Microsoft Word - Volume37_final insna.org | Issues 1&2 | Volume 37 | 7 Networks and Institutionalization: A Neo-structural Approach Emmanuel Lazega Institut d’Etudes Politiques de Paris, France Centre de Sociologie des Organisations – CNRS EUSN 2017 Mainz keynote address1 for Connections Revised 10 Dec 2017 Abstract This paper is the text prepared for the keynote address of the EUSN 2017 conference in Mainz, Germany. A short presentation of concepts reflects in part the foundations of neo- structural sociology (NSS) and its use of social and organisational network analyses, com- bined with other methodologies, to better understand the roles of structure and culture in indi- vidual and collective agency. The presentation shows how NSS accounts for institutional change by focusing on the importance of combined relational infrastructures and rhetorics. Specific characteristics of institutional entrepreneurs who punch above their weight in institu- tionalization processes are introduced for that purpose, particularly the importance of multi- status oligarchs, status heterogeneity, high-status inconsistencies, collegial oligarchies, con- flicts of interests and rhetorics of relative/false sacrifice. Two empirical examples illustrate this approach. The first case focuses on a network study of the Commercial Court of Paris, a 450-year-old judicial institution. The second case focuses on a network study of a field- configuring event (the so-called Venice Forum) lobbying for the emergence of a new Europe- an jurisdiction, the Unified Patent Court, and its attempt to create a common intellectual prop- erty regime for the continent. For sociologists, both examples involve “studying up”: they are cases of public/private joint regulation of markets bringing together these ingredients of insti- tutionalization. The conclusion suggests future lines of research that NSS opens for the study of institutionalization, in particular using the dynamics of multi-level networks. One of the main issues raised by this approach is its contribution to the study of democratic deficits in a period of intense institutional change in Europe. Author Emmanuel Lazega is Professor of sociology at the Institut d’Etudes Politiques de Paris and a membre of the Centre de Sociologie des Organisations (CNRS). His current research projects focus on the dynamics of multi-level networks in organizations and markets, with a special focus on network modelling of social processes helping actors in such settings manage the dilemmas of their collective action (contemporary forms of solidarity, social control, learning and regulation). He is the author and co-editor of several books (for example The Collegial Phenomenon, Oxford University Press; Micropolitics of Knowledge, Aldine-de Gruyter; Conventions and Structures in Economic Organization, edited with Olivier Favereau, Edward Elgar Publishing; Multilevel Network Analysis for the Social Sciences, edited with Tom A.B. Snijders, Springer; Knowledge and Networks, edited with Johannes Glückler and Ingmar Hammer). His publications can be downloaded from http://www.elazega.fr. 1 I would like to thank the committees of EUSN 2017, particularly Marina Hennig, for the invitation to present this keynote address. Connections Networks and Institutionalization 8 | Volume 37 | Issues 1&2 | insna.org Introduction The outline of this address is the following. It will first briefly present a few concepts that belong to the foundations of neo- structural sociology (NSS) (for a synthesis, see Lazega, 2006, 2012a). The purpose of this sociology is to use social and organi- zational network analyses, combined with other methodologies, to better understand the roles of structure and culture in the processes of collective agency. It will then look at how NSS contributes to under- standing institutional change as collective agency by focusing on the importance of relational infrastructures and rhetorics. It will next identify specific characteristics of institutional entrepreneurs who punch above their weight in institutionalization processes: the importance of high-status inconsistencies, collegial oligarchies, mul- ti-status oligarchs, conflicts of interests and rhetorics is hereby introduced. Two empir- ical examples will be provided to illustrate this approach. For the sociologist, both ex- amples involve ‘studying up’: they are cas- es of public/private joint regulation of markets looking at these elements of insti- tutionalization. The first case focuses on the Commercial Court of Paris, a 450-year- old judicial institution. The second case focuses on the emerging European Unified Patent Court. The conclusion suggests fu- ture lines of research that NSS opens up based on the study of institutionalization, in particular in the dynamics of multi-level networks. One of the main issues raised by this approach is that of democratic deficits in a period of intense institutional change in Europe. Foundations of neo-structural sociology Neo-structural sociology brings together, very generally, structure, culture and ac- tion (both individual and collective action) in the organizational society. It relies on sophisticated analyses of socio- organizational networks, if possible multi- level and dynamic, and enriched with data on culture and behavior2. The focus is on collective action, and especially on model- ling its generic social processes (learning and socialization, particularistic solidarity and discriminations, social control and conflict resolution, regulation and institu- tionalization). To theorize and model these processes, it is important to use the notion of relational infrastructures of collective action, particularly of collective action that relies on personalized ties between peers for coordination, elsewhere called collegi- ality3. In order to frame the issue theoreti- cally, relationships, as indicators of inter- dependencies, are defined as both channels for resources with, and symbolic or moral commitments to, exchange partners. It is because relationships have both dimen- sions that they can become infrastructural. Indeed, stabilized relationships are usually part of structures that are identified locally, in the entourage of the actors, as relational sub-structures with which we are all famil- iar, such as direct reciprocity, indirect reci- procity, transitivity, etc.; but also part of what can be called relational infrastruc- tures, such as vertical patterns of differen- tiations (mainly heterogeneous forms of status coexisting in complex ways) and horizontal patterns of differentiations (mainly social niches in a system of nich- es) at a morphological level. In particular, relational infrastructures combined with norms facilitate generic social processes helping members manage the dilemmas of their collective actions. Using combina- 2 Contemporary neo-structuralism is different from the structuralism of Claude Lévi- Strauss and of the 1960s mainly because it relies on a theory of individual and collective action to provide a new fundamental link between structure and process. The articulation between culture and structure, however, was already present in the ‘old’ structuralism. 3 This also means that understanding collec- tive action in ideal, typically speaking bu- reaucratized settings, based on routines, hie- rarchy and impersonal ties, does not need much social network analysis. Networks and Institutionalization Connections insna.org | Issues 1&2 | Volume 37 | 9 tions of relational infrastructures and rela- tional sub-structures in social and organi- zational networks, we have modelled sev- eral generic processes4, such as solidarities with direct and indirect reciprocity in vari- ous multiplex networks; social control in lateral control regimes using personal rela- tionships for collective goals; collective learning with various kinds of advice seek- ing networks; and regulation and institu- tionalization as political processes, that are the focus of this talk. Thus, NSS reframes the recursive dynamics of social structure and social processes. A focus on relational infrastructures in institutionalization processes Institutions are commonly and broadly de- fined as rules, norms, and beliefs that de- scribe reality for actors, explaining what is and is not, what can be acted upon and what cannot, and how (Hall & Taylor, 1996; Hoffman, 1999; Scott, 2007; Bathelt & Glückler, 2014). The focus is on the norms, mutual expectations and beliefs about issues and the legitimate organiza- tional ways in which related behavior has to be normed and governed. A classical question in the social sciences is: ‘How do such institutions emerge?’ Our work in this area owes much to collaboration, initiated by Harrison White in 1995, with institu- tional economists, in particular Olivier Fa- vereau (see Favereau and Lazega, 2002; Lazega 2016a). In the era of ‘governance’ (weakening of command and control framework of States) that we witnessed over the last decades, this neo-structural sociology has blossomed in economic so- ciology, by focusing on collective action in joint (public authorities and private actors, including business) regulation of markets. Since 1996, NSS has owed much to Julien 4 Neo-structural publications about these pro- cesses including theoretical, empirical and methodological work developing a neo- structural theory of institutionalization are available online: www.elazega.fr > http://elazega.fr/?m=201709 Brailly, Catherine Comet, Fabien Eloire, Guillaume Favre, Lise Mounier, Jaime Montes-Lihn, Mohamed Oubenal, Elise Penalva-Icher, Alvaro Pina-Stranger, Tom Snijders, Paola Tubaro, Marta Varanda, and many others. Contemporary thinking about the emergence of institutions is dominated in sociology by a variety of neo-institutional perspectives focusing on ‘institutional en- trepreneurs’ (DiMaggio, 1988) who elabo- rate on taken-for-granted cultural catego- ries, classifications, rules and procedures that include beliefs and codes stabilizing action into routines. In this literature, struc- tural dimensions of regulation are largely ignored. Yet in his work on what he calls ‘precarious values’, Selznick (1957) al- ready provides an early combination of a structural and an institutional perspective in sociology. A precarious value is one that is essential to the viability of the collectivi- ty but in which most members may have no direct stake. In this illustration of the entanglement of structure and culture, a value is therefore precarious because it is always in danger of losing its flag carriers and representatives, that is, the active sup- port by organized interest groups and elites that helps preserve it as a candidate for priority on the list of all competing values. We argue that, in organizational so- cieties (Perrow, 1991), where power is ex- tremely concentrated, this relational and cultural perspective enriches the study of institutionalization processes by focusing on their often collegial, elitist and person- alized nature. In this process, the selection of priority norms and the personalized se- lection of the authorities who champion them is an important process in the crea- tion of frames of reference that become taken for granted over time, and thus insti- tutionalized. Therefore, the creation of an oligarchy of actors, who are able to guide the regulatory process and to mobilize fol- lowers by helping them align with new rules, are key underlying mechanisms that belong to the institutionalization process as theorized by Selznick and encapsulated in his notion of precarious values. Connections Networks and Institutionalization 10 | Volume 37 | Issues 1&2 | insna.org Today the focus is on the process of regulation and institutionalization. There is something remarkable in the way relational infrastructures are mobilized in political, regulatory, institution building processes: their efficiency comes from the constitu- tion of collegial oligarchies whose mem- bers take advantage of situations of con- flicts of interests to concentrate power (Lazega, 2012b). High-status inconsistencies and conflicts of interests: Punching above their weight in institutionalization process We are not equal in our capacity to defend our regulatory interests, even in egalitarian systems. Very generally speaking, neo- structural sociology starts with the princi- ple that divergent interests and forced in- terdependencies complexify the regulatory process and development of institutions. It requires an understanding of interdepend- encies as relational infrastructures that are often conflictual, dynamic and multi-level. Modelling social processes in terms of networks adds much to the reflection on the way individual and collective agents defend their regulatory interests in institu- tionalization processes, and therefore to joint regulation as a social and political process. Selznick’s connection between structure and culture is still illuminating today with respect to understanding institu- tion building by small networks of top- level institutional entrepreneurs that we call collegial oligarchies. At a very high level of generality, defining norms for col- lective action, i.e. the political process in a collectivity, depends upon who are the ac- tors promoting these rules, what are their strategies to carry out this regulation with- in the system of their interdependencies, and what are the relational infrastructures that they are able to create and mobilize to do that. In this spirit, neo-structural sociol- ogy looks at institution building by explor- ing the relational dimension of ‘institution- al work’ (Lawrence, Suddaby and Leca, 2011) or political work (Lahille, 2015; Christopoulos and Ingold, 2015), i.e. at the system of interdependencies in collegial oligarchies of institutional entrepreneurs involved in institutional framing. In particular the neo-structural ap- proach to regulatory activity reveals the ways in which strategic agents politicise their exchanges, especially by building re- lational infrastructures, for example, social niches or forms of social status. Control- ling those relational infrastructures gives them a structural position that enables them to frame or guide the negotiation of rules, including building forms of consen- sus and normative alignments that are more or less long lasting. Observing the inconsistencies between these forms of sta- tus proves to be a powerful tool in the analysis of regulatory process5. The notion that different, heterogeneous dimensions of status could be socially inconsistent has a long history in sociology (Hughes, 1945; Lenski, 1954)6. It helps with substantiating the complex link between political work and position within the structure because political actors, as institutional entrepre- neurs, try to both accumulate power and increase their legitimacy. Indeed, in the po- litical process, it is not enough to simply assert that the strongest impose their own rules. Rather, neo-structural analyses show that it is often improbable agents occupy- ing heterogeneous and inconsistent dimen- sions of social statuses, i.e. in a position of ‘conflict of interest’, who have the greatest influence in the political work of institu- tional transformation, of defining Selz- nick’s ‘precarious values’ as priority rules. 5 These inconsistencies are also the reason for difficulties in properly specifying status in statistical network models (Lazega, Mou- nier, Snijders and Tubaro, 2012). 6 “Coexistence of a number of parallel verti- cal hierarchies which are usually imper- fectly correlated with each other. (…) Certain units may be consistently high or consistenly low, while others may com- bine high standing with respect to certain status variables with low status with respect to other variables” (Lenski, 1954:405). Networks and Institutionalization Connections insna.org | Issues 1&2 | Volume 37 | 11 Examples of how relational infrastructure, particularly complex and inconsistent forms of status, matter in institutionaliza- tion of norms in micro- or macro-political controversies (Lazega, 2001: chapter 8) abound. Analyses based on this approach show, for example, how financial indus- tries play a specific role of ‘discrete regu- lators’ (Huault & Richard, 2012). The study of regulation networks shared be- tween public bodies and private agents shows how proactive and how capable of guiding regulatory work the latter are. The financial sector is not the only powerful agent in terms of institutional work but its traditionally dual (economic and political) character gives it a specific role in many regulatory processes. This is precisely be- cause of its capacity to benefit from status heterogeneity, high-status inconsistencies and conflicts of interest. This allows it to become – in part – its own regulator, and to dominate and capture, for example, ju- dicial institutions. Structure, culture and broken promises: Combining high-status inconsistency with rhetorics of sacrifice Why are members with several, heteroge- neous, high and inconsistent forms of sta- tus – multi-status oligarchs – particularly efficient as institutional entrepreneurs when operating in collegial oligarchies and using conflicts of interests? Neo-structural sociology has long contributed to research in this area (compared, for example, with Owen-Smith & Powell, 2008; Glückler, Suddaby & Lenz, forthcoming) not only by framing institutionalization in these terms but by explaining the efficiency of high- status inconsistency. This efficiency relies on the agents’ ability to lose status on one dimension and use a rhetoric of sacrifice to ‘manage the losers’ of the institutionaliza- tion process (Lazega, 2001) – i.e. to man- age actors who stand to lose out from changes in the rules. These multi-status ol- igarchs are not just boundary spanners: they are improbable linchpins with high probability of being either sidelined or punching above their weight. They have to choose, for example, between broadening their constituency (with a view to the long- term stability of the institution) and estab- lishing oligarchic closure. Indeed, since actors organize their collective action around projects and rules, changing the rules is equivalent to break- ing promises made to these actors. Using a position of high-status inconsistency is ef- ficient in terms of institution building when actors are able to combine a form of power (control of resources that others need, i.e. finance, expertise, technique, time, law, etc.) with a form of legitimacy (discourse on behalf of the collective about the value of the new norms that is consid- ered credible and compatible with its over- all project). This credibility is increased when change of rules is presented as a cause of loss of status for the institutional entrepreneurs themselves. This loss of sta- tus is often presented as personal ‘sacri- fice’ of status for the common good. But this loss is very relative, if not false, when this ‘sacrifice’ jeopardizes one dimension of status without jeopardizing the other high and uncorrelated dimension of status. Combined with other factors, this justifica- tion of broken promises is thus more likely to obtain normative alignment from the losers of the process – those who had pre- viously organized themselves around the former rules – without forcing the entre- preneurs to lose all forms of status. Losing out on one dimension of status (while still keeping the other dimension) is thus equivalent to rhetorical creation/purchase of much needed legitimacy. Electoral poli- tics rely very often on this rhetoric of ‘loss’ of status that institutional entrepreneurs claim to accept for the general interest. Ex- amples of such ‘sacrifices’ are provided in the two illustrations below. Collegial oligarchy, weak culture and normative alignments The strength of these actors (their Weberi- an Herrschaft) is increased when they are able to join forces in what we call a colle- Connections Networks and Institutionalization 12 | Volume 37 | Issues 1&2 | insna.org gial oligarchy. Institutional entrepreneur- ship is strengthened in a regulatory college bringing together several persons with high-status inconsistency and conflicts of interests. This helps to draw a wide variety of constituencies into the process. Togeth- er, multi-status oligarchs become both ver- tical and horizontal linchpins in multiple, potentially conflicting domains and levels of regulation. Together they are able to ne- gotiate, select and stabilize not only the formulation of new norms, but also the conventions and interpretations of these priority norms (Favereau and Lazega, 2002). The main reason for which the creation of a collegial oligarchy is efficient in bringing together structure, culture and collective agency is that a group of hetero- geneous leaders, even fraught with initial disagreements, diverse constituencies, or antipathies, can evolve over time. Years, often decades, of common and discrete collaboration create proximities, personal- ized relationships and fragile structural equilibria where mutual critique decreases over time. Being in a collegial regime of personalized relationships reduces the ca- pacity to challenge others’ normative choices, to disagree. Members use their personalized relationships because they fa- cilitate discussion over time – even when these are not necessarily quiet relation- ships. The important feature of this colle- gial and oligarchic closure is that it ex- cludes stakeholders that would not agree with the regulatory solutions and compro- mises hammered out as ‘weak culture’ (Schultz and Breiger, 2010). This notion suggests that one can draw new and differ- ent actors (foreigners) into this process with the right choice of words. Oligarchic closure and cross-cutting networks have consequences for political participation: if you are not at the table, you are on the menu. Applications and developments in neo- structural economic sociology Regulation and institutionalization – ap- proached from the perspective of the strong political influence of ‘improbable’ multi-status oligarchs with high-status in- consistency and rhetorical skills – are among the processes that we have studied the most7 in neo-structural economic soci- ology. It has been productive to invest in this field of research because it was char- acterised by an era of ‘governance’. In this era, the command and control framework of states weakened and business was trying even harder than usual to take advantage of high-status inconsistency and conflicts of interests to build its own self-contained normative spaces where it could define the rules and the dual dimension of joint regu- lation of markets without politicians and citizens interfering with their institutional entrepreneurship. To illustrate this neo-structural ap- proach, let me present two cases of institu- tionalization of new norms in business- related judicial institutions (Lazega, 2003), one French and one European, looking at how relational infrastructures and high- status inconsistencies are mobilized in these institutionalization processes. These are studies of how business takes ad- vantage of high-status inconsistency, con- flicts of interests and the dual dimension of joint regulation of markets, most often with strong but discreet help from public authorities8. In these studies, the main in- stitutional entrepreneurs are judges who are not shy of exposing the political di- 7 We have tried to combine these elements in a cycle of conferences called Réseaux et ré- gulation that lasted 20 years (1996–2016) and that was set up and organized together with Lise Mounier. 8 It is not the place for a detailed description of face-to-face interviews (with sociometric questions, reconstitutions of careers, open- ended questions, vignettes, etc.) of actors involved in institutional entrepreneurship (renegotiation of new norms) and the usual difficulties of ‘studying up’. Networks and Institutionalization Connections insna.org | Issues 1&2 | Volume 37 | 13 mension of their work. The first study looks at how the financial industry has captured a judicial institution, the Com- mercial Court of Paris (2000–2005). The second study looks at the currently emerg- ing European Unified Patent Court. Both raise anew Selznick’s (1949) old issues of how to design public institutions in a world of governance (to use today’s vocabulary) where an institution becomes a different thing to different people, and where each stakeholder pushes towards goal drift. Promoting norms at the Commercial Court of Paris (2000–2005) The Commercial Court of Paris is a 450- year-old court specialized in commercial litigation and bankruptcy. It is the first- level judicial institution enforcing, creating and updating business norms. The 1807 French Code de Commerce was written by its President. Heterogeneous, high and in- consistent dimensions of the social status of the judges in this court come from the fact that they are all business persons, of- ten successful executives or entrepreneurs, and at the same time officers of justice with the highest possible probity expected from (sworn-in) civil servants/judicial judges. The institution is truly judicial, and delivers fast, cheap, pragmatic, precedent- based justice using a special ‘practical’ procedure. Judges are not just business people, but also voluntary lay judges (50% with a law degree) who are endorsed by an industry and coopted by their peers. 45% are retired and 55% are still paid by their employers or own their own company (in 2000). The composition of the court is characterised by an over-representation of the financial industry (38% of all judges are bankers and insurers). The building sector, another highly litigious sector, sends in 10% of the judges. There are many examples of how these judges do punch above their weight. One example is provided by the story of how they saved the French banking sector from bankruptcy in the 1990s (Lazega, Mounier, Lemaire, 2017). Another example is the very resili- ence of this institution thanks to its built-in problem of conflicts of interests and insti- tutional capture. This court has a strong formal or- ganization, with a rule that judges rotate each year around its carrousel of 20 cham- bers. We have used the metaphor of the spinning top to model the combination of these organized mobilities (rotation) and emergence of a pecking order in the dy- namics of the advice networks among all these judges (Lazega, Lemercier, Mounier, 2006). High-status inconsistency is obvi- ous and conflicts of interests are institu- tionalized (Lazega & Mounier, 2012), as shown by the composition of the chambers in this court in 2000, illustrated by Figure 1. In this figure, bankers are in pink. Banks are the main creditors in the economy; the very simple fact that so many bankers sit in bankruptcy chambers illustrates the level of institutional capture. Connections Networks and Institutionalization 14 | Volume 37 | Issues 1&2 | insna.org Figure 1: Status inconsistency and conflicts of interests at the Commercial Court of Paris: Composi- tion of Chambers in 2000. Judges-bankers in pink However, the domination of bankers also extends, beyond bankruptcy chambers, to epistemic domination in the other activities of this court. To show this, we measured the advice network among these judges in 2000, 2002 and 2005, as represented in Fi- gure 2. An analysis of this dataset using Snijders and Nowicki’s (1997) stochastic blockmodelling shows that, in these net- works, bankers with a law degree are con- sistently the most central advisors for their peers. Statistical analyses (mentioned above) of the dynamics of this structure confirm how this collegial oligarchy con- trols the court epistemically. Bankruptcy chambers Bankruptcy chamber Chamber of Appeal against decisions made by bankruptcy chambers Networks and Institutionalization Connections insna.org | Issues 1&2 | Volume 37 | 15 Figure 2: How do bankers exercise epistemic control in the court and dominate as multi-status oli- garchs? Cyclical dynamics (centralization-decentralization-recentralization) of status in advice net- works among all judges Red: Super-central core, Green: 1st Semi-periphery, Yellow: 2nd Semi-periphery, Blue: Periphery Wave 1 Wave 2 Wave 3 The three figures visualize the cyclical dynamics (centralization-decentralization-recentralization) of status in the advice network among the judges in the court. Core, semi-peripheries and periphery iden- tified with Snijders & Nowicki’s (1997) stochastic blockmodelling. Source: Lazega, Sapulete & Mou- nier (2011). Connections Networks and Institutionalization 16 | Volume 37 | Issues 1&2 | insna.org The evolution of social and epistemic sta- tus in these cyclical dynamics is indeed characterized by the capacity of this hand- ful of members to keep their epistemic au- thority and maintain their highest centrality scores over time: they surf at the top of a wave of centralization, decentralization and recentralization of the system that was measured in this case. Thus, and importantly from a neo- structural perspective focused on institu- tionalization, the multiple dimensions of status of these specific actors in this sys- tem are high, inconsistent and stable rela- tive to the cyclical dynamics of the system. This capacity creates a competitive ad- vantage in the struggle to define the priori- ty norms in the court. Bankers with a law degree not only exercise epistemic control but also use this control to promote specific norms in the court. We measured normative controver- sies in the court and asked each judge to read jurisprudential cases summarizing these controversies, to use their discretion in judicial decision making, and to discuss them with us in face-to-face interviews. One of the cases concerned ‘predatory prices’ and assessment of punitive damag- es in a case of unfair competition. We found that money does talk in this institu- tionalization process: bankers were spread- ing self-serving norms of non-punitivity for such damages. Snijders’ Siena models (Lazega, Mounier, Snijders and Tubaro, 2012) of increasing super-centrality of bankers with a law degree over time helped us track the expected dynamics, ex- posing the fact that no pure normative ho- mophily was driving this spread of non- punitivity. It was rather reflexive align- ment of the majority of judges on the nor- mative choices of the super-central bankers (multi-status oligarchs) who achieved normative steering of the court by building epistemic authority thanks to their super- centrality combined with predictable rheto- rics of status sacrifice: ‘We [banks] could benefit from the current excess of punitivi- ty, but punitivity in principle is bad for the economy’. Other judges aligned their nor- mative choices with those of these multi- status oligarchs. This institutionalization process is not driven by pure normative choices. It is rather driven by normative choices com- bined with specific relational infrastructur- al dynamics in two processes. Firstly, by a dialectic of overload for the super-central members (who cannot answer all the re- quests for advice) and normative conflicts (Lazega, Sapulete and Mounier, 2011). Secondly, by the carrousel of chambers across which judges are rotated once a year, with many non-random infractions to the rule and a limited time horizon for all, including for the members in the core emerging from the movement. Driven by rotation and induced by delegation and turnover in these advice networks, oscilla- tion (centralization, decentralization, re- centralization) produces a form of rewiring that signals to most judges whose norma- tive choices survive in these normative controversies. This first example confirms the capac- ity of the neo-structural theory to account for this institutionalization. Our second ex- ample also involves judges punching above their weight in institution building. Institutionalization of new norms for the European Unified Patent Court This second case is about institutionaliza- tion of common norms at the emerging Eu- ropean Unified Patent Court that is meant to build a unified intellectual property (IP) regime for all European countries. It is based on the same approach as in the pre- vious example, this time with judges work- ing at the international level to build this judicial institution (scheduled to come into existence in 2018). It starts with a familiar pattern of EU institution building, begin- ning with the failure of European national governments to agree on common solu- tions and to concentrate enough power at the European level to impose such institu- tions (Dehousse, 1999; Thatcher, 2002). Here institutional entrepreneurs are corpo- rate lawyers specialized in patents, repre- Networks and Institutionalization Connections insna.org | Issues 1&2 | Volume 37 | 17 sentatives of a public/private regulator (the European Patent Office, EPO) and activist European judges with high-status incon- sistency (civil servants openly involved in lobbying and political work). The position of these judges can also be characterized by high-status inconsistency because they openly cross the lines of division of pow- ers – thus aggrandizing their role in de- mocracy. This lobbying for the creation of a transnational court to enforce a legal in- strument, the European patent created in 1973, was accompanied by work trying to create a common interpretation (‘harmoni- zation’) of the European patent. Indeed, the normative controver- sies, both substantive and procedural, about the interpretation of this patent con- cern the assessment of the ‘inventive step’ of an invention, the determination of the scope of protection afforded by the patent, the involvement of technical experts, and the use by judges of personal rules consid- ering patents either as exceptions to the rule of copying or as rewards for the inven- tor. These disagreements led to a frag- mented normative space in European IP. An example concerning an anti-depressant drug called Escitalopram is a good illustra- tion. Generic supplier companies have long sought the revocation of a basic patent held by a pharmaceutical company that first synthesized this molecule. A Dutch court in The Hague decided on complete cancel- lation of the product and process claims for lack of an ‘inventive step’. In Germany, the Bundespatentgericht decision was the same. However, a decision of the French Tribunal de Grande Instance in Paris was that the product claims by the pharmaceu- tical company were valid. In the United Kingdom, a first instance court decided to invalidate the product claims, but for a dif- ferent reason (namely insufficiency of dis- closure). The case is still regularly revisit- ed. This fragmentation facilitates forum shopping by multinational companies, and the emergence of ‘zombie’ patents. Prodded and assembled by the cor- porate lawyers and by EPO, activist Euro- pean IP judges started to network and con- vene at the so-called Venice Forum, a field-configuring event, to lobby for the construction of this European court and work together on harmonizing their inter- pretations by hammering out a ‘European Compromise’. Where governments failed, a small collegial oligarchy of super-central judges emerges from these events and is poised to create the new European IP re- gime. This collegial oligarchy of judges is perceived by their peers as primi inter pares who should sit on the future Court of Appeal of the UPC and make decisions that will create a common jurisprudence. Three different networks of European judges were measured at the Venice Fo- rum: the network of national and foreign peers (present at the Venice Forum) with whom judges personally discussed patent issues; the network of national and foreign peers whose decisions they read; and the network of national and foreign peers whose decisions they had actually cited at least once in their own decisions. Figures 3.1 to 3.4 identify the most central high- status inconsistency Venice Forum judges assembled in a ‘conclave’9, or collegial ol- igarchy that was both central in the three networks of transnational social exchanges observed among these judges at the Venice Forum, and in a fourth network of judges considered to be representatives of the fu- ture uniform position about patents in Eu- rope (henceforth called ‘uniform net- work’). Note that the reference network (respondent i cites explicitly – in his/her own decisions – decisions made by foreign judges j) is the sparsest network because explicit reference to work of foreign judges is forbidden in some countries (for exam- ple France), but it remains the main net- work in terms of influence. 9 This word was used by lawyers and judges themselves. For example: ‘On a personal level, cohabitation – almost in a conclave – allowed us to get to know and appreciate each other (…) Patent judges from the main European intellectual property countries confronted their points of view, sometimes very frankly, but always with courtesy’. Connections Networks and Institutionalization 18 | Volume 37 | Issues 1&2 | insna.org Figures 3.1–3.3: The Venice Forum collegial oligarchy: Three different networks of European judges measured at the Venice Forum Figure 3.1: Reading network Figure 3.2: Discussion network Figure 3.3: Reference network Networks and Institutionalization Connections insna.org | Issues 1&2 | Volume 37 | 19 Figure 3.4: Uniform network. Judges perceived by their peers as closest to a future uniform European position (the future ‘European Compromise’, if any) *: Super-central judges. Source: Lazega (2012) Multi-status oligarchs, i.e. super- central UK, German, Dutch judges, domi- nate this heterogeneous set of 38 Venice Forum judges. Losers are French, Southern and Central European judges. Super-centrality in the uniform network is explained at 99% by centrality in the three other networks, by member- ship in a block of countries sharing the same kind of capitalism, and by judges’ use of experts in the legal procedure (a dis- criminant feature separating the UK from the Continent at the time). Analysis of the judges’ positions in the controversy com- bined with their positions in the networks shows a slightly chaotic situation: personal ties between judges across borders do not lead (automatically) to convergence and ‘harmonization’. At the time (2009), judg- es actually disagreed on several key issues with the peers that they had selected as the future representatives of the Unified Euro- pean position. These super-central judges at the core of the network, the future rule makers according to their peers, also did not agree with each other (yet). This analy- sis of the Venice Forum relational infra- structure identified the judges who later came up with the 2016 Rules of Procedure of the UPC, a compromise based on rhetor- ical/possible ‘sacrifices’ of distinctive pro- cedural features of each big country’s na- tional ‘legal culture’: orality and adversari- al procedures by the UK; saisie- contrefaçon by France; and bifurcation by German judges. Lazega, Quintane and Casenaz (2016) provide ERGM analyses of the emergence of this uniform network and identify the costs for each of the Ven- ice Forum judges in terms of alignment upon the kind of judge whom they have se- lected in this uniform network. Based on future ‘forced’ normative alignments, win- ners of the institutionalization process are Germany and the UK, losers are Southern and Central Europe. Here again, this case confirms the capacity of the neo-structural theory of institutional entrepreneurship outlined above to account for the construc- tion of this institution. Conclusion To sum up, people are not equal in their capacity to defend their regulatory inter- ests: some punch above their weight by bringing together structure and culture to shape collective agency, process by pro- cess, particularly regulatory and institu- tionalization processes in political work. Connections Networks and Institutionalization 20 | Volume 37 | Issues 1&2 | insna.org Neo-structural institutionalism (Lazega, 2001, 2012a, 2016) shows how helpful it is to identify specific relational infrastruc- tures in socially organized settings to help model processes of regulation and institu- tionalisation of norms and practices. Insti- tutionalization is characterised by specific social dynamics of oligarchical negotiation of ‘precarious values’ (Selznick, 1957) and stabilization in interpretation of the rules en vigueur. In these dynamics, institutional entrepreneurs with heterogeneous and in- consistent forms of social status (measured also in network terms) can have a particu- lar influence in promoting their regulatory interests: when defining priority rules; when using a rhetoric of relative sacrifice to build legitimacy and manage the losers; when articulating and synchronizing regu- lation levels as vertical linchpins. High- status inconsistency is indeed the main characteristic of relational infrastructures mobilized in institutionalization of contro- versial norms. The added value of taking into account the real complexities of high- status inconsistencies was illustrated in two different cases. Analogies between the two cases are not superficial: the same concepts (relational infrastructure, status heterogeneity and inconsistencies, conflicts of interests, multi-status oligarchs, collegi- al oligarchy, rhetorics of building legitima- cy and credibility) and measurements prove useful in accounting for institution- alization. In the neo-structural model of the regulatory process, and in politics in gen- eral, individuals do not just represent themselves. Here they represent organiza- tions, professions, their country, their type of capitalism, or even legal cultures10. This means that the process is necessarily a multi-level one, with superimposed forms of collective agency, justifying the scien- tific work at the level of granularity pre- 10 Legal cultures could be considered to be a level of collective agency if they are defined as dramaturgies, i.e. as involving a text, a scenario, players, roles, strategies, skills and all the ingredients of a play. sented here. For example, judges and coun- tries represent different levels. Within- country relations between judges as nodes differ essentially from between-country re- lations between judges, which is distinct from between-country relations, where countries are the nodes. What is therefore needed in further exploration of this insti- tutionalization process is a combined dy- namic and multi-level perspective (Snijders, 2016, 2017). Indeed, neo- structural institutionalism requires new and richer kinds of data structures, modelling the emergence of an institution using more powerful stochastic actor-oriented models for dynamic multi-level networks, and per- haps new measures. For example, since these dynamics are multi-level, synchroni- zation11 issues between levels also emerge in institutionalization. Related to synchro- nization and its costs, since institutionali- zation is so oligarchical and driven by closed and collegial elites, how should the widening democratic deficit that comes at- tached to the process, and that has been criticised in the public debate over the last few years – especially in relation with the public/private dimension of such institu- tions – be questioned? Such questions show that it is necessary, for a better un- derstanding of the relation between struc- ture and process, to articulate a neo- structural institutionalism (that exploits the critical potential of network analyses of dynamic multi-level systems of interde- pendences) with the political and cultural institutionalisms that prevail in sociology today. In that respect, much remains to be done. 11 Synchronization refers here to the construc- tion of intermediary-level relational infras- tructures where actors combine temporali- ties of collective action taking place above and below their intermediary-level, and try to behave in a way compatible with their own interpretations of norms coming from both upper and lower levels (Lazega, 2016c). Networks and Institutionalization Connections insna.org | Issues 1&2 | Volume 37 | 21 References Bathelt, H. & Glückler, J. (2014). Institutional change in economic geography. Progress in Human Ge- ography, 38(3), 340-63. Christopoulos, D. & Ingold, K. (2015). Exceptional or just well connected? Political entrepreneurs and brokers in policy making. European Political Sci- ence Review, 7(3), 475-498. Dehousse, R. (1999). L'Europe par le droit, Critique internationale, 2,133-50. DiMaggio, P. (1988). Interest and agency in institu- tional theory. In: L. Zucker (ed.), Institutional pat- terns and culture. Cambridge, MA: Ballinger Pub- lishing Company, 3-22. Favereau, O. & Lazega, E. (eds.) (2002). Conventions and Structures in Economic Organization: Mar- kets, Networks, and Hierarchies. Cheltenham: Edward Elgar Publishing. Glückler, J., Suddaby, R. & Lenz, R. (eds.) (2018). Knowledge and Institutions. Heidelberg: Springer. Hall, P.A. & Taylor, R. (1996). Political science and the three new institutionalisms. Political studies, 44, 936-57. Hoffman, A. J. (1999). Institutional evolution and change: Environmentalism and the US chemical industry. Academy of Management Journal, 42, 351-71. Huault, I. & Richard, Ch. (eds.) (2012). Finance: The Discreet Regulator. How Financial Activities Shape and Transform the World, London: Pal- grave-Macmillan. Hughes, E. C. (1945). Dilemmas and contradictions of status. American Journal of Sociology, 50, 353-59. Lahille E. (2015). Le politique dans la Théorie de la Régulation: Bilan et Perspective, Colloque La Théorie de la Régulation à l’épreuve des crises, Paris, 9-12 juin 2015, Université Paris-Diderot- Inalco. Lawrence, T. B., Suddaby, R. & Leca, B. (2011). In- stitutional Work: Refocusing Institutional Studies of Organization, Journal of Management Inquiry 20(1): 52-58. Lazega, E. (2001). The Collegial Phenomenon, Ox- ford: Oxford University Press. Lazega, E. (2003). Networks in legal organizations: On the protection of public interest in joint regula- tion of markets, Oratie for Wiarda Chair, Wiarda Institute Publications, Faculty of Law, Utrecht University. Lazega, E. (2006). Capital social, processus sociaux et capacité d’action collective, in A.Bevort and M.Lallement (eds.), Capital social: Echanges, ré- ciprocité, équité, Paris, La Découverte, 213-25. Lazega, E. (2012a). Sociologie néo-structurale. In: R.Keucheyan et G.Bronner (eds.), Introduction à la théorie sociale contemporaine. Paris: Presses Universitaires de France. Lazega, E. (2012b). Time to Shrink to Greatness? Networks and Conflicts of Interests in Large Pro- fessional Firms, Revue für postheroisches Man- agement, 10, 68-77. Lazega, E. (2012c). Learning from lobbying: Mapping judicial dialogue across national borders among European intellectual property judges. Utrecht Law Review, http://www.utrechtlawreview.org, 8(2) (May) 2012. Lazega, E. (2016a). Réseaux et régulation: Pour un in- stitutionnalisme néo-structural, Revue de la Régu- lation : Capitalisme, institutions, pouvoirs, https://regulation.revues.org/11902#text Lazega, E. (2016b). Joint ‘anormative’ regulation from status inconsistency: A multilevel spinning top model of specialized institutionalization, in Margaret S. Archer (ed.), Anormative Regulation in the Morphogenic Society, Springer, 169-90, Lazega, E. (2016c). Synchronization costs in the or- ganizational society: Intermediary relational infra- structures in the dynamics of multilevel networks, in E.Lazega and T.A.B. Snijders (eds.), Multilevel Network Analysis for the Social Sciences, Spring- er, 47-77. Lazega, E., Mounier, L. & Lemaire, S. (2017). Finan- cial pragmatism and bankers’ politics: The duality of the “Zombies” debate in bankruptcy proceed- ings at the Commercial Court of Paris (2000- 2005). In: V.Boussart (ed.), Finance at Work, London: Routledge. Lazega, E. & Mounier, L. (2012). Networks of institu- tional capture. In: B.Vedres & M.Scotti (eds.), Networks in Social Policy Problems. Cambridge: Cambridge University Press, 124-37. Lazega, E., Lemercier, C. & Mounier L. (2006). A spinning top model of formal structure and infor- mal behaviour: Dynamics of advice networks in a commercial court. European Management Review, 3, 113-22. Lazega, E., Sapulete, S. & Mounier, L. (2011). Struc- tural stability regardless of membership turnover? The added value of blockmodelling in the analysis of network evolution. Quality & Quantity, 45, 129-44. Lazega, E., Mounier, L., Snijders, T.A.B. & Tubaro, P. (2012). Norms, status and the dynamics of ad- vice networks. Social Networks, 34, 323-32. Lazega, E., Quintane, E. & Casenaz, S. (2016). Colle- gial Oligarchy and Networks of Normative Alignments in Transnational Institution Building: The Case of the European Unified Patent Court. Social Networks, 48, 10-22. Lenski, G. E. (1954). Status crystallization: a non- vertical dimension of social status. American Soci- ological Review, 19(4), 405-13. Owen-Smith, J. & Powell, W.W. (2008). Networks and Institutions, in Greenwood, R., Oliver, C., Sahlin-Andersson, K. & Suddaby, R. (eds.). The Sage Handbook of Organizational Institutional- ism, 596-623, London: Sage. Connections Networks and Institutionalization 22 | Volume 37 | Issues 1&2 | insna.org Perrow, Ch. (1991). A society of organizations. Theo- ry and Society, 20, 725-62. Scott, W. R. (2007). Institutions and organizations. Thousand Oaks: Sage. Schultz, J. & Breiger, R. L. (2010). The strength of weak culture. Poetics, 38(6), 610-624. Selznick, P. (1949). TVA and the grass roots: A study in the sociology of formal organization, 3, Berke- ley: University of California Press. Selznick, Ph. (1957). Leadership in Administration. Evanston, Ill.: Row, Peterson & Co. Snijders, T. A., & Nowicki, K. (1997). Estimation and prediction for stochastic blockmodels for graphs with latent block structure. Journal of classifica- tion, 14(1), 75-100. Snijders, T.A.B. (2016). The multiple flavours of mul- tilevel issues for networks, in Lazega, E. and Snijders, T.A.B. (eds.), Multilevel network analy- sis for the social sciences. Springer International Publishing, 15-46. Snijders, T.A.B. (2017). Stochastic Actor-Oriented Models for Network Dynamics. Annual Review of Statistics and Its Application, 4, 343-63. Thatcher, M. (2002). Analysing regulatory reform in Europe. Journal of European Public Policy, 9(6), 859-72.