Competitive Cooperation

This paper examines the institutional and social dimensions of cooperation in the Alliance of Science Organisations, the central corporatist stakeholder in German science policy, in the 1970s and 1980s, which were a crucial period for this committee. In doing so, this essay mainly focuses on the way science organizations interact with each other, as well as with national politics. The Federal Ministry of Research invited the Alliance to regular meetings and thereby fostered its involvement into political decision-making processes. Consequently, the question of who belonged to the Alliance came into the focus of different players. Although the members of the Alliance themselves decided on the composition of their committee, they were not able to completely insulate themselves from external demands. Including new members into the Alliance had a destabilizing effect on the carefully balanced distribution of power within this committee, as will be shown through the case study on the admission of the Association of the Major Research Centers (Arbeitsgemeinschaft der Großforschungseinrichtungen, AGF) in 1976. In order to restabilize the situation, the members of the Alliance tried to exclude the AGF from certain issues. At the same time, the AGF itself was keen on being regarded as an equal partner and thus strove for its inclusion. This complex interplay of cooperative and competitive actions finally resulted in the institutionalization of the Alliance.


Introduction: Cooperation, Competition and Their Effects on Processes of Institutionalization
The political thinking of the Federal Republic of Germany in the s and early s was characterized by a phase of planning euphoria and the desire for closer links between science and politics (Trischler : -; Orth : -). This in turn was shaped both by the perception of living in an age dominated by science and technology and by the hope of being able to solve general social problems with the help of scientific policy advice (Rudloff : -; Schimank : -; Seefried : -). Various ministries, as well as the parliament,  increasingly consulted experts and scientific findings were incorporated into the development of political programs.  Participation in political decisionmaking processes enabled scientific stakeholders to actively shape politics. Consequently, although the spheres of science and politics follow their own internal logics, the two are closely interrelated and serve as resources for each other (Ash , : -, ; Bourdieu : -). Since the s, the Federal Republic of Germany has increasingly seen both a scientification of politics and at the same time a politicization of science-and thus an even closer intertwining of those areas. These two processes are by no means mutually exclusive but can be regarded as complementary, sometimes even mutually dependent (Rudloff : -; Weingart , : -). Consequently, a quantitative increase in scientific policy advice and an increasing institutionalization of different advisory formats can be observed during this time (Rudloff : -).
It is not a coincidence that this period also witnessed the informal establishment and gradual institutionalization of a committee  that remains influential in German science policy up until today: the Alliance of Science Organisations.  According to its own statements, the Alliance "brings together the top research organisations in Germany" (Hochschulrektorenkonferenz ). Its members regularly meet to discuss key issues of science policy or research funding. Furthermore they coordinate their financial interests, as all of them are funded by the public sector.  However, these institutions are extremely heterogeneous with regard to their basic legal status, their internal organization and their function within the German science system: Representatives of (non-)university research, research funding agencies and, finally, scientific advisory bodies come together in the Alliance and raise different concerns in the joint meetings. Hence, the question arises as to how access to and cooperation within the Alliance was (and is) regulated. In order to examine this question, this study focuses on the s and s, which represented a formative period for the Alliance. During these years, the Alliance underwent an initial phase of formalization of its cooperation. Simultaneously it had to deal for the first time with serious destabilization in its internal organization caused by the admission of a new member.
One might assume on the one hand that the members of the Alliance are in direct competition with each other for the limited financial resources of the federal and state governments. In addition, there are other areas of competition, such as for the best scientists or for top positions in rankings (Nickelsen ; Nickelsen & Krämer ; Brankovic et al. ). In this respect, the decision by the science organizations to act cooperatively appears all the more surprising, especially since their goals may overlap in many areas, which may further encourage competition and hinder cooperation (Simmel ; Vogt : -; Werron ; May ). On the other hand, the name "Alliance of Science Organisations" suggests that the committee's main purpose is to foster close cooperation.
When defining cooperation and competition with a focus on the participants' respective aims, the interconnection between these two modes of interaction becomes visible: In both cases, the players are striving for the same goal. While cooperation implies that the actors can only achieve their common goal when working together, competition implies a negative interdependency: Competitors can only achieve their aim if others fail to do so (Nickelsen ; Nickelsen & Krämer ; Soutschek & Nickelsen ; Simmel ). Yet in practice, we can observe the intertwining of these two modes of interaction within the Alliance of Science Organisations as it is neither a purely cooperative committee, nor do its members only compete with each other. Instead, the Alliance is a particularly good illustration of the instability of both modes of action as well as of the tense interconnection between cooperation and competition (Nickelsen & Krämer ). The internal cooperation between the members of the Alliance is subject to constant negotiation processes and is characterized by considerations as to whether a goal can be achieved better by acting cooperatively or individually. As a committee, the Alliance of Science Organisations, in turn, operates in other competitive constellations, in which, for example, access to resources of power and influence in setting the agenda for German science policy are of great importance. The joint appearance of the presidents and chairpersons of the most influential science organizations has always made their statements particularly effective, which is why the Federal Minister of Research invited the Alliance to meetings on a regular basis within the so-called Presidential Circle.  The Alliance thus is not only an intermediary actor of self-governing research, but also a science policy advisory body, which primarily serves the preliminary and informal coordination of interests between the science organizations and the ministry regarding structural changes in German research policy. This privileged access to the relevant political stakeholders in the field of science and research policy in turn makes membership in the Alliance of Science Organisations highly attractive to those standing outside. Consequently, the Alliance operates in the field of tension between cooperation and competition. The committee regularly has to face the question of whether the involvement of other actors strengthens its position, or if the inclusion of further perspectives in the joint discussions weakens its voice. Hence, this article examines how decisions about potential inclusion or exclusion were made: who decided on membership in the Alliance of Science Organisations and who determined the opening or closing of the committee? Which actors were regarded as capable of satisfying the membership requirements and therefore were allowed to enter this elite circle? And how do these processes of exclusion and inclusion affect the inner workings of the Alliance? On a broader conceptual level, two important questions arise: In what way are cooperative and competitive practices interrelated in the Alliance, the central coordinating committee of the science organizations? And how did these social practices influence the German science system on an institutional level? By exploring these central topics, this article will contribute to discussions on the dynamics of cooperation and competition, focusing in particular on their social and institutional dimensions.
The first part of this essay aims to clarify the extent to which the Alliance of Science Organisations can be considered an institutionalized coordination and advisory body, despite the informality of its meetings. In a first step, the development and progressive formalization of the Alliance in the s and s will be examined, which has received little attention in historical and social-science research so far. In doing so, the institutionalization of the meetings of the so-called Presidential Circle will be discussed as a corporatist attempt to shape German science politics. The process of institutionalization formalizes and consolidates structures, and thus regulates inclusions and exclusions, also regarding participation in political advisory bodies (Wenninger et al. : -).
In the second part, a specific case study of the practices of inclusion and exclusion is analyzed. This section concerns the relations of the Association of the Major Research Centers (Arbeitsgemeinschaft der Großforschungseinrichtungen, AGF)  to the established German science organizations. The inclusion of this association, which later developed into the Helmholtz Association, represented the first enlargement of the Alliance. For the analysis of the Alliance's complex relation to the AGF, this article will draw on some historical and social-science studies on the development of the German research system in general as well as on studies on the history of the AGF in particular.  However, the Alliance of Science Organisations has so far only appeared on the margins of studies dealing with individual science organizations that are members of the Alliance.  For this reason, this paper is based primarily on previously unexamined archival collections on the Alliance and the Presidential Circle.  The aim of this paper is to shed light on the social and institutional level of cooperation in the Alliance of Science Organisations and, in doing so, to examine the interactions of the science organizations within the committee, as well as in relation to the central actors of the Federal Government in the field of science policy. This article argues that the Alliance can be understood as a structurally rather conservative committee which initially often perceived changes as disrupting its established patterns of interaction. In order to restabilize the situation, exclusionary and competitive practices were used by the members of the Alliance (Stucke ; Hohn ). Thus, this article aims to analyze how practices of inclusion and exclusion have been deployed by the stakeholders of the Alliance to meet different expectations and to balance a variety of interests at the same time.
Only this complex interplay of cooperative and competitive practices-so the hypothesis-finally resulted in the institutionalization of the Alliance.

Institutionalization of the Alliance of Science Organisations
For a long time, the Alliance has been perceived as a highly informal organization. The committee has been described by external observers as some sort of "secret gatherings of the Knights of the Roundtable" or as a club "that attempts to pull strings behind the scenes" (van Bebber : ). Although the Alliance has remained a rather informal committee up to this day with no office of its own, no permanent staff or press department and an external appearance that is relatively restrained, there have been signs of a gradual institutionalization of this cooperative committee since the s.

Formalization of Cooperation
The beginnings of the Alliance of Science Organisations date back to the late s, when the West German Rectors' Conference (WRK)  and the German Research Foundation (Deutsche Forschungsgemeinschaft, or DFG) had an intensive exchange on the plans of the federal and state governments to create the German Science and Humanities Council (Wissenschaftsrat, or WR) and tried to influence its design (Bartz : -; Orth : -; Stamm : -; Stucke : -).  At the same time the Federal Government had become active in the funding of research, as indicated by the creation of the Federal Ministry for Nuclear Affairs (Bundesministerium für Atomfragen, BMAt) in  and the transformation of the latter into the Federal Ministry for scientific Research (Bundesministerium für wissenschaftliche Forschung, BMwF) in . The creation of these bodies changed the dynamics in the German science system and should finally lead to the establishment of the Alliance of Science Organisations: In response to efforts of Atomic Minister Siegfried Balke to convince the presidents of the Max Planck Society (MPG), DFG and WRK to support his plan to create a Federal Research Department, these three science organizations moved closer together in order to protect themselves from any influence by the ministry in terms of regulating or controlling the content of research (Orth : -; Stucke : -; Stamm : -).
At a meeting of the Senate, Adolf Butenandt, president of the MPG, explained that one should "not take a position on the question of whether Germany required a special ministry for research individually" but rather seek out cooperation with the DFG and the WRK.  Subsequently, the three science organizations, the MPG, DFG and WRK, continued their meetings from  onwards with the involvement of the WR to discuss "foundational questions about science policy on a regular basis. "  The joint decision to formalize their hitherto irregular meetings marks the birth of the Alliance of Science Organisations and can be understood as a reaction to the expansion of the federal government in the field of science policy (Schimank : -).
The emergence of a new external negotiating partner in the federal government resulted in the first consolidation of this informal group, which had previously met mainly on the occasion of other meetings or at a joint dinner. A look at the frequency of the Alliance's meetings over a longer period of time ( Fig. ) confirms the impression of a slow institutionalization process beginning in the s, as their meetings became increasingly regular during these years. Officially, these meetings were referred to as "Meetings of the Presidents, Chairpersons and General Secretaries of the Science Organizations, "  but the shorter name "Holy Alliance" or "Alliance" quickly became common among its members (Klofat ). The reference to the Holy Alliance founded in  by Russia, Prussia and Austria, which had committed itself to mutual loyalty and cooperation despite all differences in political and confessional orientation was surely not accidental (Fehrenbach : -; Pyta ).

Institutionalization of the Presidential Circle as an Advisory Body on Science Policy
For its part, federal politics reacted to the formation and consolidation of the Alliance and sought to exchange views with the presidents of the science organizations. After taking office, the first Federal Minister for Scientific Research (BMwF), Hans Lenz, contacted the presidents of the science organizations and met with them occasionally.  However, it was his successor, Gerhard Stoltenberg, who intensified the relations between his department and the self-governing organizations with the aim of "regularly discussing the problems of science. "  The establishment of regular consultations between the ministry and the presidents of the Alliance can also be understood as the formalization of their consultative role in processes of political decision-making.  The Alliance benefited from the fact that they had already exchanged views on research policy issues with other ministries, including the Atomic Ministry, even before the BMwF was founded and had thus established the corporatist approach of mediating interests.  To a certain extent, the selfgoverning organizations also benefited from the absence of a pre-defined political program or instruments in matters of research funding, which further encouraged the close coordination between politics and science. Through the joint efforts of the presidents of the self-governing organizations, the Alliance was able to intervene in its political environment and influence the shaping of German research policy (Braun : -; Stucke : -).
Consequently, Stoltenberg strengthened the already existing corporatist structures of German science policy by introducing regular consultations. At the same time, he created a framework for this informal exchange for the first time with the so-called "fireplace rounds" (Kaminrunden), as the meetings of the Presidential Circle were referred to in the s and s (Trischler & vom Bruch : -; Weßels : -). In  Stoltenberg described the meetings with the presidents of the Alliance as an "extensive exchange of opinions, " which he hoped would continue "in casual succession. "  The informal nature of the talks was to remain the defining characteristic of the Presidential Circle for a long time to come (Patzwaldt & Buchholz : ; Schimank : -). In , for example, the staff of the Federal Ministry of Research and Technology (Bundesministerium für Forschung und Technologie, BMFT), which had been created in , noted on behalf of minister Horst Ehmke, who wanted to renew the meetings with the presidents, that the Presidential Circle was an "informal discussion group. "  Ehmke himself emphasized in the invitation to the presidents of the Alliance that he saw the joint talks as an opportunity "to discuss questions of research and technology policy in an intimate setting. "  In , Volker Hauff, then Research Minister, praised the presidents' "great pool of experience, " which would find its way into the "elaboration of research policy. "  Hauff 's statement reveals for the first time a certain commitment-albeit a vaguely formulated one-to the issues the group discussed together. The s were characterized by frequent changes at the top of the BMFT. Despite the fact that most ministers remained in office for no more than two years,  a trusting relationship began to develop between the ministry and the Alliance during the phase of the Social-Liberal Coalition (-). The high level of personnel continuity among the ministry officials who valued the presidents of the science organizations as reliable partners in questions of research policy proved beneficial. A briefing of the new minister, Andreas Bülow, in  illustrates the extent to which ministry officials valued the Alliance. Bülow was not very familiar with the field of research policy and the staff tried with the utmost effort to convince him of the high importance of the Alliance as an "informal advisory body" and to prepare him for the meeting with regard to its contents.  Two years later, on the occasion of another change at the top of the Research Ministry, the importance of the meetings with the presidents of the Alliance was outlined even more precisely: The Presidential Circle was now characterized as an "advisory committee of the BMFT that met roughly two times each year. "  In addition, the incoming minister Heinz Riesenhuber assured the presidents of the Alliance during a joint meeting of their role as "top councilors, " through whom "the essential views of the science organizations would be realized in research policy. "  Riesenhuber, a chemist and proven expert on research policy, would remain in office for more than ten years and had great interest in strengthening the ministry's relationship to the presidents of the Alliance. The contacts between the ministry and the Alliance became closer in the second half of the s and the early s. This resulted in an increasing regularity of their meetings (Fig. ), which accelerated the institutionalization of the Presidential Circle in the following years. In addition to the increasing regularity of the meetings, one can find further signs of an institutionalization beginning in the mid-s: For example, the BMFT started to send out official invitations with a preset agenda for each meeting. The participants also came carefully prepared for the individual items to be discussed in these meetings. In the ministry, the respective (sub-)heads of the departments prepared detailed consultation notes for the Federal Minister of Research on the planned topics in advance. In the Alliance, the preliminary discussion of the Presidential Circle and the coordination of a common line towards the ministry became a fixed part of the internal agenda.  In comparison to the Alliance of Science Organisations, the meetings of the Presidential Circle were of a more official nature. In fact, the Presidential Circle can be considered a gathering of some of the most important stakeholders from the two areas of science and politics. Therefore, the formalization of cooperation and internal workflows on both sides began earlier while also proceeding more quickly. By the early s, the perception of the Presidential Circle-and thus also of the Alliance-had changed massively. Originally conceived as a casual discussion forum, it developed into an institutionalized, in-house advisory body of the BMFT, whose ad-vice found its way into the ministry's research policy programs as a matter of course.
However, it is difficult to measure the influence of the Alliance in terms of specific successes or cases in which the presidents put their mark on decisions regarding science policy. Instead, the joint discussions were characterized by an open and trusting exchange of ideas: The minister usually gave a short overview of his plans and sought the science organizations' advice. In this way, the Alliance was able to contribute its ideas and priorities before specific political programs were elaborated.  The BMFT trusted the presidents of the science organizations in matters of science policy and regularly consulted them, for example, when developing general funding priorities for the next few years.  Therefore, trust is a key term to describe the relationship between science and politics as represented in the Presidential Circle. The minister trusted the presidents of the Alliance to provide him with the necessary (and valid) information for his subsequent political actions, while the science organizations for their part trusted that the minister would act in their interest (Luhmann ; Frevert ). Furthermore, the ministry did not hesitate to ask the presidents to get in touch with other politicians or the public in order to support their common goals. Thus, one can conclude that the relationship between the ministry and the science organizations can be characterized as having been highly cooperative, although the Alliance was also dependent on the ministry for its funding.
Through their regular consultations in the Presidential Circle the Alliance had privileged access to the ministry. Moreover, the BMFT incorporated the Alliance's proposals into the design of its research policy. This inclusion of the expertise of the Alliance's members in political decisionmaking processes further enhanced the reputation of the science organizations that comprised the Alliance and thus contributed to consolidating their position at the top of the German science system.

The Composition of the Alliance: Inclusion and Exclusion Using the Example of the Association of the Major Research Centers (AGF)
With the formalization of cooperation within the Alliance, its gradual institutionalization and recognition as a top advisory body by the Federal Ministry of Research, the committee was also given a higher profile in the political and scientific communities (Hohn & Schimank : -). This not only increased interest in its work, but also brought its compo-sition into the focus of various actors. Questions of membership became a central issue of scientific prestige and power in research politics.
This resulted-sometimes through indirect pressure from outside, as will be shown below-in the inclusion of previously excluded science organizations. At the same time, however, the Alliance tried to keep the circle of its members as small as possible in order to preserve its own ability to act. This necessitates a tension-filled simultaneity of opening and closing tendencies, which will now be examined in more detail using the example of the relationship between the AGF and the Alliance of Science Organisations.

Context: The Association of the Major Research Centers (AGF)
In , the ten major research centers existing in the Federal Republic of Germany founded the Association of the Major Research Centers (Arbeitsgemeinschaft der Großforschungseinrichtungen, AGF) (Hoffmann & Trischler : -; Szöllösi-Janze b: -).  The foundation of the AGF thus coincided with a phase of increasing resource scarcity resulting from the economic crisis which began in the s. Although the concrete effects may have varied from case to case, financial growth slowed down noticeably and, in some cases, even stagnated completely (Hohn & Schimank : -). The experience of the economic crisis and budget tightening fostered more competitive behavior among the science organizations, especially when dealing with a newly formed stakeholder, whose budget from federal funds was many times higher than that of the other Alliance members. In these tense times, the major research centers began to consolidate their identity as a separate type of non-university research, which the other science organizations observed with a critical eye. However, over the subsequent years the AGF was not able to fully realize its political ambitions: Until the s, the chairman only had an office with minimal staff which he administered on a part-time basis; the individual research centers continued to remain legally independent. Thus, the powers of the chairman of the AGF in internal matters were initially not particularly far-reaching (Hoffmann & Trischler : -). This limited capacity for action was of course apparent to his colleagues in the Alliance and so they mockingly asked whether the AGF's chairman must first seek permission from his association before joining the Alliance. 

Inclusion Through Politics: the Admission of the AGF into the Presidential Circle
In , two years after the division of the department for education and science (BMBW) from the one for research and technology (BMFT), the Minister of Research Horst Ehmke decided to intensify the relations be-tween the science organizations and his department. Furthermore, he also enlarged the group of participants and invited the chairman of the AGF to join (Ritter et al. ; Szöllösi-Janze a; Szöllösi-Janze & Trischler ; Hoffmann & Trischler ).  The members of the Alliance initially showed reservations towards the newest negotiating partner within the Presidential Circle. Nevertheless, the ministry had no direct influence on the composition of the Alliance and on whom the science organizations invited to their internal meetings, even though the integration of the AGF into the Alliance was definitely in the ministry's interest (Szöllösi-Janze b: -, -). One year later, federal politics confirmed its position towards the AGF and thus strengthened its position within the German science system: Since the establishment of the WR in , only the WRK, the DFG and the MPG had been entitled to send the Federal President a list of suggestions for appointments to this council. In , following the amendment of the administrative agreement, the AGF was also granted the right to submit suggestions for appointments to the WR (Bartz : -; Röhl : ; Rusinek : ).

The AGF and the Alliance: Reluctant Inclusion
This amendment now required a vote between the Alliance and the AGF regarding the nominations for the WR. The organizations entitled to suggest appointments needed to draw up a joint list of people suited to fill the vacant positions to be submitted to the Federal President (Röhl : -).
At first, the members of the Alliance and the AGF discussed the appointment lists prepared in the respective subcommittees separately. Within the Alliance, the presidents of the WRK, DFG and MPG finally agreed on a common list of names, while the DFG took over the task of coordinating this list with the AGF.  This procedure proved to be impractical and was therefore discussed at a meeting of the Alliance in : MPG president Reimar Lüst remarked with a certain displeasure that the Alliance would have to consult with Karl Heinz Beckurts, chairman of the AGF, on the nominations for the Scientific Commission of the WR.  His phrasing indicated that the coordination of the list of proposals with the AGF was initially met with disapproval because this process prolonged the procedure of appointing new members to the WR.
Consequently, DFG president Heinz Maier-Leibnitz suggested inviting Beckurts as a guest to their joint meetings.  However, the invitation of the AGF's chairman to the Alliance's meetings did not come about as a result of internal conviction or even recognition of the growing importance of the AGF. Rather, this decision was shaped by a certain necessity and pragmatism that aimed to avoid unnecessary and additional work. It is also remarkable that Beckurts was initially not intended to become a member of the Alliance. While the other participants were addressed as representatives of their respective organizations, the members of the Alliance insisted on inviting Karl Heinz Beckurts explicitly as a guest and not as an official representative of the AGF. The fact that the Alliance nevertheless decided to invite Beckurts to the joint meetings was probably helped along by the high scientific reputation he enjoyed as a very successful nuclear physicist. He was furthermore the chairman of the KfA Jülich and was thus certainly one of the most influential science managers within the AGF (Rusinek : -, -). Consequently, Beckurts benefited from his personal and professional authority, since the guest status that the Alliance granted him was directly linked to his person (Bourdieu : -). However, this did not change the Alliance's concerns about including the AGF as an organization. Their reservations were fueled by different factors: The members of the Alliance considered the work carried out in the major research centers to be to a certain extent research commissioned by the state. Hence, they feared that the inclusion of the AGF would undermine their claim of autonomy. Furthermore, the Alliance was afraid of a growing competition in matters of association policy due to the fact that the major research centers had founded the AGF and thus started acting cooperatively. This in turn, put the carefully balanced distribution of power between the science organizations and the division of their fields of activity to the test. At the same time, the major research centers remained legally independent. That is why the AGF was regarded as an association with very little authority. Consequently, the question was raised as to whether the AGF would be able to act as a negotiating partner of equal status (Szöllösi-Janze b: -, -). The AGF nonetheless interpreted the invitation as a first step towards its official recognition by and inclusion in the Alliance. Membership in the Alliance gave the AGF the opportunity to participate in the processes of consensually influencing political decisions. The committee was perceived as a forum that allowed the science organizations to lend their interpretations a sense of validity which went beyond the sensitivities of a single institution and therefore was accepted by the political authorities and taken into considerations. Herwig Schopper, chairman of the AGF beginning in , announced at a general meeting that "the conversations in the so-called "Alliance" have flourished" (Szöllösi-Janze b: ).  Indeed, during Schopper's term of office, not only the "coordination with other science organizations, such as the Alliance" was intensified, but also "bilateral actions" were started, for example in the form of a joint recommendation with the WRK.  Internally, the AGF had the feeling that they had succeeded in further improving their association. 

Exclusion Despite Opening
Although the Alliance had slowly opened up to the AGF since  due to indirect external pressure from the Federal Ministry of Research, this transformation took place neither in a tension-free environment nor in a linear manner.
Even in the s, more than four years after the first invitation of its chairman to the exclusive meetings of the Alliance, the AGF continued to see itself being pushed into an outsider position. Thus, Gisbert zu Putlitz, the new chairman of the AGF, noted in an internal memo regarding the nomination of new members for the WR in : The Federal President receives information about suggestions for the coming appointments to the German Council of Science and Humanities. The documents used by Mr. Lüst, Mr. Turner and Mr. Seibold are not in my possession.  Zu Putlitz therefore experienced a similar dynamic to the one Herwig Schopper had faced a few years earlier: While the relationship between the AGF and the Alliance seemed to have stabilized under his predecessor-especially towards the end of his term of office-the situation suddenly changed again when the office was handed over. It becomes clear that the reputation of the office-holder among the established science organizations was directly linked to the respective person and was not automatically passed on to his successor (Szöllösi-Janze b: -). Instead, the successor remained highly dependent on his own personal and professional prestige being recognized by the Alliance. Especially at the beginning of his term of office, he was initially confronted with their reservations about the organization he was chairing. In zu Putlitz's case, for example, this was expressed by the fact that he did not receive circulating documents in the run-up to joint meetings. Moreover, the founding members of the Alliance did not hesitate to express their considerable reservations in the joint meetings.  However, this rejection was not only expressed in a subtle way in the internal circle. On the contrary, his exclusion was also openly displayed. For example, zu Putlitz reported the following from a meeting of the Presidential Circle when the budget situation of the Federal Ministry of Research and funding cuts were discussed: The BMFT reports that the budget for personnel and expenses at the major research centers will be reduced [. . . ] Lüst und Seibold used this opportunity to unleash complaints about the quality of the major research centers.  The situation described is remarkable in that the presidents of the MPG and DFG did not hold back their criticism of the AGF and its work in the presence of the Federal Minister of Research and his state secretaries, as well as in the presence of the president of the AGF. The question of the distribution of financial resources was still incendiary in the early s and had a destabilizing effect on the fragile cooperation with the AGF. In times of general funding shortages, the MPG and DFG perceived the major research centers as competitors for state resources and feared losses for their own organizations. Although the AGF had already been officially accepted into the circle of the so-called Holy Alliance at this point in time, its founding members made the newcomer aware of the limits of acceptance and its position as an outsider. Completely excluding the AGF from the joint negotiations was, however, not a viable alternative since the Alliance did not want to offend the Ministry of Research.

Gradual Acceptance
The other science organizations perceived the rise of the AGF and its establishment as a new pillar in the German research system as a threat to their own position. Another partner that brought its own wishes and ideas into the negotiations threatened to reduce the influence of the long-standing members of the Alliance. These dynamics had also become apparent to the members in the Alliance-with the MPG taking the lead in this respect-and to the AGF. In order to defuse the situation, Reimar Lüst, president of the MPG, finally invited his colleague zu Putlitz, chairman of the AGF, to a discussion in January : At the outset of this conversation, Mr. Lüst expressed his concerns and fears that a sort of competition might arise between the major research centers and the Max Planck Society that would be mutually unbeneficial. The special status of the Max Planck Society must be recognized.  Fears that the MPG's own position-whether in financial or scientific terms or in matters of science politics-could be threatened by the AGF seemed to influence the MPG's behavior. However, the AGF could not be completely ignored, especially given the interest of the Federal Ministry of Research in their involvement, which is why the MPG finally opted for a joint consultation. Direct competition between the MPG and the AGF was now apparently regarded as a problem. The MPG perhaps hoped it would be easier to reach its goal in close cooperation with the AGF, especially when dealing with the ministry.
Zu Putlitz for his part was well aware of the importance of a cooperative relationship to the MPG as one of the leading institutions within the Alliance and tried to dispel the concerns immediately by emphasizing that he saw "no conflict at all" concerning "the special status and special rights of the Max Planck Society. "  Instead, he emphasized that they shared a common goal, which could only be achieved together.  Whether this conversation was the trigger for the stronger integration of the AGF into the Alliance cannot be said with certainty. However, from about  onwards, various episodes of inclusion can be observed with regard to the AGF.
A look at the chairs at the meetings of the Alliance of Science Organisations (Fig. ) shows that in the s the AGF only chaired one single meeting, whereas in the s it chaired one meeting roughly every two years. Compared to the total number of annual meetings, this at first glance may not seem like convincing evidence for inclusion.
However, if we compare the data with the chairs of the WR, one of the founding members of the Alliance, a different picture emerges. The first feature that stands out is the discrepancy in the s. During this period, the WR organized at least one, sometimes even two or three meetings a year. In the s, the bars gradually even out.
Since the WR chaired  meetings of the Alliance between  and ,  this development shows that the AGF was able to establish itself within the Alliance-at least in regard to the chairmanship (Fig. ). By taking over the chairmanship, the AGF was therefore in a position to set the main focus of the meeting and to significantly shape the agenda in a way that went beyond mere participation.
In the following years, the AGF was also increasingly involved in the elaboration of joint statements, for example in a  statement concerning deregulation in non-university research.  Even the MPG, which did not hold back its criticism of the AGF at the beginning of the s, showed itself to be increasingly cooperative: When the members of the Alliance discussed a change in tax law that AGF and FhG wanted to initiate together, the MPG announced that it would refrain from objecting to the plans even though it did not correspond with its interests. In return for this favor, the MPG demanded that the other members of the Alliance sign the MPG's declaration on collaborative research.  The atmosphere in the Alliance had obviously changed: Now the founding members had granted the AGF the right to act as an equal partner. A further indication of the AGF's inclusion into the Alliance during this period is its leadership in the preparation of joint statements. To cite just one prominent example, the AGF coordinated joint actions and public statements of the Alliance regarding animal protection in the early s.  While each of the science organizations had their own priorities, the founding members entrusted the AGF with managing the statement. This close involvement in the workings of the Alliance shows that the AGF was finally perceived as an equal-ranking member of the Alliance. After all, the statements published by the Alliance were public actions, and thus reached a select audience in science and politics.

Conclusion
Examining the practices of inclusion and exclusion allows us to better understand the internal dynamics of the Alliance and trace the constant balancing act between cooperation and competition. This also enables us to draw conclusions regarding the social dimension of cooperation between the science organizations in the Alliance. In this view, we can observe the simultaneity of modes of action that may appear contradictory at first glance: practices of inclusion and exclusion coexist, as do modes of cooperation and competition. Since the Alliance is an informal committee that brings together the most important science and research organizations in Germany, it may initially appear as being solely designed for cooperation and thus for including various players. This cooperation, however, also requires an exclusionary element with respect to other institutions that are not represented in the Alliance and, consequently, with respect to their interpretations and perspectives. Investigating the history of the Alliance in the s and s also showed that the internal relationships between the members were by no means exclusively cooperative; questions of hierarchy and competition for resources shaped their collaboration in many ways.
The formalization of cooperation between the science organizations in the Alliance by fixing a meeting schedule and carefully preparing joint meetings came as a reaction to the growing attempts of the federal government to shape science policy since the s. Since its foundation, the Federal Ministry of Research has tried to maintain a close exchange with the presidents of the science organizations assembled in the Alliance. The ministry's aim was to consult with the science organizations on important issues in order to ensure the legitimacy of its decisions regarding research policy in the eyes of a wider public. This also meant that the Alliance was officially accepted by the federal government as being a reliable committee of experts, which the minister and his officials trusted in questions of research policy. Moreover, the politicians also hoped that the decisions achieved in agreement with the Alliance would be implemented internally by the science organizations. The relationship between the participants in the Presidential Circle can be described as an exchange of ideas in an atmosphere of great trust. These meetings offered the Alliance the opportunity to actively shape German science policy together with the ministry. However, the nature of these discussions changed in the course of the s, when the Alliance became explicitly recognized as the BMFT's advisory body. Consequently, the institutionalization of the cooperative consultations resulted in the active inclusion of scientific expertise from the political side.
At the same time, changing constellations in the German science system made it necessary to admit new players to the Alliance and thus resulted in the opening of this committee. Probably the most tension-filled inclusion of a science organization in the Alliance was its first expansion: The admission of the AGF in . As an independent player in science policy, the AGF was a comparative newcomer who was invited to the Presidential Circle by the Federal Ministry of Research. The inclusion of the AGF into the joint negotiations took place in a time of scarce resources. The uncertainty about the allocation of financial resources and about the role of the AGF in German science politics resulted in different elements of exclusion.
The initiative of federal politics to grant the AGF the right of nomination for members of the WR and its admission to the Presidential Circle forced the Alliance to pursue closer coordination with the AGF. This finally led to the invitation of the AGF's chairman to the meetings of the Alliance. In theory, the members of the Alliance themselves decided on the composition of their committee. However, the Alliance could not completely ignore the demands of external parties. In the following decades, further steps towards the inclusion of the AGF took place. Among other changes, the AGF began to assume leadership of meetings and coordinate cooperative actions with other members.
Nonetheless, this process of inclusion was by no means a linear development. This is hardly surprising, since the inclusion of the AGF disrupted the well-established cooperation of the hitherto unchallenged four large science organizations. While mostly external actors-such as the Federal Ministry-suggested including certain organizations in the Alliance, its members at times acted to exclude newcomers and limit their influence. After all, accepting a new partner in the Alliance implied an additional perspective that had to be included in the negotiation process. In turn, new members threatened to weaken the positions of the four founding members. This initially had a destabilizing effect on the carefully balanced relationship of power in the Alliance (Bourdieu : -). The more or less subtle attempts of a partial exclusion of the AGF on different levels can therefore be understood as an attempt to stabilize the prevailing distribution of power in the German science system. The Alliance's reaction to the change in its internal dynamics spurred by the inclusion of the AGF was more than a mere demonstration of power for its own sake: Membership in the Alliance also implied the possibility of shaping German science pol-icy in a corporatist way. As a result, the status of new members remained dependent on the established science organizations of the Alliance; their assessments of whom they considered capable of participating in this circle and, consequently, of supporting their common interests in a cooperative manner proved decisive (Goffman : -).
The moments of exclusion in certain areas of cooperation which the AGF had to face should be interpreted with regard to their symbolic dimension. By including and excluding other science organizations, the Alliance began to increasingly define itself, its mission and its elite membership; these developments finally led to the institutionalization of the committee. The analysis of the internal relationships within the Alliance has shown that inclusion at one level could trigger exclusion at another level, sometimes involving the same players. Cooperation can therefore be understood as an unstable mode of interaction from which actors may withdraw at any time (Nickelsen : -).

Acknowledgements
I would like to thank the anonymous reviewers and the editorial staff of NTM for their helpful suggestions on the first version of this essay. I am especially grateful to Helmuth Trischler for his help and constructive suggestions during the revision of the manuscript. Furthermore, I also thank the members of the DFG Research Group "Cooperation and Competition in the Sciences" (FOR ), who discussed an early draft of the text with me, and the members of the Research Program "History of the Max Planck Society" (Max Planck Institute for the History of Science, Berlin). Finally, I wish to thank Anna Lehner, Ramona Pohlmann and Fabienne Will for their valuable comments and suggestions and Moritz Schlenker for his support during the research process.
Funding Open Access funding enabled and organized by Projekt DEAL.
Open Access This article is licensed under a Creative Commons Attribution . International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/./.

Competitive Cooperation
Artikel/Articles Endnotes  For example, in the form of commissions of enquiry on issues such as future nuclear energy policy (-) or on the opportunities and risks of genetic engineering (-).  For example, the results of the working group "In Vitro Fertilisation, Genome Analysis and Gene Therapy" (-) convened by the Federal Minister of Justice and the Federal Minister of Research and Technology were incorporated into the drafting of the Embryo Protection Law (Bogner : ).  Due to its informal qualities, it is difficult to define the Alliance of Science Organisations in Germany and its work conceptually. When speaking of the Alliance's internal organization and of the way in which the members cooperate with each other the term "committee" is sometimes used. The term "advisory body" is also used to describe its relationship to politics. Both terms were coined by the actors themselves during the time span on which this paper focuses and therefore seem promising for characterising the Alliance in more detail. For the use of the German term Gremium, which can be translated with the English term "committee" see e.