, p. 833).

Complementarities in the Contextualized

Learning Model

The present study is based on a stage model of organizational learning, one that embeds learning processes in the social context that sustains them. To show that learning is a social accomplishment, it draws attention to the conditions and circumstances that shape the knowledge-seeking behavior of an organization, the distribution and interpretation of the acquired knowledge, and the integration of that knowledge with organizational memory. The turn to the social constitu-tion of learning has at least two consequences. First, it points to issues of power and politics and how they interact with processes of organizational learning. Second, it highlights the processes of creating, sustaining, and changing intersubjective meanings. Because the present study strives to develop a more socially aware stance on learning and knowing, it contributes to a recurrent debate about the essence of organizational learning. In a discussion about what learning is and where to locate it, Easterby-Smith, Snell, and Gherardi (1998, p. 264) describe the differing views as a divergence between two conceptuali-zations of organizational learning: one drawing on individual and cognitive con-structs, the other portraying organizational learning as a sociocultural process.

148

Proponents of the latter perspective criticize adherents of the former one for their emphasis on information-processing and their treatment of information as an objective, transferable good (Child & Heavens 2001).

Organizational learning from boundary-spanning activities with external stakeholders is about the efforts organizations make to cope with changing developments and trends in their environments. To provide fruitful ground for examining the linkages between the constituent processes of learning, it must be approached as a process that is essentially political (Friedberg 2003). Boundary-spanning activities are a source of power because they determine the perceptions of external stakeholders with which an organization engages. Moreover, they “always represent a source of potential disturbance of [the organization’s] internal functioning and hence a major and unavoidable source of uncertainty” (Crozier & Friedberg 1980, p. 41). In this view, acquiring knowledge through boundary-spanning activities moves the organization into a “zone of uncertain-ty” controlled by certain members of an organization. But to affect the capacity of the relevant parties to become involved in learning and to influence their interest in doing so, the knowledge must be relevant to them and to the issues at hand (Hanft 1996). Relevance can be created through effective handling of the communication and information flow among the organization’s units and mem-bers, and through the judicious use of organizational rules. The way in which these sources of power are mobilized among the specific actors depends on how the distinct issues in each part of the learning process are handled. These variables are what give rise to distinct forms of learning.

Analysis of learning processes on the basis of the contextualized stage model is an attempt to overcome the weaknesses of focusing on the information-processing in Huber’s (1991) model of learning. The factors that have been identified to explain how a set of functionally differentiated boundary-spanning activities trigger distinctive qualities of learning reflect the tension inherent in the contextualized model of learning. The evidence of these activities blurs the boundaries between the distinct learning stages by showing that similar factors, including regulatory aspects and perceptual filters, affect the distinct stages simultaneously. So does the relationship between the politics of organizational knowledge and learning. The findings indicate that learning from boundary-spanning activities takes place in a landscape of interests and differential power relations. The regard for issues of power and politics cannot be limited to the sense-making of knowledge gains; it must encompass the overall learning pro-cess. The study also provides a fresh look at the analytical value of examining organizational learning in terms of stages. The learning stages—the acquisition, distribution, interpretation, and storage of knowledge—are distinct in the sense that each involves different issues, including the achievement of openness and the stability of the status quo. When learning is examined from the perspectives

149

of the different issues in question, the approach opens ways to understand what the accomplishment of learning consists of.

The divergence between cognitive and sociocultural perspectives on organizational learning is at least partially offset by the advantages of using the contextualized learning model for an empirical investigation of learning pro-cesses. Combining both viewpoints within one framework links the creation of meaning to political processes, connects organizational thought to action, and captures learning processes in large formal organizations. Orientations centering on “meaning creation within a political framework” (Coopey & Burgoyne 2000, p. 871) have limited applicability to an examination of how organizations as a whole may learn, for they tend to reduce the complex phenomenon of learning to the generation and change of intersubjectivity. The combined approach calls for more than an analysis of learning in groups or subgroups within organi-zations and may therefore be seen as a useful starting point for an empirical investigation into the “development of shared meanings across wider and more strategically dominant collectivities” (Easterby-Smith et al. 1998, p. 265). Learning in organizations as a whole is, according to Easterby-Smith and col-leagues, a major gap in the learning discourse. Future research may contribute to the development of the contextualized model of organizational learning. A fruitful effort to do so pertains to the identification of complementarities in the underlying assumptions about the processual nature of learning, issues of power and politics, and agency. The proposed discussion about ontological distinctions depends in part on what one considers to be the essence of an “organization” (Argyris & Schön 1978, Scherf-Braune 2000).

Organizational Learning and the Institutionalization

of Organizational Life

This study argues for conceiving organizational learning as a process that is embedded in the social context of organizations, including the interpersonal relations within and across their external boundaries. The conditions and cir-cumstances of the selectiveness in the knowledge-seeking behavior of organi-zations and the transformation of the acquired knowledge into an organizational property give rise to distinct qualities of organizational learning. The theoretical model helps identify the organizational factors facilitating and inhibiting learning processes through boundary-spanning activities. But the model does not sufficiently explain the origins of the deciphered facilitators and barriers. The factors examined in this study to explain the scope and direction of learning tend to be treated as emanating from within the focal organization rather than from the society in which it is institutionally embedded. The empirical evidence suggests that boundary-spanning activities are governed by regulatory and

150

normative features of the wider social environment and, hence, by externally fixed institutions. They can be seen as constraints on the learning behavior of an organization because they affect what it is able to learn and how it goes about its learning processes (see Geppert 2000, Wiesenthal 2000).

Institutional perspectives stress the tendency of organizations to conform to the expectations and practices of societal institutions, for it promotes their legitimacy and provides them with vital resources (Meyer & Rowan 1977). Institutions are defined in this perspective as normative obligations that take on a rule-like status in social thought and action. Institutions are the rules of the game and organizations are the players (North 1990, pp. 3–4). In the case of the European Commission, such normative obligations include supranationalism, agenda-setting, and autonomy from national influence. These are general expec-tations based on the constitutional and house rules of the Commission (Hooghe 2005). Assuming that most behavior in organizations is based on routines and axiomatic aspects of organizational life, proponents of institutional approaches (DiMaggio & Powell 1991, Meyer & Rowan 1977, Scott 1992) argue that organizations incorporate externally fixed institutions in their behavior to gain legitimacy and reduce the likelihood of dramatic fluctuations in products, techniques, or policies. From this perspective, organizations are institutionally constituted rather than constructs of more or less rational decisions (DiMaggio & Powell 1991, March & Olsen 1989).

Integrating institutional perspectives into theories of organizational learn-ing helps explain the influence of societal constraints on the ability of organi-zations to engage in learning processes. It may then be argued that societal institutions affect the scope for organizational learning understood as an “overarching normalizing process that shapes organizations through the influ-ence of societal ‘templates’ of values, beliefs, norms and structures” (Coopey & Burgoyne 2000, p. 873). In this view, organizational learning through boundary-spanning activities is a means of winning the support of an organization’s environment because it helps identify and meet the needs and interests of external groups. Institutionalism also implies rather low confidence in an organi-zation’s ability to shape the direction and nature of learning, but the findings of the present study support the assertion that learning generates legitimacy for the Commission’s implementation powers and agenda-setting. On way in which the Commission performs its role as the “steering body of the world’s most encompassing supranational regime” (Hooghe 2005, p. 862) is through routin-ized regard for Member State interests during all stages of the policy process. Beyond that obligation, the Commission generates “hegemonial discourses” (Schaper-Rinkel 2003, p. 231) on the content and direction of European policies. Belying institutionalism’s skepticism that self-directed learning is possible of organizations, the Commission engages in proactive agenda-setting in a widen-

151

ing range of policy areas as a response to the manifold and often inconsistent expectations of governments and the stakeholders from business and society.

The Commission is institutionalized, but it is also the product of its organizational dynamics. It is an actor that cannot be “artificially separated from the process that sets constraints on [its] behavior” (Majone 1989, p. 96). As the members of the political Commission colleges come and go, the Commission’s internal staff in the various Directorates General and other services continuously pursue their goals within the given institutional framework, attempting to modify it to their benefit. Commission learning surely emanates from the wider social environment in which it is embedded, but that learning is ultimately the result of internal factors. It culminates in the bureaucratization of European integration, which appears impersonal and technocratic and is far away from public scrutiny.

152

References

Abels, G. (2000): Strategische Forschung in den Biowissenschaften. Der

Politikprozeß zum europäischen Humangenomprogramm. Berlin: edition sigma.

Aldrich, H. A. (1979): Organizations & Environments. Englewood Cliffs: Prentice-Hall.

Adams, S. (1980): “Interorganizational Processes and Organization Boundary Activities”. In: Research on Organizational Behavior, 2, pp. 321-355.

Ancona, D. G. and D. F. Caldwell (1992): “Bridging the Boundary: External Activity and Performance in Organizational Teams”. In: Administrative

Science Quarterly, 37(4), pp. 634-665.

Argyris, C. and D. A. Schön (1978): Organizational Learning: A Theory of

Action Perspective. Reading, MA: Addison-Wesley.

Axelrod, R. (1976): “The Analysis of Cognitive Maps”. In: Axelrod, R. (ed.) Structure of Decision. The Cognitive Maps of Political Elites. Princeton University Press, pp. 55-73.

Bach, M. (1999): Die Bürokratisierung Europas. Frankfurt am Main: Campus.

Barnett, M.N. and M. Finnemore (1999): “The Politics, Power, and Pathologies of International Organizations”. In: International Organization, 53(4), pp. 699-732.

Bender, G. (1999): “Technologische Innovation als Form der europäischen Integration. Zur Entwicklung des europäischen Mobilfunkstandards”. In: Zeitschrift für Soziologie, 28(2), pp. 77-92.

Berthoin Antal, A. (1997): “Organizational learning processes in downsizing”. Wissenschaftszentrum Berlin für Sozialforschung, Discussion Paper FS II 97-113.

Berthoin Antal, A. (1998): “Die Dynamik der Theoriebildungsprozesse zum Organisationslernen”. In: Albach, H., M. Dierkes, A. Berthoin Antal and K. Vaillant (eds.) Organisationslernen – institutionelle und kulturelle

Dimensionen. Berlin: edition sigma, pp. 31-52.

Berthoin Antal, A. (2000): “Types of Knowledge Gained by Expatriates”. In: Journal of General Management, 26(2), pp. 32-51.

Berthoin Antal, A. (2003): “Organizational Learning: How Can it Help the Biodiversity Field Evolve?” Workshop on the ‘Incentive Politics in the

153

First 10 Years of the Convention on Biodiversity’, 17-19 July 2003, Wissenschaftszentrum Berlin für Sozialforschung.

Berthoin Antal, A., M. Dierkes and S. Helmers (1993): “Unternehmenskultur. Eine Forschungsagenda aus Sicht der Handlungsperspektive.” In: Dierkes, M., L. von Rosenstiel and U. Steger (eds.) Unternehmenskultur in Theorie

und Praxis. Frankfurt am Main: Campus, pp. 200-218.

Berthoin Antal, A. and K. Böhling (1998): “Expatriation as an Underused Resource for Organizational Learning.” In: Albach, H., M. Dierkes, A. Berthoin Antal and K. Vaillant (eds.) Organisationslernen – institutionel-

le und kulturelle Dimensionen. Berlin: edition sigma, pp. 215-236.

Berthoin Antal, A., M. Dierkes, J. Child and I. Nonaka (2001a): “Introduction. Finding Paths through the Handbook”. In: Dierkes, M., A. Berthoin Antal, J. Child and I. Nonaka (eds.) Handbook of Organizational Learning and

Knowledge. Oxford: Oxford University Press, pp. 1-7.

Berthoin Antal, A., M. Dierkes, J. Child and I. Nonaka (2001b): “Organizational Learning and Knowledge: Reflections on the Dynamics of the Field and Challenges for the Future”. In: Dierkes, M., A. Berthoin Antal, J. Child and I. Nonaka (eds.) Handbook of Organizational Learning and

Knowledge. Oxford: Oxford University Press, pp. 921-939.

Berthoin Antal, A. and C. Krebsbach-Gnath (2001): “Consultants as Agents of Organizational Learning: The Importance of Marginality”. In: Dierkes, M., A. Berthoin Antal,, J. Child and I. Nonaka (eds.) Handbook of

Organizational Learning and Knowledge. London: Oxford University Press, pp. 462-483.

Berthoin Antal, A., U. Lenhardt and R. Rosenbrock (2001): “Barriers to Organizational Learning”. In: Dierkes, M., A. Berthoin Antal, J. Child and I. Nonaka (eds.) Handbook of Organizational Learning and Knowledge. Oxford: Oxford University Press, pp. 865-885.

Blackler, F. and S. McDonald (2000): “Power, Mastery and Organizational Learning”. In: Journal of Management Studies 37(6), pp. 833-851.

Böhling, K. (2001): “Zur Bedeutung von 'boundary spanning units' für Organi-sationslernen in internationalen Organisationen”, Wissenschaftszentrum Berlin für Sozialforschung, Discussion Paper FS II 01-101.

Böhling, K. (2002): “Learning from Environmental Actors about Environmental Developments – The Case of International Organizations”, Wissenschafts-zentrum Berlin für Sozialforschung, Discussion Paper FS II 02-110.

Brown, M. L. (2000): “Scientific uncertainty and learning in European Union environmental policymaking”. In: Policy Studies Journal, 28(3), pp. 576-596.

154

Brunsson, N. (1989): The Organization of Hypocrisy. Talk, Decisions and

Actions in Organizations. Chichester: John Wiley & Sons.

Brücker, A., C. Joerges, J. Neyer and S. Schlacke (1997): “Formen europäischer Risikoverwaltung: Komitologie im Bereich von Lebensmitteln und technischen Gütern”. In: König, T., E. Rieger and H. Schmitt (eds.) Euro-

päische Institutionenpolitik. Frankfurt am Main: Campus, pp. 178-199.

Büchel, B. and S. Raub (2001): “Media Choice and Organizational Learning”. In: Dierkes, M., A. Berthoin Antal, J. Child and I. Nonaka (eds.) Hand-

book of Organizational Learning and Knowledge. Oxford: Oxford University Press, pp. 518-534.

Child, J. (1997): “Strategic Choice in the Analysis of Action, Structure, Organi-zations and Environment: Prospect and Retrospect”. In: Organization

Studies, 18(1), pp. 43-76.

Child, J. and C. Smith (1987): “The Context and Process of Organizational Transformation – Cadbury Limited in its Sector”. In: Journal of Manage-

ment Studies, 24(6), pp. 565-593.

Child, J. and S. Heavens (2001): “The Social Constitution of Organizations and its Implications for Organizational Learning”. In: Dierkes, M., A. Berthoin Antal, J. Child and I. Nonaka (eds.) Handbook of Organizational

Learning and Knowledge. Oxford: Oxford University Press, pp. 308-326.

Christiansen, T. (1997): “Tensions of European governance: politicized bureau-cracy and multiple accountability in the European Commission”. In: Journal of European Public Policy, 4(1), pp. 73-90.

Cini, M. (1997): “Administrative Culture in the European Commission: The Cases of Competition and Environment”. In: Nugent, N. (ed.): At the

heart of the Union. Studies of the European Commission, London: MacMillan Press, pp. 71-88.

Cohen, W. M. and D. A. Levinthal (1990): “Absorptive Capacity: A New Perspective on Learning and Innovation.” In: Administrative Science

Quarterly, 35, pp. 128-152

Cook, S. D. N. and D. Yanow (1993): “Culture and Organizational Learning”. In: Journal of Management Inquiry, 2(4), pp. 373-390.

Coopey, J. (1995): “The Learning Organization, Power, Politics and Ideology”. In: Management Learning, 26(2), pp. 193-213.

Coopey, J. and J. Burgoyne (2000): “Politics and Organizational Learning”. In: Journal of Management Studies, 37(6), pp. 869-885.

Cram, L. (1994): “The European Commission as a Multi-Organization: Social Policy and IT Policy in the EU”. In: Journal of European Public Policy, 1(2), pp. 196-217.

155

Crozier, M. and E. Friedberg (1980): Actors and Systems. The Politics of

Collective Action. Chicago, London: University of Chicago Press.

Czarniawska-Joerges, B. (1996): “The Process of Organizing”. In: Warner, M. (ed.) International Encyclopedia of Business and Management. London, New York: Routledge, pp. 3966-3981.

Daft, R. L. and K. E. Weick (1984): “Toward a Model of Organizations as Inter-pretation Systems”. In: Academy of Management Review, 9, pp. 284-295.

DiBella, A.J., D.C. Nevis and J.M. Gould (1996): “Understanding Organi-zational Learning Capability”. In: Journal of Management Studies, 33(3), pp. 361-379.

Dierkes, M. (1988): “Unternehmenskultur und Unternehmensführung. Konzept-ionelle Ansätze und gesicherte Erkenntnisse”. In: Zeitschrift für Betriebs-

wirtschaft, 58(5/6), pp. 554-575.

Dierkes, M. (1997): Technikgenese. Befunde aus einem Forschungsprogramm. Berlin: edition sigma.

Dierkes, M., K. Hähner and A. Berthoin Antal (1997): Das Unternehmen und

sein Umfeld. Wahrnehmungsprozesse und Unternehmenskultur am Bei-

spiel eines Chemiekonzerns. Frankfurt am Main: Campus.

Dierkes, M. and L. Marz (1998): “Leitbilder als Katalysatoren des Organisa-tionslernens”. In: Albach, H., M. Dierkes, A. Berthoin Antal and K. Vaillant (eds.) Organisationslernen – institutionelle und kulturelle Dimen-

sionen. Berlin: edition sigma, pp. 373-400.

Dierkes, M., L. Marz and C. Teele (2001): “Technological Visions, Techno-logical Development, and Organizational Learning”. In: Dierkes, M., A. Berthoin Antal, J. Child and I. Nonaka (eds.) Handbook of Organizational

Learning and Knowledge. Oxford: Oxford University Press, pp. 282-301.

DiMaggio, P.J. and W.W. Powell (1991): “Introduction”. In: Powell, W.W. and P.J. DiMaggio (eds.) The New Institutionalism in Organizational Ana-

lysis. Chicago, London: University of Chicago Press, pp. 1-38.

Dodgson, M. (1993): “Organizational Learning: A Review of Some Literatures”. In: Organization Studies, 14(3), pp. 375-393.

Easterby-Smith, M., R. Snell and S. Gherardi (1998): “Organizational Learning: Diverging Communities of Practice?” In: Management Learning, 29(3), pp. 259-272.

Easterby-Smith, M., M. Crossan and D. Nicolini (2000): “Organizational Learn-ing: Debates Past, Present and Future.” In: Journal of Management

Studies, 37(6), pp. 783-796.

156

Edmondson, A. and B. Moingeon (1998): “From Organizational Learning to the Learning Organization”. In: Management Learning, 29(1), pp. 5-20.

Edwards, G and D. Spence (1997): “The Commission in Perspective”. In: Edwards, G. and D. Spence (eds.) The European Commission, Maple House: Cartermill International Ltd., pp. 1-32.

Elizalde, J (1992): “Legal Aspects of Community Policy on Research and Technological Development (RTD).” In: Common Market Law Review, 29, pp. 309-346.

Falke, J. (2000): “Komitologie – Entwicklung, Rechtsgrundlagen und erste empirische Annäherung”. In: Joerges, C. and J. Falke (eds.) Das Aus-

schusswesen der Europäischen Union. Baden-Baden: Nomos, pp. 43-159.

Feld, W.J. and R.S. Jordan (1994): International Organizations. A Comparative

Approach. Westport, London: Praeger.

Feldman, M. and B. Pentland (2003): “Reconceptualizing Organizational Rou-tines as a Source of Flexibility and Change”. In: Administrative Science

Quarterly, 48(3), pp. 94-118.

Filion, N. and H. Rudolph (1999): “Power, Control and Organisational Learn-ing.” Wissenschaftszentrum Berlin für Sozialforschung, Discussion Paper FS I 99–104.

Fischer, F. (1990): Technocracy and the Politics of Experience. Managerial and

policy perspectives. London: Sage.

Fischer, F. (2003): Reframing Public Policy. Discursive Politics and Deliber-

ative Practices. Oxford: Oxford University Press.

Fox, A. (1974): Beyond Contract. Work, Power and Trust Relations. London: Faber and Faber Limited.

Freeman, R. E. (1984): Strategic Management. A Stakeholder Approach. Boston: Pitman.

Friedberg, E. (1995): Ordnung und Macht. Dynamiken organisierten Handelns. Frankfurt am Main: Campus.

Friedberg, E. (2003): “Mikropolitik und Organisationelles Lernen”, in: Brentel, H, H. Klemisch and H. Rohn. (eds.) Lernendes Unternehmen. Wiesbaden: Westdeutscher Verlag, pp. 97-108.

Geppert, M. (2000): Beyond the learning organization. Paths of organizational

learning in the East German context. Aldershot: Gower.

Gherardi, S. and D. Nicolini (2001): “The Sociological Foundations of Organi-zational Learning”. In: Dierkes, M., A. Berthoin Antal, J. Child and I. Nonaka (eds.) Handbook of Organizational Learning and Knowledge, Oxford: Oxford University Press, pp. 35-60.

157

Gherardi, S., D. Nicolini and F. Odella (1998): “Towards a Social Under-standing of how People Learn in Organizations. The Notion of Situated Curriculum”. In: Management Learning, 29(3), pp. 273-297.

Giddens, A. (1984): The Constitution of Society: Outline of the Theory of

Structuration. Cambridge: Polity Press.

Gnyawali, D.R. and A.C. Stewart (2003): “A Contingency Perspective on Organizational Learning: Integrating Environmental Context, Organi-zational Learning, and Types of Learning”. In: Management Learning 34(1), pp. 63-89.

Gordenker, L and P.R. Saunders (1978): “Organisation Theory and International Organisation”. In: Taylor, P. and A. J. R. Groom (eds.) International

Organisation. A Conceptual Approach. London: Pinter, pp. 84-107.

Grande, E. (1997): “Auflösung, Modernisierung oder Transformation? Zum Wandel des modernen Staates in Europa”. In: Grande, E. and R. Prätorius (eds.) Modernisierung des Staates. Baden-Baden: Nomos, pp. 45-63.

Granovetter, M. S. (1985): “Economic Action and Social Structure: The Problem of Embeddedness.” In: American Journal of Sociology, 91, pp. 481-510.

Haas, E. B. (1990): When knowledge is power. Three models of change in

international organizations. Berkeley: University of California Press.

Haas, E. B. (1964): Beyond the Nation State. Functionalism and International

Organization. Stanford: Stanford University Press.

Haas, P. M. (2000): “International Institutions and Social Learning in the Management of Global Environmental Risks.” In: Policy Studies Journal, 28(3), pp. 558-575.

Haas, P. M. (1992): “Introduction: Epistemic Communities and International Policy Coordination.” In: International Organizations, 26(1), pp. 1-35.

Haibach, G. (2000): “Governance by Committee: The Role of Committees in European Policy Making and Policy Implementation.” In: Eipascope, no. 2, pp. 38-42.

Hanft, A. (1996): “Organisationales Lernen und Macht – Über den Zusammen-hang von Wissen, Lernen, Macht und Struktur.” In: Schreyögg, G. and P. Conrad (eds.) Managementforschung 6, Berlin, New York: de Gruyter, pp. 133-162.

Hardy, C (1996): “Understanding Power: Bringing about Strategic Change.” In: British Journal of Management, Special Issue, 7, pp. S3-S16.

158

Hardy, C. and S. Clegg (1996): “Some dare call it power”. In: Hardy, C. and S. Clegg (eds.) Handbook of Organization Studies. London: Sage, pp. 622-640.

Hartlapp, M. (2005): Die Kontrolle der nationalen Rechtsdurchsetzung durch

die Europäische Kommission – Politik, Verbände, Recht: Umsetzung

europäischer Sozialpolitik, Band 3. Frankfurt am Main: Campus.

Hedberg, B. (1981): “How Organizations Learn and Unlearn”. In: Nystrom, P. and W. Starbuck (eds.) Handbook of Organizational Design. New York: Oxford University Press, pp. 3-27.

Héritier, A. (1997): “Policy-making by Subterfuge: interest accommodation, innovation and substitute democratic legitimation in Europe.” In: Journal

of European Public Policy, 4(1), pp. 171-189.

Hernes T. (2003): “Enabling and Constraining Properties of Organizational Boundaries”. In: Paulsen, N. and T. Hernes (eds.) Managing Boundaries

in Organizations. Multiple Perspectives. Hampshire, New York: Palgrave Macmillan, pp. 35-54.

Hernes, T. and N. Paulsen (2003): “Introduction: Boundaries and Organization”. In: Paulsen, N. and T. Hernes (eds.) Managing Boundaries in Organiza-

tions. Multiple Perspectives. Hampshire, New York: Palgrave Macmillan, pp. 1-13.

Hoeser, U. (2002): “In search of projects: the politics of multilateral develop-ment organizations”. European Group of Organization Studies (EGOS) Conference Paper, 4-7 July 2002, Barcelona.

Hooghe, L. (2005): “Several Roads Lead to International Norms, but Few Via International Socialization: A Case Study of the European Commission”. In: International Organization, 59(4), pp. 861-898.

Huber, G. P. (1991): “Organizational Learning. The Contributing Processes and the Literatures”. In: Organization Science, 2(1), pp. 88-115.

Jachtenfuchs, M. and B. Kohler-Koch (1996): Europäische Integration, Opladen: Leske+Budrich.

Jevnaker, B H. (2003): “Industrial Designers as Boundary Workers.” In: Paulsen, N. and T. Hernes (eds.) Managing Boundaries in Organizations.

Multiple Perspectives. Hampshire, New York: Palgrave Macmillan, pp. 110-128.

Kerlen, C. (2003): Problemlos beraten? Die Problemdefinition als Startpunkt

organisationalen Lernens. Berlin: edition sigma.

Kieser, A., N. Beck and R. Taino (2001): “Rules and Organizational Learning: The Behavioral Theory Approach”. In: Dierkes, M., A. Berthoin Antal,

159

J. Child and I. Nonaka (eds.) Handbook of Organizational Learning and

Knowledge. Oxford: Oxford University Press, pp. 598-625.

Kim, L. (2001): “Absorptive Capacity, Co-opetition, and Knowledge Creation”. In: Nonaka, I. and T. Nishiguchi (eds.) Knowledge Emergence. Oxford: Oxford University Press, pp. 270-285.

Kissling-Näf, I., D. Marek and P. Gentile (1994): Politikorientierte Lernprozes-

se: Analysekonzept zur empirischen Erhebung im Feld. Lausanne: Cahier de L'IDHEAP, no. 131.

Kissling-Näf, I. and P. Knoepfel (1998): “Lernprozesse in öffentlichen Poli-tiken”. In: Albach, H., M. Dierkes, A. Berthoin Antal and K. Vaillant (eds.) Organisationslernen – institutionelle und kulturelle Dimensionen. Berlin: edition sigma, pp. 239-268.

Klimecki, R. G. and M. Thomae (1997): “Organisationales Lernen. Eine Bestandsaufnahme der Forschung”. Management Forschung und Praxis, Universität Konstanz, no. 18, [http://www.vb.uni-konstanz.de/kops/ volltexte/2000/396] (downloaded 7 July 2006).

Knill, C. (2001): “Private governance across multiple arenas: European interest associations as interface actors”. In: Journal of European Public Policy, 8(2), pp. 227-246.

Kohler-Koch, B. (1997): “The European Union Facing Enlargement. Still a System sui generis?” In: Mannheimer Zentrum für Sozialforschung, Arbeitspapiere, AB3/no. 20.

Krebsbach-Gnath, C. (1996): Organisationslernen. Theorie und Praxis der

Veränderung. Wiesbaden: Deutscher Universitätsverlag.

Kuusi, O. and M. Meyer (2002): “Technological generalizations and leitbilder - the anticipation of technological opportunities”. In: Technological Fore-

casting & Social Change, 69, pp. 625-639.

Laffan, B. (1997): “From Policy Entrepreneur to Policy Manager: the Challenge facing the European Commission”. In: Journal of European Public

Policy, 4(3), pp. 422-438.

Lane, C. (2001): “Organizational Learning in Supplier Networks”. In: Dierkes, M., A. Berthoin Antal, J. Child and I. Nonaka (eds.) Handbook

of Organizational Learning and Knowledge. Oxford: Oxford University Press, pp. 699-715.

LaPalombara, J. (2001): “Power and Politics in Organizations: Public and Private Sector Comparisons”. In: Dierkes, M., A. Berthoin Antal, J. Child and I. Nonaka (eds.) Handbook of Organizational Learning and Know-

ledge. Oxford: Oxford University Press, pp. 557-581.

160

Levinthal, D. A. and J. G. March (1993): “The Myopia of Learning”. In: Strate-

gic Management Journal, 14, pp. 95-112.

Levitt, B. and J. G. March (1988): “Organizational Learning”. In: Annual

Review of Sociology, 14, pp. 319-340.

Lequesne, C. (2000): “The European Commission: A Balancing Act between Autonomy and Dependence”. In: Neunreither, K. and A. Wiener (eds.) European Integration after Amsterdam: Institutional Dynamics and

Prospects for Democracy. Oxford: Oxford University Press, pp. 36-51.

Luhmann, N. (1964): Funktionen und Folgen formaler Organisationen. Berlin: Duncker & Humblot.

Lukes, S. (1974): Power. A Radical View. London, New York: MacMillan.

Maitlis, S. (2005): “The Social Processes of Organizational Sensemaking”. In: The Academy of Management Journal, 48(1), pp. 21-49.

Majone, G. (1989): Evidence, Argument, and Persuasion in the Policy Process. New Haven, London: Yale University Press.

Malek, T. and L. Hilkermeier (2001): “The European Commission as a Learning Organization? Theoretical Considerations and Empirical Ideas”. European Consortium for Political Research Workshop Paper, 6-11 April 2001, Grenoble.

March, J. G. (1991): “Exploration and Exploitation in Organizational Learning”. In: Organization Science, 2, pp. 71-87.

March, J. G. and H. A. Simon (1958): Organizations. New York: John Wiley & Sons.

March, J. G. and J. P. Olsen (1975): “The Uncertainty of the Past: Organiza-tional Learning under Ambiguity”. In: European Journal of Political

Research, 3, pp. 147-171.

March, J. G. and J. P. Olsen (1989): Rediscovering Institutions. The Organi-

zational Basis of Politics. New York: The Free Press.

Mazey, S. and J. Richardson (1997): “The Commission and the Lobby”. In: Edwards, G. and D. Spence (eds.): The European Commission. Maple House: Cartermill International Ltd, pp. 178-212.

McDonald, M (2000): “Accountability, Anthropology and the European Com-mission”. In: Strathern, M. (ed.) Audit Cultures, Anthropological Studies

in Accountability, Ethics and the Academy. London, New York: Routledge, pp. 106-132.

Metcalfe, L. (1996): “Building capacities for integration: the future role of the Commission”. In: Eipascope, no. 2, pp. 2-8.

161

Meyer, J. W. and B. Rowan (1977): “Institutionalized Organizations: Formal Structure as Myth and Ceremony”. In: American Journal of Sociology, 83(2), pp. 340-363.

Messner, D. (1998): “Die Transformation von Staat und Politik im Globali-sierungsprozess”. In: Messner, D. (ed.) Die Zukunft des Staates und der

Politik. Bonn: Dietz, pp. 14-43.

Meuser, M. and U. Nagel (1991): “ExpertInneninterviews – vielfach erprobt, wenig bedacht.” In: Garz, D. and K. Kraimler (eds.) Qualitativ-

Empirische Sozialforschung. Konzepte, Methoden, Analysen. Opladen: Westdeutscher Verlag, pp. 411-471.

Mintzberg, H. (1983): Structure in Fives: Designing Effective Organizations. London: Prentice-Hall.

Münch, R. (1995): Das Projekt Europa. Zwischen Nationalstaat, regionaler

Autonomie und Weltgesellschaft. Frankfurt am Main: Suhrkamp.

Ness, G. D. and S. R. Brechin (1988): “Bridging the gap: international organiza-tions as organizations”. In: International Organization, 42(2), pp. 245-273.

Neergaard, P. (1992): Management Accounting and Uncertainty. Frederiksberg: Samfundslitteratur.

Neyer, J. (2000): “Justifying Comitology: The Promise of Deliberalization”. In: Neunreither, K. and A. Wiener (eds.) European Integration after

Amsterdam: Institutional Dynamics and Prospects for Democracy. Oxford: Oxford University Press, pp. 112-128.

Nonaka, I. and H. Takeuchi (1995): The Knowledge Creating Company. New York: Oxford University Press.

Nonaka, I., R. Toyama and P. Byosière (2001): “A Theory of Organizational Learning and Knowledge Creation: Understanding the Dynamic of Creating Knowledge”. In: Dierkes, M., A. Berthoin Antal, J. Child and I. Nonaka (eds.) Handbook of Organizational Learning and Knowledge. London: Oxford University Press, pp. 491-517.

North. D. C. (1990): Institutions, Institutional Change and Economic Perform-

ance. Cambridge: University Press.

Paulson, N. and T. Hernes (eds.) (2003a): Managing Boundaries in Organi-

zations. Multiple Perspectives. Hampshire, New York: Palgrave Macmillan.

Paulsen, N. and T. Hernes (2003b): “Epilogue: A Reflection and Future Directions”. In: Paulsen, N. and T. Hernes (eds.) Managing Boundaries in

Organizations. Multiple Perspectives, Hampshire, New York: Palgrave Macmillan, pp. 302-308.

162

Pentland, B. T. and H. H. Rueter (1994): “Organizational routines as grammars of action”. In: Administrative Science Quarterly, 39, pp. 484-510.

Peterson, J. and E. Bomberg (1999): Decision-Making in the European Union. London: MacMillan.

Peterson, J. and M. Sharp (1998): Technology Policy in the European Union. London: MacMillan.

Pierson, P. (1996): “The Path to European Integration. A Historical Institu-tionalist Analysis”. In: Comparative Political Studies, 29(2), pp. 123-163.

Pollack, M. A. (1998): “The Engines of Integration? Supranational Autonomy and Influence in the European Union”. In: Sandholtz, W. and A. Stone Sweet (eds.) European Integration and Supranational Governance. Oxford: Oxford University Press, pp. 217-249.

Pompidou, A., J. Affenzeller, P. Boekholt, R. Cochran, F. Garas, G. Hains, and K. J. Linden (2000): Five Year Assessment Report Relates to the Specific

Program: User-Friendly Information Society. Covering the Period 1995 –

1999. [http://www.cordis.lu/ist/publications.htm] (downloaded 10 October 2000).

Popper, M. and R. Lipshitz (2000): “Organizational Learning: Mechanisms, Culture, and Feasibility”. In: Management Learning, 31, pp. 181-196.

Radaelli, C.M. (2003): “The Open Method of Coordination: A new governance architecture for the European Union?” Swedish Institute for European Policy Studies, Report no. 1.

Ragin, C. (1994): Constructing Social Research. Thousand Oaks: Pine Forge Press.

Risse, T. (2000): “‛Let's Argue!’: Communicative Action in World Politics”. In: International Organization, 54(1), pp. 1-39.

Roobeek, A.J.M. (1990): Beyond the Technology Race. An Analysis of

Technology Policy in Seven Industrial Countries. Amsterdam: Elsevier.

Sabatier, P.A. (1993): “Advocacy Koalitionen, Policy Wandel und Policy Lernen: Eine Alternative zur Phasenheuristik”. In: Héritier, A. (ed.) Policy

Analyse: Kritik und Neuorientierung. PVS-Sonderheft 24, Opladen, pp. 116-148.

Sandholtz, W. (1998): “The Emergence of a Supranational Telecommunications Regime”. In: Sandholtz, W. and A. Stone Sweet (1998) (eds.) European

Integration and Supranational Governance. Oxford: Oxford University Press, pp. 134-163.

163

Sbragia, A. M. (ed.) (1992): Euro-Politics. Institutions and Policymaking in the

“New” European Community. Washington D.C.: The Brookings Institu-tion.

Schaper-Rinkel, P. (1999): “Zur politischen Produktion von Sachzwängen: Die europäische Informationsgesellschaft als regulativer Rahmen zur Siche-rung der Wettbewerbsfähigkeit”. In: PROKLA 1/1999, pp. 29-53.

Schaper-Rinkel, P. (2003): Die europäische Informationsgesellschaft. Techno-

logische und politische Integration in der europäischen Politik. Münster: Westfälisches Dampfboot.

Scherf-Braune, S. (2000): Organisationales Lernen. Ein systemtheoretisches

Modell und seine Umsetzung. Wiesbaden: DUV.

Schmitter, P. C. (1996): “Imagining the Future of the Euro-polity with the Help of New Concepts.” In: Marks. G, F.W. Scharpf, P.C. Schmitter and W. Streeck (eds.) Governance in the European Union. London: Sage, pp. 121-150.

Schreyögg, G. and P. Eberl (1998): “Organisationales Lernen: Viele Fragen, noch zu wenig neue Antworten”. In: DBW, 58(4), pp. 516-536.

Scott, W. R. (1992): “The Organization of Environments: Network, Cultural, and Historical Elements”. In: Meyer, J. W. and W. R. Scott (eds.) Organi-

zational Environments, Ritual and Rationality. London: Sage, pp. 155-175.

Selwyn, N. (2001): “‛E-Establishing’ an Inclusive Society? Technology, Social Exclusion and UK Government Policy Making”. In: Journal of Social

Policy, 31(1), pp. 1-20.

Senarclens, P. de (2001): “International Organisations and the Challenges of Globalisation”. In: International Social Science Journal, December 2001, pp. 509-522.

Senge, P. M. (1990): The Fifth Discipline. New York: Doubleday.

Shrivastava, P. (1983): “A Typology of Organizational Learning Systems”. In: Journal of Management Studies, 20(1), pp. 7-28.

Spence, D. (1997): “Staff and personnel policy in the Commission”. In: Edwards, G. and D. Spence (eds.) The European Commission, 2nd edition. London: Cartermill, pp. 68-100.

Stone, D. (ed.) (2000): Banking on Knowledge. The Genesis of the Global

Development Network. London, New York: Routledge.

Stone Sweet, A. and W. Sandholtz (1998): “Integration, Supranational Gover-nance, and the Institutionalization of the European Polity”. In:

164

Sandholtz, W. and A. Stone Sweet (eds.) European Integration and

Supranational Governance. Oxford: Oxford University Press, pp. 1-26.

Stopford, J. M. (2001): “Organizational Learning as Guided Responses to Market Signals”. In: Dierkes, M., A. Berthoin Antal, J. Child and I. Nonaka (eds.) Handbook of Organizational Learning and Knowledge. Oxford: Oxford University Press, pp. 264-281.

Strassheim, H. (2004): “Wissensgenerierung und Wissenstransfer in Netzwerken der lokalen Beschäftigungspolitik”. In: Edeling, T., W. Jann and D. Wagner (eds.) Wissenssteuerung und Wissensmanagement in Politik

und Verwaltung. Wiesbaden: VS Verlag, pp. 57-89.

Sutcliffe, K. (2001): “Organizational Environments and Organizational Informa-tion Processing.” In: Jablin, F.M. and L.C. Putnam (eds.) The New Hand-

book of Organizational Communication. Advances in Theory, Research

and Methods. Thousand Oaks: Sage, pp. 197-230.

Tacke, V. (1997): “Systemrationalisierung an ihren Grenzen – Organisations-grenzen und Funktionen von Grenzstellen in Wirtschaftsorganisationen”. In: Schreyögg, G. and J. Sydow (eds.) Gestaltung von Organisations-

grenzen. Berlin: de Gruyter, pp. 1-44.

Thompson, J. D. (1962): “Organizations and output transactions”. In: American

Journal of Sociology, 67, pp. 309-324.

Thompson, J. D. (1967): “Boundary spanning in different types of organi-zations”. In: J. D. Thompson (ed.) Organizations in action. Social science

bases of administrative theory. New York: McGraw-Hill, pp. 66-73.

Tushman, M. L. and M. L. Scanlan (1981): “Boundary Spanning Individuals: Their Role in Information Transfer and their Antecedents”. In: Academy

of Management Journal, 24(2), pp. 289-305.

Ventriss, C. and J. Luke (1988): “Organizational Learning and Public Policy: Towards a Substantive Perspective”. In: American Journal of Public

Administration, 18(4), pp. 337-355.

Walsh, J. G. (1995): “Managerial and Organizational Cognition: Notes from a Trip Down memory Lane”. In: Organization Science, 6(3), pp. 280-321.

Weick, K. E. (1991): “The Nontraditional Quality of Organizational Learning”. In: Organization Science, 2(1), pp. 116-124

Weick, K. E. and M. G. Bougon (1986): “Organizations as Cognitive Maps. Charting Ways to Success and Failure”. In: Sims, H. P. and D. A. Gioia (eds.) The Thinking Organization. San Francisco, London: Joey-Bass Publications, pp. 102-135.

165

Weick, K. E. and S. J. Ashford (2001): “Communication and Organizational Learning”. In: Jablin, F. M. and L. Putnam (eds.) New Handbook of

Organizational Communication. Thousand Oaks: Sage, pp. 704-731.

Weiser, M. (1993): “Some Computer Science Issues in Ubiquitos Computing”. In: Communications of the ACM, 36(7), pp. 75-84.

Wiesenthal, H. (1995): “Konventionelles und unkonventionelles Organisations-lernen: Literaturreport und Ergänzungsvorschlag”. In: Zeitschrift für

Soziologie, 24(2), pp. 137-155.

Wiesenthal, H. (2000): “Neun Thesen zur Konzipierung tiefgreifenden Orga-nisationslernens”, Workshop “Kollektives Lernen und Governance”, 24 June 2005, Technische Universität Darmstadt.