ORGANIZATIONAL
TRANSFORMATION AND
SCIENTIFIC CHANGE THE
IMPACT OF INSTITUTIONAL
RESTRUCTURING ON
UNIVERSITIES AND
INTELLECTUAL INNOVATION
RESEARCH IN THE SOCIOLOGY OF
ORGANIZATIONS
Series Editor Michael Lounsbury
Recent Volumes
Volume 25 The Sociology of Entrepreneurship
Volume 26 Studying Difference between Organizations Comparative
Approaches to Organizational Research
Volume 27 Institutions and Ideology
Volume 28 Stanfordrsquos Organization Theory Renaissance 19702000
Volume 29 Technology and Organization Essays in Honour of Joan Woodward
Volume 30A Markets on Trial The Economic Sociology of the US Financial
Crisis Part A
Volume 30B Markets on Trial The Economic Sociology of the US Financial
Crisis Part B
Volume 31 Categories in Markets Origins and Evolution
Volume 32 Philosophy and Organization Theory
Volume 33 Communities and Organizations
Volume 34 Rethinking Power in Organizations Institutions and Markets
Volume 35 Reinventing Hierarchy and Bureaucracy From the Bureau to
Network Organizations
Volume 36 The Garbage Can Model of Organizational Choice Looking
Forward at Forty
Volume 37 Managing lsquoHuman Resourcesrsquo by Exploiting and Exploring
Peoplersquos Potentials
Volume 38 Configurational Theory and Methods in Organizational Research
Volume 39A Institutional Logics in Action Part A
Volume 39B Institutional Logics in Action Part B
Volume 40 Contemporary Perspectives on Organizational Social Networks
Volume 41 Religion and Organization Theory
RESEARCH IN THE SOCIOLOGY OF ORGANIZATIONS
VOLUME 42
ORGANIZATIONALTRANSFORMATION ANDSCIENTIFIC CHANGE
THE IMPACT OFINSTITUTIONAL
RESTRUCTURING ONUNIVERSITIES ANDINTELLECTUALINNOVATION
EDITED BY
RICHARD WHITLEYUniversity of Manchester United Kingdom
JOCHEN GLASERTU Berlin Germany
United Kingdom North America Japan
India Malaysia China
Emerald Group Publishing Limited
Howard House Wagon Lane Bingley BD16 1WA UK
First edition 2014
Copyright r 2014 Emerald Group Publishing Limited
Reprints and permission service
Contact permissionsemeraldinsightcom
No part of this book may be reproduced stored in a retrieval system transmitted in
any form or by any means electronic mechanical photocopying recording or
otherwise without either the prior written permission of the publisher or a licence
permitting restricted copying issued in the UK by The Copyright Licensing Agency
and in the USA by The Copyright Clearance Center Any opinions expressed in the
chapters are those of the authors Whilst Emerald makes every effort to ensure the
quality and accuracy of its content Emerald makes no representation implied or
otherwise as to the chaptersrsquo suitability and application and disclaims any warranties
express or implied to their use
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
ISBN 978-1-78350-684-2
ISSN 0733-558X (Series)
Certificate Number 1985ISO 14001
ISOQAR certified Management Systemawarded to Emerald for adherence to Environmental standard ISO 140012004
CONTENTS
LIST OF CONTRIBUTORS ix
ADVISORY BOARD xi
EDITORSrsquo INTRODUCTIONRichard Whitley and Jochen Glaser 1
INSTITUTIONAL CHANGE AND THE TRANSFORMATIONOF UNIVERSITIES AS STRATEGIC ACTORS
THE IMPACT OF INSTITUTIONAL REFORMS ON THENATURE OF UNIVERSITIES AS ORGANISATIONS
Richard Whitley and Jochen Glaser 19
EMPOWERMENT OF FRENCH UNIVERSITIES BYFUNDING AND EVALUATION AGENCIES
Christine Musselin 51
FUNDING INDIVIDUALS CHANGINGORGANISATIONS THE IMPACT OF THE ERC ONUNIVERSITIES
Jakob Edler Daniela Frischer Michaela Glanz andMichael Stampfer
77
WHERE HAVE ALL THE SCIENTISTS GONEBUILDING RESEARCH PROFILES AT DUTCHUNIVERSITIES AND ITS CONSEQUENCES FORRESEARCH
Grit Laudel and Elke Weyer 111
v
MERGER MANIA IN SCIENCE ORGANIZATIONALRESTRUCTURING AND PATTERNS OFCOOPERATION IN AN ACADEMICRESEARCH CENTRE
Julien Barrier 141
CONTRADICTORY CONSEQUENCES OF INSTITUTIONALCHANGES ON INTELLECTUAL INNOVATION IN
THE PUBLIC SCIENCES
INSTITUTIONAL CONDITIONS AND CHANGINGRESEARCH PRACTICES IN SWITZERLAND
Martin Benninghoff Raphael Ramuz Adriana Gorga andDietmar Braun
175
COLD ATOMS HOT RESEARCH HIGH RISKSHIGH REWARDS IN FIVE DIFFERENT AUTHORITYSTRUCTURES
Grit Laudel Eric Lettkemann Raphael RamuzLinda Wedlin and Richard Woolley
203
HIGHLY ADAPTABLE BUT NOT INVULNERABLENECESSARY AND FACILITATING CONDITIONSFOR RESEARCH IN EVOLUTIONARYDEVELOPMENTAL BIOLOGY
Grit Laudel Martin Benninghoff Eric Lettkemann andElias Hakansson
235
PATH DEPENDENCE AND POLICY STEERING INTHE SOCIAL SCIENCES THE VARIED IMPACT OFINTERNATIONAL LARGE SCALE STUDENTASSESSMENT ON THE EDUCATIONAL SCIENCES INFOUR EUROPEAN COUNTRIES
Jochen Glaser Enno Aljets Adriana Gorga Tina HedmoElias Hakansson and Grit Laudel
267
vi CONTENTS
WHERE TO GO FOR A CHANGE THE IMPACT OFAUTHORITY STRUCTURES IN UNIVERSITIES ANDPUBLIC RESEARCH INSTITUTES ON CHANGES OFRESEARCH PRACTICES
Jochen Glaser Enno Aljets Eric Lettkemann andGrit Laudel
297
COMPUTER CORPUS LINGUISTICS ANINNOVATION IN THE HUMANITIES
Lars Engwall Enno Aljets Tina Hedmo andRaphael Ramuz
331
HOW DO INSTITUTIONAL CHANGES AFFECTSCIENTIFIC INNOVATIONS THE EFFECTS OFSHIFTS IN AUTHORITY RELATIONSHIPSPROTECTED SPACE AND FLEXIBILITY
Richard Whitley 367
viiContents
LIST OF CONTRIBUTORS
Enno Aljets Department of Sociology University ofBremen Germany
Julien Barrier Triangle - CNRS research unit EcoleNormale Superieure de Lyon France
Martin Benninghoff Institut drsquoEtudes Politiques etInternationales University of LausanneSwitzerland
Dietmar Braun Institut drsquoEtudes Politiques etInternationales University of LausanneSwitzerland
Jakob Edler Manchester Business School University ofManchester United Kingdom
Lars Engwall Department of Business Studies UppsalaUniversity Sweden
Daniela Frischer Faculty of Education McGill UniversityMontreal Canada
Michaela Glanz Art | Research | Service Academy of FineArts Vienna Austria
Jochen Glaser Center for Technology and Society TUBerlin Germany
Adriana Gorga Institut drsquoEtudes Politiques etInternationales University of LausanneSwitzerland
Tina Hedmo Department of Business Studies UppsalaUniversity Sweden
Elias Hakansson Department of Economic History UppsalaUniversity Sweden
ix
Grit Laudel Center for Higher Education PolicyStudies University of Twente EnschedeGermany
Eric Lettkemann Department of Sociology TU BerlinGermany
Christine Musselin Sciences Po and CNRS Paris France
Raphael Ramuz Institut drsquoEtudes Politiques etInternationales University of LausanneSwitzerland
Michael Stampfer Vienna Science and Technology FundVienna Austria
Linda Wedlin Department of Business Studies UppsalaUniversity Sweden
Elke Weyer Center for Higher Education PolicyStudies University of Twente EnschedeThe Netherlands
Richard Whitley Manchester Business School University ofManchester United Kingdom
Richard Woolley Institute for Innovation and KnowledgeManagement Research PolytechnicUniversity of Valencia Spain
x LIST OF CONTRIBUTORS
ADVISORY BOARD
SERIES EDITORS
Michael LounsburyAssociate Dean of Research
Thornton A Graham ChairUniversity of Alberta School of Business and National
Institute for Nanotechnology Alberta Canada
ADVISORY BOARD MEMBERS
Howard E AldrichUniversity of North CarolinaUSA
Stephen R BarleyStanford University USA
Nicole BiggartUniversity of California atDavis USA
Elisabeth S ClemensUniversity of Chicago USA
Jeannette ColyvasNorthwestern University
Barbara CzarniawskaGoteborg University Sweden
Gerald F DavisUniversity of Michigan USA
Marie-Laure DjelicESSEC Business SchoolFrance
Frank R DobbinHarvard University USA
Royston GreenwoodUniversity of Alberta Canada
Mauro GuillenThe Wharton School University ofPennsylvania USA
Paul M HirschNorthwestern University USA
Brayden KingNorthwestern University
Renate MeyerVienna University of Economicsand Business AdministrationAustria
Mark MizruchiUniversity of Michigan USA
Walter W PowellStanford University USA
xi
Hayagreeva RaoStanford University USA
Marc SchneibergReed College
W Richard ScottStanford University USA
Haridimos TsoukasALBA Greece
xii ADVISORY BOARD
EDITORSrsquo INTRODUCTION
Richard Whitley and Jochen Glaser
Since World War II the funding and governance of higher education andscientific research systems have undergone a number of major changes inmost industrialised capitalist societies In particular the state and otherfunding agencies have become more proactive in seeking to steer the direc-tion of academic research universities are being encouraged to be moreaccountable and strategic in their behaviour and the commercialisation ofresearch has taken new forms (Whitley Glaser amp Engwall 2010 Ziman2000) These changes represent significant shifts in the organisation of thesciences and have altered the nature of universities as strategic actors inmany countries including measures that increase their autonomy from thestate and strengthen their internal managerial governance (KruckenKosmutzky amp Torka 2007 Paradeise Reale Bleiklie amp Ferlie 2009Whitley 2012) In transforming the organisational contexts in whichresearch is carried out these reforms have had substantial consequences forthe dynamics of scientific change
These changes have been accompanied by a rapid increase in the typesand numbers of formal organisations involved in the production coordina-tion and evaluation of published scientific knowledge Research organisa-tions themselves have not only grown in numbers but are also more variedin their structures and missions (eg Merz amp Biniok 2010) and haveformed associations that organise lobbying or collaboration The wide-spread transition from recurrent to project funding for scientific research
Organizational Transformation and Scientific Change The Impact of Institutional
Restructuring on Universities and Intellectual Innovation
Research in the Sociology of Organizations Volume 42 115
Copyright r 2014 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN 0733-558Xdoi101108S0733-558X20140000042001
1
has increased the importance of funding agencies which channel moneyfrom the state or from industry to science (Benninghoff amp Braun 2010Braun 1998) Science policy making has additionally become based on alarge number of advisory bodies intermediary and lobby organisations(Van der Meulen amp Rip 1998) and the institutionalisation of formal eva-luations of research performance has given rise to agencies that oversee orconduct these (Whitley amp Glaser 2007) Researchers also make use ofcommercial organisations especially publishers (Thompson 2005) and pro-viders of research equipment materials and services (Kleinman 2003pp 9395) Finally professional organisations of academics have becomemore active in both organising collaboration and lobbying for the interestsof researchers (Schofer 1998 Simpson 2002)
With these developments control over two key processes crucial forknowledge production the allocation of resources and of reputations has become increasingly channelled through formal organisations so thatboth the strategic direction of research goals and changes in technical prac-tices depend more and more on organisational processes As a conse-quence access to and influence over formal organisations become crucialfor gaining authority over the selection of research goals and the use that ismade of research results as in the central role of advisors to funding agen-cies in the allocation of scarce resources for projects
These changes in funding and governance have been investigated byscholars in science studies higher education studies and governance from avariety of perspectives Analyses of changing sciencesociety relationshipshave focused on the emergence of a new social contract between scienceand society (Guston amp Keniston 1994) the usefulness of research for eco-nomic innovations (Berman 2012 Geiger amp Sa 2005) and the levelling outof funding in what has been termed lsquosteady state sciencersquo (Ziman 1994)Studies of university systems have been mostly concerned with the changesin their formal governance structures particularly the implementation ofsome new public management ideas (eg Amaral Lynn Meek amp MarheimLarsen 2003 Braun amp Merrien 1999 Krucken et al 2007 Marginson ampConsidine 2000 Paradeise et al 2009 Whitley amp Glaser 2007)
Specific attention has also been paid to authoritative agencies such aspublic and private funding councils and to new governance instrumentssuch as national systems of research evaluation (Martin amp Whitley 2010Whitley amp Glaser 2007) Additionally the roles of research councils asintermediary organisations between the state and scientists have enjoyedconsiderable attention (Braun 1998 Rip 1994 Van der Meulen 2003)Some analyses have used principal-agent theory for analysing relationships
2 RICHARD WHITLEY AND JOCHEN GLASER
between science policy funding councils and researchers (Braun amp Guston2003 Guston 1996 Van der Meulen 1998) while others have adoptedneo-institutionalist frameworks for the study of the international diffusionof governance structures of higher education (Drori Meyer Ramirez ampSchofer 2003) and for analysing the evolving lsquoactorhoodrsquo of universities(Krucken amp Meier 2006 Weingart amp Maasen 2007) or governance-oriented approaches for analysing changes in state-university relationshipsand university management (eg Schimank 2005 Whitley 2008 2012)
Although these approaches to institutional change in higher educationand the sciences have contributed important insights most studies havefocused on formal shifts in governance and the introduction of new instru-ments for state steering rather than undertaking comparative analyses ofhow such changes are affecting the nature of universities as organisationsand their behaviour as strategic actors in differently organised highereducation systems Additionally we still lack a theoretical synthesis of theprocesses through which the numerous changes in governance and organi-sation are affecting the conduct and content of research and there havebeen very few studies that have dealt empirically with the consequences ofgovernance and authority changes on the development of different kinds ofintellectual innovations in different sciences thus integrating the analysisof macro-level institutional change with more micro-analyses of shifts inresearch strategies and careers
It is also doubtful whether we can truly understand the nature of organi-sational change in public science systems without systematically taking intoaccount the specific nature of scientific work There is a paradox hereAlthough students of the organisational impact of governance changes onresearch practices know that this varies greatly between fields this varia-tion is rarely explicitly included in studies of organisational and scientificchange Achieving an understanding of the differential impact of organisa-tional transformations on scientific change across fields of research requireslsquobringing work back inrsquo the analysis (Barley amp Kunda 2001) In their pleato bring work back into organisational analysis Barley and Kunda notedtwo troublesome strategies in the conceptualisations of post bureaucraticorganisations namely conceptual inversion the contrasting of traditionalforms of organising with an allegedly new form and environmentalism accounting for new organisational forms and practices solely as responsesto changing environments (ibid pp 7779) Both strategies have in com-mon that they distance themselves from the detailed analysis of work prac-tices and rest more on the persuasive power of metaphor than on empiricalevidence (idem)
3Editorsrsquo Introduction
Both conceptual inversion and environmentalism can be identified inorganisational sociology and higher education approaches to the changinggovernance of research Overcoming these problems requires as Barleyand Kunda put it the blurring of conceptual boundaries between organisa-tional theory and other disciplines Applied to our topic it means aninterdisciplinary collaboration between organisational sociology and thesociology of science that can be understood as the sociology of scientificwork This collaboration poses an interesting conceptual challenge becausethe sociology of science would maintain that scientific research is conductedin organisations but by communities Thus bringing work back into thestudy of formal organisations in science requires re-thinking the relation-ship between two different types of social orders and the role they playin the production of scientific knowledge Current organisational sociologyrecognises the importance of communities for organisational processes(Marquis Lounsbury amp Greenwood 2011a) However the attempt tointegrate communities into organisational theory by conceptualising themas lsquoinstitutional ordersrsquo and lsquotypes of organisingrsquo (Marquis Lounsbury ampGreenwood 2011b) risks obscuring fundamental differences between theways in which these social forms operate differences that are highlightedby Knorr-Cetinarsquos (1999) description of lsquoepistemic culturesrsquo
The contributions to this volume address the interaction between organi-sational transformation and scientific change and the complex interplaybetween social forms in these processes at two different levels The firstpart of the book focuses on organisational transformation and asks howthis transformation is shaped by and shapes in turn the nature of scientificwork In their contribution on the impact of institutional reforms on thenature of universities as organisations Whitley and Glaser address aproblem that is not only implicit to much of the higher education policydiscourse but also reverberates in higher education and organisationresearch Do the higher education reforms currently under way in manycountries lead to universities resembling firms in their authority structures
The authors start from the observation that there is no such thing aslsquothersquo firm or lsquothersquo university Instead firms vary considerably in the author-ity exercised by owners and managers that they are granted by other socialactors and in their need to share this authority with employees if theywant to build distinctive organisational capabilities Universities in nationalhigher education systems also vary considerably in their external andinternal authority sharing and different types of universities need to bedistinguished in any analysis of authority structures Underlying thesedifferences however is the basic delegation of authority over the goals
4 RICHARD WHITLEY AND JOCHEN GLASER
and conduct of research and teaching that universities necessarily grant toacademics This delegation results from the uncertainties inherent to theproduction of published scientific knowledge and to the fact that the pri-mary social context for the formulation of goals workflow interdependen-cies and the evaluation of results is the scientific community
The complex relationships between organisations and scientific commu-nities are highlighted by Musselinrsquos analysis of the role of evaluations in thegovernance of French universities The higher education reforms in Francehave granted the university management considerable formal authorityover the internal affairs of universities This authority however is neces-sary but not sufficient for university management to influence decision-making about research and teaching activities It needed to be combinedwith a second element of higher education reforms namely evaluationsbased on peer reviews if the new authority of the university leadership wasto become effective Musselinrsquos case studies in three French universitiesdemonstrate how these evaluations boosted the authority of universitymanagers by legitimising actions intended to achieve positive evaluationsBeing based on peer reviews these evaluations are difficult for researchersto dismiss By strengthening the authority of university management theyamplify the effects of governance reforms
The study by Edler et al shows that much of the effect of organisationalchange on science results from its impact on individual members of otherorganisations rather than through inter-organisational interactions Byselectively funding the research of individuals and groups within organisa-tions a funding council (in their study the European Research CouncilERC) may simultaneously affect a research organisationrsquos reputationexternal funding and internal funding structure workforce and profile ofwork At the same time funding councils affect the plans and careers ofindividual researchers for which their current organisation might or mightnot be a resource
The responses by organisations and within them are similarly complexThey depend on the organisationrsquos interests and action capabilities lsquoIn-betweenrsquo organisations in terms of their collective coordination andresource capabilities showed the most changes in their internal strategieswhile powerful universities didnrsquot need to respond to the changes in theirenvironment produced by a single research council and powerless oneswere unable to respond Despite such strategic responses universities hadlittle control over the changes in research profiles resulting from significantgrants which was strongest in universities whose overall research perfor-mance was weak because acquiring one strong research group in such an
5Editorsrsquo Introduction
organisation meant significant shifts in their research profiles as well as bal-ances of resources and reputation In some cases organisations respondedto this lack of control and to the very low success rates by attempting tofilter grant applications for quality as understood by current elites poten-tially undermining the very intentions of the ERC to support transforma-tive research
Laudel and Weyer use a case study of the changing Dutch science systemto illustrate the impact of homogeneous institutional environments forresearch on scientific communities Dutch universities are not only expectedto develop distinct research profiles but to incorporate the statersquos thematicpriorities in these profiles The Dutch new public management reformshave provided universities with the capabilities to implement such or otherpriorities by importing or removing whole groups or departments Fundingorganisations face similar expectations and additionally a pressure towardsfunding lsquousefulrsquo research These combined expectations leave few niches forresearch that is basic but does not yield highly visible publications easilyAs a result collaboration patterns within universities might be destroyedand quasi-market failures emerge that make fields disappear at the nationallevel
In his reconstruction of a merger between French laboratories Barrierinvestigates responses to a recent trend in the French science systemDuring the last two decades the CNRS has been concentrating its fundingin larger units which creates a strong institutionalised expectation tobecome bigger Rationales of better visibility to the CNRS grant fundingagencies and commercial actors as well as science and regional policy actorswere invoked in conflicts over the merger of small laboratories to form alarger unit Barrier shows how organisational structures are used byscientists as a resource in their struggle to legitimise their research and tosecure funding The very same structures were expected to influence pat-terns of research collaboration and preferred strategies within organisa-tional boundaries The lack of collaboration between groups prior to themerger contributed to tensions between key actors that endangered thewhole process Although these difficulties were overcome the impact ofthe new structures on patterns of internal collaboration was only significantin a few very specific cases which depended on the content of research
The second part of the book presents results of a comparative project onthe impact of changing authority relations in the public sciences onconditions supporting the development of different kinds of scientific inno-vations The effects of governance funding and authority shifts on oppor-tunities for researchers to develop innovations in their fields were
6 RICHARD WHITLEY AND JOCHEN GLASER
compared across countries and sciences The countries included Germanyas a country in which governance reforms have been implemented relativelylate the Netherlands as a small country which was one of the firstEuropean states to begin introducing new public management changes andSweden and Switzerland which are also small and have introduced suchreforms to a more limited extent and later than the Netherlands
The comparison of innovations was intended to cover a wide range ofepistemic properties of fields and changes in research practices This wasachieved by selecting innovations in the physical sciences life sciencessocial sciences and humanities The development of these innovationsneeded to be sensitive to authority relations Since the main channelsthrough which authority is exercised are the allocation of resources andreputation innovations were selected whose development required signifi-cant resources posed reputational risks due to conflicting opinions in thescientific community about them or both An additional requirement wasthat each innovation had to be present in each of the four countries
Observing all criteria led to the selection of four innovations Bose-Einstein condensation (BEC) of cold atom gases was first realised experi-mentally in 1995 by two US research groups It took other research groupsmore than two years to replicate the first experiments In the late 1990s and2000s research about and with Bose-Einstein condensates became arapidly growing research field Evolutionary developmental biology (EDB)is an interdisciplinary perspective that grew out of the observation that evo-lutionary theory was insufficient to explain new experimental results in the1970s Since then evolutionary and developmental biologists have con-ducted a wide range of theoretical and comparative experimental researchintended to understand the interaction between the development of organ-isms and the evolution of species Comparative international large-scalestudent assessment (ILSA) is a politically motivated innovation in educa-tional science Although these comparative assessments originally were lit-tle more than a service to state education policymakers it had an impacton national educational science communities which were tasked with car-rying out the assessments and linked their research to this task Finallycomputerised corpus linguistics (CCL) the building and analysis of largedigital databases of words phrases and sentences became an innovationin the four countries at different points in time since the 1960s dependingon national research traditions in linguistics and language studies
The investigation of the development of four innovations in four coun-tries provided rich empirical material which is used for comparisons indifferent dimensions An important analytical tool for linking the content
7Editorsrsquo Introduction
of research to institutional and organisational conditions in many of thecomparisons is the concept of protected space This term refers to the per-iod of time in which scientists have discretion over the use of neededresources including their own efforts to pursue particular problems andapproaches before having to produce publishable and collectively valuedresults It incorporates authority over the choice and formulation of whichtopics to study how to do so and how to obtain and manage resourcesAt the macro-level of national science systems opportunities to developinnovations vary depending on the level or lsquoamountrsquo of protected spaceafforded to researchers and on its distribution between qualified researchersin a particular organisation or scientific community Protected space thusbecomes an important dimension in which public science systems can becompared (see Whitley in this volume)
At the micro-level the definition of protected space is the same but theanalytical emphasis is on the opportunities of researchers to build protectedspace for their research in different national science systems organisationsand scientific communities These opportunities depend on their organisa-tional positions success with funding agencies and priorities and prefer-ences of organisations and scientific communities Individual protectedspaces vary in the time horizons for which they afford discretion toresearchers and in the amount of resources over which this discretion canbe exercised (see the contributions by Laudel et al Engwall et al and thecomparison of organisations by Glaser et al in this volume see alsoGlaser Laudel amp Lettkemann 2014)
The applicability of protected space at both system and individual levelsprovides a means of integrating individual knowledge production processeswith the organisational conditions for these processes and the conditionsprovided by national science systems It also enables the analytical integra-tion of the influence of scientific communities and formal organisations onthe development of innovations
A complementary approach is to compare the costs of changing researchpractices This perspective was applied by Benninghoff et al to examinethe development of the four innovations in one country SwitzerlandPartly owing to its size the Swiss science system had not developed allfour research traditions from which these innovations emerged While itbelonged to the leading countries in EDB due to its strong biosciencesthe study of BEC ILSAs and CCL was not taken up by Swiss scientists asquickly because the leaders of the respective fields had incompatibleresearch priorities The authors found that the relative affluence of theSwiss public science system was a key factor supporting the development of
8 RICHARD WHITLEY AND JOCHEN GLASER
these innovations over time This condition had the most impact in thesciences because Swiss universities were able to support the above-averagecosts of the infrastructures required for these innovations and were able toimport innovations as was the case with BEC for which a leading Germanresearcher was recruited Finally the good funding of researchers at theseuniversities enabled at what the authors call partial switches that is thebeginning of new lines of research alongside traditional established onesThe comparison of the four innovations in Switzerland demonstrates thatnew public management reforms do not always restrict the diversity ofresearch if introduced in the context of sufficient funding and autonomyof universities
In their paper on the development of BEC Laudel et al trace the devel-opment of this innovation in five countries including Spain in addition tothe four countries that were part of the original comparative study For thefirst 10 years after its initial realisation BEC required an amount ofresources that significantly exceeds the average of the field The time it tookto achieve experimental success was difficult to predict and could extend toseveral times the common three-year term of a project This requirement ofhigh levels of protected space made it an excellent exemplar for the impactof changing authority relations on the opportunities for researchers todevelop this innovation In all five countries opportunities to develop inno-vations that needed considerable levels of protected spaces were limited toprofessors because only they combined permanent positions with thecontrol of the infrastructure that formed part of the experimental setup
For this to happen at all it was secondly necessary that the country hada research tradition of studying cold atoms which existed in Germany andthe Netherlands but not in the other countries A third necessary conditionwas access to grant funding which in the mid- to late 1990s required scien-tific communities to allocate large grants to Bose-Einstein experimentsalthough the majority opinion still was that realising BEC was either notpossible or after it had been achieved not interesting any moreFinally the comparison showed that funding for basic research has becomeprecarious in some countries which is why the research with Bose-Einsteincondensates is growing quite unevenly in the five countries investigated
EDB differs in interesting ways from BEC Laudel et al show thatbecause it is based on a change of approach that can be applied to a rangeof experimental research in evolutionary and developmental biology theepistemic pathways to the development of this innovation are various andthe protected space required by the transitions ranges from zero to morethan a million euros in the resource dimension and has time horizons that
9Editorsrsquo Introduction
are similar to those required by BEC As a consequence the governancesystems of the four countries provided different niches in which differentkinds of transitions to EDB could be accomplished Minimum require-ments at the international level were the acceptance by established scientificcommunities that certain standard experiments could not be conductedwithout the organisms studied by evolutionary developmental biologistsAt the national level minimum requirements included support for basicresearch topics that were not currently in fashion support that eroded inthe Netherlands Finally the considerable level of protected space neededfor undertaking comparative experimental research in EDB involved theability of research organisations to provide above-average investmentsin infrastructure in the form of breeding facilities for the new species ofinterest Swiss universities and German non-university research instituteswere best equipped to provide these investments In Sweden large grantsprovided a functional equivalent for a few researchers
The innovation in the social sciences comparative large-scale studentassessments exemplifies the importance of institutionalised research tradi-tions for the national development of an innovation While the researchtraditions in the natural sciences were internationalised and nationaldevelopments depended on the presence or absence of these research tradi-tions in each country the educational sciences are predominantly nationalresearch fields that have developed largely independently of each other AsGlaser et al show the impact of the international initiative to compare stu-dent achievements could take any one of four forms (a) accepted as a validactivity (b) considered to be falling behind the state of the art in thenational scientific community (c) seen as largely irrelevant or (d) used asan opportunity to establish the new field depending on the research tradi-tions and priorities in each of the four countries In all four though stateinterest guaranteed that the data for the international comparison wouldbe collected but the nature of the data made using them for educationalresearch difficult because there was little intellectual interest in internation-ally comparative studies and research traditions of educational sciencerequired longitudinal rather than cross sectional data This is why interna-tional comparative student assessments had a significant impact only inGermany where they were used by those who conducted them to establishquantitative studies as a legitimate enterprise of educational science
A different kind of comparison is employed by Glaser et al in theirstudy of how different types of research organisation provided opportu-nities for developing innovations The authors exploit the fact that some ofthese innovations were developed in both universities and state-funded
10 RICHARD WHITLEY AND JOCHEN GLASER
public research institutes in Germany and the Netherlands By comparingthe development of BEC EDB and ILSA in universities and publicresearch institutes the authors are able to identify five characteristic pat-terns of authority sharing that rendered research situations comparableacross types of organisations career positions and countries They showthat although the highly autonomous positions at German public researchinstitutes may provide the best local conditions for changing research prac-tices universities are essential for the global development of any innovationbecause only the university system provides the necessary epistemic andinstitutional diversity for simultaneously developing an innovation andbuilding a career with it At the same time the analysis demonstrates thesuperiority of discretionary block funding in situations in which researcherswant to respondent quickly to developments not yet accepted by the major-ity of their community
The reconstruction of an innovation in the humanities CCL by Engwallet al highlights the importance of disciplinary structures and authorityrelations in scientific communities as well as the challenges posed by aninnovation that introduces epistemic practices that deviate from some ofthe traditions of the field Engwall et al characterise language studies asweakly integrated and name language barriers approaches empiricalobjects (written vs spoken language modern vs historical language use)and aims (understanding language use vs support of education) as differ-ences that contribute to the fragmentation CCL and the concomitantapproach to the empirical study of languages were at odds with the theninfluential Chomskyan attempt to integrate language studies around theanalysis of generative grammars In spite of this friction the diversity oflanguage studies and their institutionalisation in separate departmentsacross Europe made it possible for CCL to develop in a variety of ways lar-gely unhampered by such opposition National traditions of research andcross-national contacts with those scholars who advanced CCL were thekey influences on how CCL developed Both early and more recentdevelopments also illustrate a feature of CCL that linked it to organisa-tional and national authority relations Its resource intensive nature parti-cularly for constructing large corpora created some problems becauseuniversities and national funding agencies had to arrange funding for thebuilding of corpora as expensive infrastructure a concept that is relativelynew to the humanities This problem was originally solved through grantsin Sweden through a state-funded institute in Germany and through com-bined grant and university funding in the Netherlands The difficulties ofobtaining infrastructure funding from research councils in the Netherlands
11Editorsrsquo Introduction
and Switzerland indicate that the changes in epistemic practices were stillnot fully recognised by authoritative agencies
In the final paper of this part Whitley integrates the comparative studiesby providing an account of how changing authority relations in the publicsciences affect the opportunities for the development of scientific innovationsHe does this in two steps In the first step he outlines the concepts of pro-tected space and flexibility of standards governing resources and reputationsas intermediate variables linking changes in authority relations to the neces-sary conditions for the development of different kinds of scientific innova-tions Considering the impact of the major changes in authority relationsoutlined at the beginning of this introduction on these two intermediate vari-ables emphasises their divergent and sometimes contradictory consequencesfor scientistsrsquo opportunities for developing innovations Owing to varyingstrengths and speed of changes in authority relations in different publicscience systems the influence of changes in authority relations varies consid-erably between differently organised public science systems This impactfurthermore varies depending on the content of the innovation and the fieldsin which they occur
In the second step he suggests how the different patterns of develop-ment of the four innovations investigated in the comparative casestudies can be understood in terms of the relationships between thestructure of authority relations and levels of protected space and flexibilitytaking account of the epistemic contexts in which they occur Variationsin the means of accessing protected space and other resources fordifferent researchers in each public science system help to explain contrast-ing innovation patterns These connections have a number of implicationsfor the consequences of more recent funding reductions and governancechanges
Taken together the contributions to this volume reposition researchorganisations and their environments in the approaches of organisationalsociology and higher education research Bringing work back in the studyof organisational transformation in public science systems amounts to twomajor changes First it requires looking beyond the organisations in whichresearch is conducted and research policies are constructed All theseorganisations are overlaid by scientific communities and their authoritystructures which compete with organisational authority structures for thecontrol of research and teaching the allocation of resources and the eva-luation of results At the same time the local shaping of research andteaching by organisations contributes to an aggregate effect of publicscience systems on the research conducted by scientific communities
12 RICHARD WHITLEY AND JOCHEN GLASER
The interaction between organisations and communities constrains theauthority over research exercised in both social contexts
Second scientific work itself has properties that modify the impact ofinstitutional environments and organisational structures Research and byextension much teaching are inherently uncertain with regard to the nat-ure of problems formulated workflow interdependencies and the meaningsof their outcomes Beyond these global properties that limit the level oforganisational control of work beyond operating groups these processeshave specific epistemic properties that make their susceptibility to organi-sational control and governance specific to fields and even to individualresearch processes or innovations Bringing work back in the analysis ofscientific change is then necessary for ascertaining how specific institu-tional and organisational conditions contribute to specific trajectories ofscientific change
ACKNOWLEDGEMENTS
We would like to thank Steven Casper Ben Martin Renate MayntzCatherine Paradeise Peter Wagner and Peter Weingart who helped bring-ing this volume to fruition by reviewing the papers We are particularlyindebted to Renate Mayntz who participated in our author workshop andhelped to significantly improve several papers We would like to acknowl-edge the help of David Bree Martin Durrell Uwe Schimank JurgenEnders Mary McGee Wood Sarah Moore and Mike Lounsbury
REFERENCES
Amaral A Lynn Meek V amp Marheim Larsen I (Eds) (2003) The higher education man-
agerial revolution Dordrecht Kluwer
Barley S R amp Kunda G (2001) Bringing work back in Organization Science 12(1) 7695
Benninghoff M amp Braun D (2010) Research funding authority relations and scientific pro-
duction in Switzerland In R Whitley J Glaser amp L Engwall (Eds) Reconfiguring
knowledge production Changing authority relationships in the sciences and their conse-
quences for intellectual innovation (pp 81110) Oxford Oxford University Press
Berman E P (2011) Creating the market university How academic science became an eco-
nomic engine Princeton Princeton University Press
Braun D (1998) The role of funding agencies in the cognitive development of science
Research Policy 27(8) 807821
Braun D amp Guston D H (2003) Principal-agent theory and research policy An introduc-
tion Science and Public Policy 30 302308
13Editorsrsquo Introduction
Braun D amp Merrien F-X (Eds) (1999) Towards a new model of governance for universities
A comparative view London Jessica Kingsley Publishers
Drori G S Meyer J W Ramirez F O amp Schofer E (2003) Science in the modern world
polity Institutionalization and globalization Stanford Stanford University Press
Geiger R Meyer J W Ramirez F O amp Schofer E (2005) Beyond technology transfer
US state policies to harness university research for economic development Minerva 43
121
Glaser J Laudel G amp Lettkemann E (2014) Hidden in plain sight The impact of
generic governance on the emergence of research fields Forthcoming In M Merz
P Sormani amp P Biniok (Eds) The local configuration of new research fields On regio-
nal and national diversity yearbook sociology of science Dordrecht Springer
Guston D H (1996) Principal-agent theory and the structure of science policy Science and
Public Policy 23 229240
Guston D H amp Keniston K (Eds) (1994) The fragile contract University science and the
federal government Cambridge MA The MIT Press
Kleinman D L (2003) Impure cultures University biology and the world of commerce
University biology at the millennium Madison WI University of Wisconsin Press
Knorr-Cetina K (1999) Epistemic cultures How the sciences make knowledge Cambridge
MA Harvard University Press
Krucken G Kosmutzky A amp Torka M (Eds) (2007) Towards a multiversity Universities
between global trends and national traditions Bielefeld transcript Verlag
Krucken G amp Meier F (2006) Turning the university into an organizational actor
In G S Drori J W Meyer amp H Hwang (Eds) Globalization and organization
World society and organizational change (pp 241257) Oxford Oxford University
Press
Marginson S amp Considine M (2000) The enterprise university Power governance and rein-
vention in Australia Cambridge Cambridge University Press
Marquis C Lounsbury M amp Greenwood R (Eds) (2011a) Communities and organiza-
tions Research in the sociology of organizations Bingley Emerald Group
Marquis C Lounsbury M amp Greenwood R (2011b) Introduction Community as an
institutional order and a type of organizing In C Marquis M Lounsbury amp
R Greenwood (Eds) Communities and organizations Research in the sociology of orga-
nizations (Vol 33 pp ixxxvii) Bingley Emerald Group
Martin B amp Whitley R (2010) The UK Research Assessment Exercise A case of regulatory
capture In R Whitley J Glaser amp L Engwall (Eds) Reconfiguring knowledge produc-
tion Changing authority relationships in the sciences and their consequences for intellec-
tual innovation (pp 5180) Oxford Oxford University Press
Merz M amp Biniok P (2010) How technological platforms reconfigure science-industry
relations The case of micro- and nanotechnology Minerva 48 105124
Paradeise C Reale E Bleiklie I amp Ferlie E (Eds) (2009) University governance Western
Europe comparative perspectives Dordrecht Springer
Rip A (1994) The republic of science in the 1990s Higher Education 28 332
Schimank U (2005) lsquoNew public managementrsquo and the academic profession Reflections on
the German situation Minerva 43 361376
Schofer E (1998) Science associations in the international sphere 18751990 The realization
of science and the scientization of society In J Boli amp G M Thomas (Eds) World
polity formation since 1875 (pp 249266) Stanford Stanford University Press
14 RICHARD WHITLEY AND JOCHEN GLASER
Simpson I H (2002) Life course patterns of national associations International Sociology
17(2) 285303
Thompson J B (2005) Books in the digital age The transformation of academic and higher
education publishing in Britain and the United States Cambridge Polity Press
Van der Meulen B (1998) Science policies as principal-agent games Institutionalization and
path dependency in the relation between government and science Research Policy 27
397414
Van der Meulen B (2003) New roles and strategies of a research council Intermediation of
the principal-agent relationship Science and Public Policy 30 323336
Van der Meulen B amp Rip A (1998) Mediation in the Dutch science system Research
Policy 27(8) 757769
Weingart P amp Maasen S (2007) Elite through rankings-The emergence of the enterprising
university In R Whitley amp J Glaser (Eds) The changing governance of the sciences
The advent of research evaluation systems Sociology of the sciences yearbook (Vol 26
pp 75100) Dordrecht Springer
Whitley R (2008) Universities and strategic actors Limitations and variations In L Engwall amp
D Weaire (Eds) The university in the market (pp 23 37) London Portland Press
Whitley R (2012) Transforming universities National conditions of their varied organisa-
tional actorhood Minerva 50 493510
Whitley R amp Glaser J (Eds) (2007) The changing governance of the sciences The advent of
research evaluation systems Sociology of the sciences yearbook (Vol 26) Dordrecht
Springer
Whitley R Glaser J amp Engwall L (Eds) (2010) Reconfiguring knowledge production
Changing authority relationships on the sciences and their consequences for intellectual
innovation Oxford Oxford University Press
Ziman J (1994) Prometheus bound Science in a dynamic steady state Cambridge Cambridge
University Press
Ziman J (2000) Real science What it is and what it means Cambridge Cambridge University
Press
15Editorsrsquo Introduction
INSTITUTIONAL CHANGE AND
THE TRANSFORMATION
OF UNIVERSITIES AS
STRATEGIC ACTORS
THE IMPACT OF INSTITUTIONAL
REFORMS ON THE NATURE OF
UNIVERSITIES AS
ORGANISATIONS
Richard Whitley and Jochen Glaser
ABSTRACT
Recent reforms to higher education systems in many OECD countrieshave focused on making universities more effective organisations in com-peting for resources and reputations This has often involved increasingtheir internal cohesion and external autonomy from the state to makethem more similar to private companies However pre-reform universi-ties differed so greatly in their governance and capabilities that theimpact of institutional changes has varied considerably between threeideal types Hollow State-chartered and Autarkic Furthermore thecombination of (a) the inherent uncertainty of scientific research under-taken for publication (b) limited managerial control over work pro-cesses and reputations and (c) the contradictory effects of some fundingand governance changes has greatly restricted the ability of universities
Organizational Transformation and Scientific Change The Impact of Institutional
Restructuring on Universities and Intellectual Innovation
Research in the Sociology of Organizations Volume 42 1949
Copyright r 2014 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN 0733-558Xdoi101108S0733-558X20140000042000
19
to function as authoritatively integrated organisations capable of devel-oping distinctive competitive competences
Keywords Institutional reforms universities ideal types hollowstate-chartered autarkic uncertainty managerial ignorance organi-sational actorhood
INTRODUCTION
Over the past three decades or so many OECD countries have introducedsubstantial reforms to the governance of research universities (ParadeiseReale Bleiklie amp Ferlie 2009) These changes usually involved the devel-opment of new relationships with the state and other extra-mural agenciesas well as shifts in authority relations within academic institutions some-times as part of more general reforms of statesrsquo administrative structuresunder the generic title of lsquoNew Public Managementrsquo or NPM (de BoerEnders amp Leistye 2007 Ferlie Musselin amp Andresani 2009 Schimank2005) Against a common background of reduced public funding of aca-demic research and teaching or at least a substantial reduction in thegrowth rates of such support states have increasingly sought to recon-struct universities as quasi-independent collective agents on whom they canrely to realise national education and research goals Often drawing onorganisational models from the private corporate sector much of therhetoric behind these changes has suggested that they should become moresimilar to companies competing in product capital and labour marketsand lsquoentrepreneurialrsquo (Clark 1998 Marginson amp Considine 2000) Indeedthe corporate model as understood in the United States during much of thefirst half of the 20th century was widely imitated as a formal structure bymany colleges and universities in that country and is being urged upon EUmember countries in the current century (Slaughter amp Cantwell 2012Thelin 2004)
In much of continental Europe and Japan where universities have beenorganised around separate discipline-based institutes that combineresearch and teaching under the authority of full professors this involvestransforming them from arenas in which the heads of these institutes com-peted for control over resources into more internally cohesive and exter-nally autonomous organisations (de Boer et al 2007 Musselin 2007)From being rather fragmented organisations in which their administrators
20 RICHARD WHITLEY AND JOCHEN GLASER
had limited responsibility for or control over the research and teachingactivities their members carried out universities in these countries areincreasingly supposed to have been transformed into organisationally sepa-rate collectivities capable of developing and implementing independentstrategies in pseudo-markets
In more departmentalised academic systems mostly in Anglophonesocieties such as the United Kingdom Australia and New Zealand similarreforms have tended to combine substantial reductions in the level of publicsupport with an intensification of competition between universities forresources and reputations as well as governance shifts that enhanced theauthority of central university managers and boards and the institutionali-sation of a wide array of quality assurance and performance monitoringsystems (Dobbins Knill amp Vogtle 2011 Marginson amp Considine 2000Martin amp Whitley 2010) Many of these changes have been intended totransform predominantly collegiate organisations into more hierarchicallymanaged ones whose senior staff could be held accountable for their per-formance in a comparable way to private companies being disciplined bycompetitive markets
As a number of national case studies have shown however few if anyof such attempts have resulted in universities developing organisationallydistinctive knowledge and capabilities that might enable them to functioneffectively as strategic actors in a comparable manner to private companies(Huisman 2009 Paradeise Reale amp Goastellec 2009) While this could beattributed to inconsistencies in the ways that these policies have been imple-mented in many states andor to time lags between institutional change andorganisational responses it also suggests that there may be some funda-mental difficulties with a corporate model being applied to universities
In particular such models tend to ignore the limited ability of adminis-trators and managers to organise and control scientific research and teach-ing activities that are highly uncertain and more influenced by externalscientific communities than by organisational hierarchies (Musselin 2007Whitley 2008) Since they are poorly qualified to judge the merits of parti-cular research goals and approaches and the results they produce man-agers tend to rely on the decisions and preferences of those conductingresearch and advanced teaching activities about which strategies to pursuewith particular knowledge and skills and how to do so They therefore areunable to coordinate and control them in organisationally specific waysthat could generate distinctive competitive advantages
It is also worth noting that in any case there is no single model of effec-tive corporate governance for universities to imitate but rather a wide
21The Impact of Institutional Reforms on the Nature of Universities
variety of corporate forms and business strategies that have dominated dif-ferent market economies and industries in different historical periods asrecently highlighted by the extensive comparative capitalisms literature(see eg Crouch 2005 Roy 1997 Schmidt 2002 Whitley 1999 2007a)Thus even if they could develop some characteristics of competitive firmsit remains unclear which kind of enterprise would be most appropriate orlikely to emerge in different circumstances
Furthermore since national higher education systems and the natureof universities in them have varied so much in the 20th century wewould expect any set of institutional reforms to have varied consequenceshowever similar they may seem initially Just as overtly similar deregula-tory reforms to financial services markets in Britain and Japan havehad quite different effects because they were introduced into differentlyorganised market economies dominated by different kinds of interestgroups (Kushida amp Shimizu 2013 Laurence 2001) so too New PublicManagement influenced reforms are likely to have varied outcomes acrossdifferent higher education systems (de Boer et al 2007 Schimank 2005)
In this article we explore these points in more detail by considering howthe most common reforms are affecting key features of universities asresearch and teaching organisations and their likely impact on the kinds oforganisations that might emerge in different societies Following a sum-mary of the key features of firms as strategic actors in competitive marketsand how these vary between institutional contexts we consider the extentto which universities in different kinds of higher education systems havediffered significantly in these respects in terms of three distinct ideal typesNext we discuss how the major changes in the governance of universitiesthat have taken place in many OECD countries seem to be affecting firsttheir strategic autonomy and second their management of research andteaching activities and consequently their establishment as coherent dis-tinct and authoritatively integrated organisations in different higher educa-tion systems
THE NATURE OF FIRMS IN MARKET ECONOMIES
Private companies are commonly seen as the critical unit of economicaction in capitalist societies because they are the key collective entitythrough which private property rights holders and their delegated agentscoordinate economic activities to generate and appropriate added value
22 RICHARD WHITLEY AND JOCHEN GLASER
It is the combination of private ownership of the rights to an incomestream from such activities with decentralised decision making by indivi-dual strategic actors and the authoritative integration and direction ofhuman and material resources through organisational routines and proce-dures that makes firms crucial economic agents in such political economies(Penrose 1959 Richardson 1998)
The ability to direct employees to undertake specific tasks through dele-gated authority from private property rightsrsquo holders is central to the orga-nisational development of distinctive competitive competences in capitalistsocieties not least because the flexibility provided by employment agree-ments enables managers to organise economic activities in different waysand to change these to suit altered circumstances (Richardson 1998) Thisflexibility facilitates the management of increasingly complex and uncertainactivities particularly innovation As Lazonick (1991) Lazonick andWest (1998) and others have suggested the planned coordination of a spe-cialised division of labour has enabled firms to build distinctive organi-sational capabilities for developing process and product innovations on acontinuing basis
However it is important to recognise that the powers duties and socio-economic functions of private companies vary considerably betweennational jurisdictions institutional contexts and over time (Milthaupt2003 Morck 2007 Roy 1997) In particular the extent to which legallyconstituted corporate entities combine unified authority limited liabilityinvestor ownership and easily transferred private property rights in a singleorganisational form is highly variable across market economies and issubject to substantial change reflecting the contested nature of the domi-nant corporate form and major institutional differences between stateand regions (DiMaggio 2001 Dobbin 1994 Gourevitch amp Shinn 2005Goyer 2011)
This variety of leading firmsrsquo governance direction and capabilitiesacross capitalist societies suggests that there is no single model of competi-tive companies that can be taken to represent the ideal organisational formfor modernised universities to follow Rather there are many differentkinds of dominant firm types that have become established and changed indifferent institutional environments These vary on a number of dimensionsthat can be summarised under two broad headings (Whitley 2010) Firstthose dealing with issues of ownership control and direction that distin-guish between the different kinds of groups and interests dominating strate-gic decision making and their impact on leading firmsrsquo priorities oftenreferred to as corporate governance (Aguilera amp Jackson 2010) Second
23The Impact of Institutional Reforms on the Nature of Universities
those dealing with the organisation and control of work activities in waysthat add significant value to inputs and generate organisation-specificknowledge and capabilities to provide competitive advantages (Kogut ampZander 1992 Metcalfe amp James 2000 Teece Pisano amp Shuen 2000)
Firm Strategic Autonomy and Governance
Considering first how the governance of leading firms can differ betweensocieties it is important to recognise that governance involves a wide rangeof interests and issues including lsquothe whole set of legal cultural and institu-tional arrangements that determine what publicly traded corporations cando who controls them how that control is exercised and how the risksand returns from the activities they undertake are allocatedrsquo (Blair 1995p 3) In particular any analysis of how the governance of firms varies andchanges has to consider how dominant interest groups and institutionsinfluence the construction direction and development of major companiesin different jurisdictions in such ways that their dominant goals and eco-nomic consequences vary This clearly involves issues of ownership andcontrol but it also includes the impact of various kinds of employees sup-pliers customers competitors and other business associates on firm man-agement and behaviour
While in principle all private companies in decentralised market econo-mies are separate decision making agents able to decide which human andmaterial resources they will use to compete in particular markets and howthey will do so in practice the extent of their independence from lsquoexternalrsquogroups and interests can vary considerably across differently organisedpolitical economies The strategic autonomy of those running firms fromoutside shareholders business partners and employee groups is muchgreater in societies where ownership is fragmented amongst portfolioholders and most business transactions including hiring and firing are con-ducted on an armrsquos length basis than in those where share ownership ismore concentrated capital is provided more by lsquomainrsquo banks than throughcapital markets trade associations are powerful and labour organisationsare significant
In the former case what might be termed isolated hierarchies are ableto pursue the interests of the firm as understood by its managers in rela-tive isolation from other organisations and investors subject to capitalmarket pressures to achieve particular levels of profitability and growth
24 RICHARD WHITLEY AND JOCHEN GLASER
This can enable them to implement radical changes in business strategythrough rapid shifts in employment and asset trading on large and liquidcapital markets as well as facilitating the growth of project-based firmsin high technology industries (Casper 2007 Whitley 2006)
In contrast in the latter kind of lsquoalliance capitalismrsquo (Gerlach 1992)dominant firms tend to be more integrated into a range of commitmentsand alliances with major shareholders sometimes as members of businessgroups business partners banks industry associations and employee groupsWhile such authority sharing with major allies can encourage long-terminvestments in training and building organisation-specific competences inparticular sectors it may also limit the speed with which firms can enter newmarkets and develop radically different technologies
Organising Work and Developing Organisational Capabilities
As many discussions have emphasised the authoritative direction and inte-gration of economic activities is a key feature of firms Hamilton andFeenstra (1995 p 56) for instance claim that firms and economic organi-sations in general are lsquoabove all authoritative organisations that structurerelationships according to established rules of conductrsquo in which partici-pants recognise that they are bound to the authoritative norms of the orga-nisation and there are effective means to enforce collective rules Howevermajor differences exist in how owners and managers use their authority tointegrate and direct economic activities and so develop distinctive kinds ofcompetitive capabilities in different firms
In particular they differ in how much different groups of employeescontribute to organisational problem solving and improving performanceas many comparisons of German Japanese and US companies have indi-cated (Aoki 2001 Cole amp Whittaker 2006 Soskice 1999) Such contribu-tions depend on both the ways in which authority over task performance isshared between management and employees and on employees being com-mitted to developing firmsrsquo organisational capabilities sometimes at theexpense of improving their own individual skills
Two important ways in which owners and managers can elicit high levelsof employee involvement in joint problem solving activities and commit-ment to the improvement of firmsrsquo collective capacity to deal with complexissues are (a) to share substantial amounts of authority with them and(b) to provide long-term organisational careers Authority sharing here
25The Impact of Institutional Reforms on the Nature of Universities
involves property rights holders and their agents delegating considerablediscretion over task performance and sometimes task organisation toskilled employees and encouraging them to contribute to product and pro-cess improvements It varies in the degree of such delegation that is theamount of discretion exercised by subordinates over task performance andproblem solving and its scope that is the range of activities and decisionsover which discretion is exercised
While these aspects are often positively correlated it is clearly possiblefor managers to delegate considerable levels of discretion over specific nar-rowly defined tasks without extending it to more general features of workorganisation and involvement in problem solving On the whole the greateris the degree and scope of such authority sharing the more firms should beable to integrate different kinds of activities and types of knowledge indealing with complex problems and to develop new routines and knowl-edge as in some Danish and Japanese companies (Aoki 1994 Koike1994 Kristensen 1996 Kristensen et al 2011)
The extent of such authority sharing and nature of employee contribu-tions to problem solving are affected by the complexity transparency andpredictability of key tasks (Mintzberg 1979 pp 348379 Simpson 1985)as reflected in many discussions of the professions whose authority overtask performance has been linked to the monopoly of access to knowledgeand lsquoindeterminationrsquo in professional knowledge (Child amp Fulk 1982pp 159162 see also Freidson 1984) In general the more indeterminateare the connections between work processes and outcomes and the moreuncertain is the value of task outcomes the more authority skilled staff canexert over how tasks are performed and organised
Long-term commitment to a firmrsquos success through contributing to itsspecific knowledge and capabilities is additionally encouraged by offeringorganisational careers for those who demonstrably make such contribu-tions on a continuing basis By tying personal futures to the growth ofthe employing organisation and making credible commitments to main-tain employment across the business cycle such careers intensifyemployee commitment to the improvement of collective competenceseven if that limits their visibility on external labour markets Where onthe other hand employment is seen as being vulnerable to market andtechnological shifts as well as to changes in ownership skilled workerswill be more concerned to improve their position on external labourmarkets by enhancing their own personal knowledge reputation andskills than on sharing knowledge and opportunities with short-termcolleagues
26 RICHARD WHITLEY AND JOCHEN GLASER
VARIETIES OF PRE-REFORM UNIVERSITIES AND
CHARACTERISTICS OF FIRMS
From this brief account of the major variable characteristics of firms inmarket economies it is clear that most 20th century universities differed agreat deal from the stereotypical large firm able to make strategic choicesand control work processes They additionally did not have to competewith each other for resources and customers by selling their services onopen markets and have not had to develop distinctive competitive advan-tages through the development of organisation-specific collective capabil-ities They have though varied considerably in the extent to which theyconstructed separate organisational identities and have been able to exertsome independence from the state between different national higher educa-tion systems in OECD countries (Clark 1983 Coleman 1999 Rothblatt ampWittrock 1993)
These differences can be summarised in similar terms to those used todescribe types of leading firms particularly the extent of senior managersrsquostrategic autonomy and the degree to which they are able to exert unifiedauthority over the organisation and conduct of work activities to generatedistinctive collective competences to identify distinct types of pre-reformuniversities At least three ideal types can readily be distinguished thatreflect the major contrasts between higher education systems and theirnational contexts in the 20th century before major institutional changeswere introduced Hollow State-chartered and Autarkic (Whitley 20082012) Their key characteristics are listed in Table 1 and will now be brieflydiscussed
University Strategic Autonomy and Governance
Hollow universities were typically parts of the state administrationAuthority over key decisions and responsibilities was concentrated in thehands of state officials with varying degrees of delegation of authority overappointments and promotions to national disciplinary elites and seniorprofessors in each university Both academic and non-academic staffremained state employees and university administrators had little or noability to alter national terms and conditions of employment
Universities approximating to this ideal type have been significant inmany countries in continental Europe as well as parts of East Asia in
27The Impact of Institutional Reforms on the Nature of Universities
both lsquoGermanicrsquo and lsquoNapoleonicrsquo higher education systems (Dobbinset al 2011 Regini 2011) These lacked or only were able to use to avery limited extent both of the constitutive features of private companiesas key economic actors strategic autonomy and organisational capabilities
Table 1 Governance and Capabilities of Three Ideal Types of Pre-Reform Universities in OECD Countries
Organisational
Governance and
Capabilities
Hollow State-Chartered Autarkic
Autonomy of senior managers from state ministries concerning
Internal
structures
Very limited Medium High
Programmes Very limited Medium High
Student
admission and
examinations
Very limited Medium High
Student numbers Very limited Limited High
Authority over resource (re-)allocation organisation structures and programmes of
State High Medium Low
Senior
management
Low Medium High
Senior academics High Medium Medium
Authority of senior
academic staff over
personnel and
academic decisions
Medium to high
often shared with
national disciplinary
elites and state
ministries
High although senior
managers can exercise
considerable influence
when they cannot agree
Medium to high with
senior managers able to
exercise considerable
influence
Authority of senior
managers over
academic careers
Very limited Limited Medium restricted by
competition for
scientific excellence and
academicsrsquo ability to
obtain external funding
Managerial ability
to develop
distinctive
organisation-
specific capabilities
None Low Low
28 RICHARD WHITLEY AND JOCHEN GLASER
Most of their financial and administrative decisions were made by stateagencies As Graham and Diamond (1997 p 23) point out
hellip permanent civil service officials on campus hellip answered to the central ministry for
finances and major policies on academic programmes In exchange for full funding the
universities served the governmentrsquos definition of the public interest
Senior university managers were typically elected from amongst the pro-fessors for limited periods or appointed directly by the state and expectedto resume their academic careers after completing their turn in office Theyhad little or no strategic independence and very limited freedom to shiftresources between activities subject areas and services let alone to estab-lish new areas of research and teaching or to close existing ones
The dominant organisational unit for research and teaching activities inthese kinds of university was the Chair-based Institute that combined intel-lectual leadership with administrative control of resources and careersunder the direct management of the professor Collaboration and coordina-tion of these activities depended on these professors being willing to worktogether rather than the exercise of organisational authority and so univer-sities were more a relatively amorphous aggregation of discipline-basedinstitutes than coherent integrated organisations
State-chartered universities in contrast had more distinct organisationalidentities separate from state ministries and employed both academic andnon-academic staff on terms that they were able to influence more directlyThe powers and responsibilities of senior managers as collective agentswere enshrined in formal charters from the state that granted them theright to select students decide the content and organisation of teachingprogrammes and examinations and award degrees Usually though thestate retained some control over student numbers through its funding ofplaces and was often the main supporter of academic research bothdirectly and through research foundations
In many cases it was also involved in the establishment and closure ofsubject departments In the United Kingdom for example the UniversityGrants Committee which was the main channel through which state sup-port was allocated to individual universities before the 1980s periodicallycarried out subject reviews that could result in staff and facilities beingmoved between universities (Martin amp Whitley 2010) and the state as themain employer of graduates usually had to agree to the creation of newmedical schools as well as frequently intervening in the provision of teachertraining courses
In many of these kinds of university their charters granted considerableauthority over the provision of degree programmes promoting of staff and
29The Impact of Institutional Reforms on the Nature of Universities
restructuring of departments to academic Senates and their committeessomewhat restricting the ability of senior managers to act independently oftheir senior colleagues Here the role of governing councils and boardswhich usually had a majority of non-academic members in determiningstrategic priorities was limited and vice-chancellors functioned less likechief executives and more like the primus inter pares model of chairmen ina predominantly collegiate culture
While not as hollow as most in continental Europe then state-chartereduniversities and their managers have been quite constrained in their strate-gic choices by state regulations and financial controls on the one handand by academic senates national union agreements and scientific elites onthe other hand Compared to most large companies their strategic auton-omy was limited particularly in countries where central state agencies pro-vided most of their income
In contrast autarkic universities have been much more autonomousfrom state agencies as independent organisations governed by boards oftrustees or university councils without requiring state approval to awardqualifications In contrast to both hollow and state-chartered universitiesautarkic universities have lsquofour essential freedomsrsquo that were enunciated bySupreme Court Justice Felix Frankfurter in a 1957 case (Thelin 2004p 343) lsquothe freedom to determine who may teach what may be taughthow it should be taught and who may be admitted to studyrsquo Howeverthey often depend on accreditation agencies to legitimate their degrees andof course had to attract fee-paying students and private donations to fundtheir activities
In the case of private US universities and more recently many publicones boards of trustees exercised considerable influence over strategicchoices especially the selection of the president who (usually) had muchgreater power over academic staff and the future of the university than hiscounterpart in Europe (Geiger 1986) As Trow (1993 p 292) puts it
In the case of the US hellip almost without exception our colleges and universities have
been created by a group of laymen who selected a president to actually direct and man-
age the day-to-day life of the institution And this founding body and its successorshellip
together with their chosen agent (the president) have occupied the leading positions of
authority and power in American colleges and universities and still do
The lsquocorporate modelrsquo of university governance in which trustees typi-cally saw themselves functioning as the board of directors delegating execu-tive powers to the president in an analogous fashion to the growing UScorporation in the first half of the 20th century became widespread in the
30 RICHARD WHITLEY AND JOCHEN GLASER
inter war period which is perhaps not too surprising given that they tendedto be senior executives themselves (Slaughter amp Cantwell 2012 Thelin2004 p 238)
Organising Work and Developing Organisational Capabilities
Turning next to consider how pre-reform universities differed in their inter-nal management of work activities and development of organisation-specific competences it is worth emphasising the limited extent to whichmanagers and trustees in any country have been able to construct competi-tive advantages in research and teaching through their coordination andcontrol of academic work This is for two major reasons First scientificresearch undertaken to contribute to collective intellectual goals throughpublication of results is inherently highly uncertain and the significance ofresearch results is often contested Researchers themselves often do notknow how their work succeeds or fails and typically are unsure what out-comes will eventuate in any precise sense Indeed in many sciences specify-ing the nature of the problem and potential ways of dealing with it areweakly codified activities that cannot readily be reduced to routine problemsolving procedures and cannot easily be assessed as worthwhile or compe-tent by outsiders Equally the meaning and significance of research resultsare often difficult to decide by specialists let alone by non-professionalsand can be subject to later reinterpretation and re-evaluation by differentgroups as many studies of scientific controversies have shown (see egHallam 1973 Pinch 1986)
Second research universities share authority over research goals andperformance evaluation with scientific communities In competing forscientific prestige and resources based on success in contributing to scienti-fic objectives they typically defer to the authority of national and interna-tional scientific elites in determining research priorities and the standardsby which academics are judged (Glaser 2010) While the extent of suchinternational reputational control of research goals and evaluation stan-dards varies between fields and over time (Whitley 2000) universities com-peting for scientific renown on the basis of their employeesrsquo intellectualcontributions have to accept the collective judgements of competentresearchers throughout the world concerning the nature of important pro-blems and the significance of results
For these reasons the scope and degree of authority sharing differ fromthose in most firms including those whose key tasks have the closest
31The Impact of Institutional Reforms on the Nature of Universities
resemblance to public research organisations namely professional servicefirms private RampD organisations and project-based firms The scope ofauthority sharing is wider because managers have no choice but to shareauthority over goal formulation workflow interdependencies and utilisa-tion of results with employees and their scientific communities Thedegree of authority sharing is also higher because in many cases very fewemployees within the organisation have a monopoly of the necessaryknowledge and skills for defining tasks and developing approaches to sol-ving them Consequently the ability of managers to plan projects and orga-nise the division of scientific labour to achieve organisational goals in acomparable manner to RampD managers in private companies is extremelylimited In this respect universities are qualitatively distinct from most pri-vate companies and indeed many formal organisations
The authority of scientific communities over research goals and perfor-mance additionally restricts universitiesrsquo ability to restructure researchteams and generate new kinds of skills by combining staff in novel waysSince disciplinary elites and gatekeepers collectively determine the nature ofcompetent contributions and research skills universities cannotreconfigure project groups around new kinds of problems in isolation fromsuch actors if they wish to compete for scientific reputations and prestige
As a consequence the high levels of managerial authority sharing withacademics in research and teaching common in universities are largelyunintended and do not necessarily greatly enhance managersrsquo abilities tointegrate different kinds of activities and types of knowledge in a similarmanner to that in many firms On the contrary the nature of researchundertaken for publication as the central organisational task leads to diver-gent and incompatible capabilities between different fields rather than theirintegration into organisation-specific competences The uncertainties inher-ent to such scientific research means that authority over the specification ofgoals is delegated to employees who in turn formulate them with referenceto different scientific communities Quite unlike most firms then managersin universities share authority over their core processes with a great numberof divergent and highly diffuse social contexts
For the same reason long-term organisational careers may not createmuch additional employee commitment to developing organisation-specificknowledge and capabilities in each university Since promotion and salaryincreases often depend on mobility between universities or at least onbeing offered posts elsewhere loyalty to individual organisations tends tobe much less than to onersquos institute and its professor as well as to the scien-tific community at large Thus few if any academics are likely to spend
32 RICHARD WHITLEY AND JOCHEN GLASER
much energy on contributing to the development of university-specific cap-abilities particularly if that is at the expense of their making significantcontributions to intellectual goals
Autarkic universities have had greater powers to alter the range and nat-ure of research and teaching programmes in response to changes in demandand resource provision than did hollow and state-chartered ones often byattracting new staff and investing in new facilities They have also beenable to encourage increased involvement with private companies and thepursuit of research commercialisation through changing incentives andrestructuring employment conditions as Terman did at Stanford (Adams2009) However their opportunities too remain limited While their man-agers can in principle invest in or abandon broadly and somewhat impre-cisely defined fields of research and teaching they have to yield authorityover exactly what is done and how it is done in these fields to the academicsthey recruit
It is worth noting here that state-chartered and autarkic universities havebeen able in principle to employ a wide range of non-academic staff provid-ing services to students acting on behalf of the university administration infund raising public relations activities and other functions under theauthority of senior managers These roles could form the basis for develop-ing distinctive organisation-specific competences that enable particular uni-versities to attract large numbers of students and external resources Themore a competitive market for such inputs develops separately from thestate the more important such competences are likely to become
The limited ability of strategic managers of universities to coordinateand control research and teaching activities highlights the contingent con-nection between high levels of organisational autonomy and managerialcontrol of work activities While variations in these two aspects of organi-sational actorhood are often correlated in private companies though by nomeans always as the examples of professional service firms and manyproject-based firms illustrate (Whitley 2006) this is usually not the case inuniversities
THE IMPACT OF INSTITUTIONAL REFORMS ON
UNIVERSITY AUTONOMY
Turning next to consider the varied ways in which recent funding and gov-ernance reforms are likely to affect different kinds of universities as
33The Impact of Institutional Reforms on the Nature of Universities
organisations and potential strategic actors we now discuss how the keychanges that are taking place in many though not all OECD statesseem to be influencing universitiesrsquo strategic autonomy Essentially thereforms combine moves to make universities more organisationally distinctand well bounded as competing organisations on the one hand withincreasingly direct attempts to steer research and teaching activitiestowards public policy objectives and intensify the monitoring of academicperformance in terms of these sometimes contradictory goals on theother hand
While the more autarkic universities in the 20th century United Stateshave not been so subject to intentional institutional reforms they havenevertheless been changed by developments in their institutional environ-ments The growth of the federal research university in the United Statesand increasing involvement of previously teaching focused organisations inthe national competition for intellectual prestige and extra-mural grants inthe 1950s and 1960s began to restrict the unilateral exercise of presidentialauthority at least in the elite institutions receiving the bulk of such grants(Thelin 2004 pp 271290) From being largely subordinate staff to theexecutive hierarchy successful researchers became more influential as indi-vidual lsquogrant swingersrsquo (Greenberg 1966) who were courted by ambitiouspresidents and increasingly able to command high salaries and supportfacilities As universities became more concerned to compete for scientificreputations on the basis of their employeesrsquo contribution to knowledgethough and researchers were more able to raise project money from exter-nal sources such as state research foundations the ability of managers tocontrol academicsrsquo behaviour has declined
As well as academics becoming more significant in the decisions ofresearch universities in the United States the growing role of national gov-ernment support in helping students gain access to higher education andincreasing state regulation of universitiesrsquo personnel and other policies havebegun to affect their autonomy While then university managers andBoards of Trustees or Regents in the United States remain remarkablyautonomous in their ability to make strategic choices and set priorities forlsquotheirrsquo organisations when compared to those in most OECD countriesthey have become more circumscribed by the interests and influence of pro-fessors and federal politicalbureaucratic organisations as well as by thevariety of accrediting agencies
In the context of institutional reforms in many other OECD countriesmoves to increase the lsquoagencyrsquo of universities included reforms to fundingpractices such as a widespread shift from allocating public funds to
34 RICHARD WHITLEY AND JOCHEN GLASER
specific activities and roles through line budgeting to providing them asblock grants transferring academic and non-academic staff from civil ser-vice status to university employment and enhancing the authority of uni-versity senior managers and supervisory boards at the expense of theprofessoriate Attempts to steer research more directly towards publicpolicy goals have been based on the establishment and strengthening ofresearch evaluation systems (Whitley amp Glaser 2007) the formalisationof various contracts between governments and universities (Dobbinset al 2011 Meier amp Schimank 2010) the steering of public researchcouncils towards socio-economic purposes and the encouragement ofresearch commercialisation
Probably the most widespread shift in the environment of universitiesacross the OECD world since the 1970s has been the decline in the level ofpublic funding of university research and teaching per student relative tothat enjoyed in the postwar decades While this may not have directlyaffected universitiesrsquo strategic autonomy and capabilities in all countries ithas been a major feature of changing state-science relationships in manyand created the opportunity for politicians and bureaucrats to undertakesignificant restructuring of higher education systems as well of course asencouraging managers to seek alternative sources of income more activelythan before (Ballarino amp Perotti 2011)
Where such reductions have been substantial they can be expected toreinforce the consequences of associated changes in universitiesrsquo governanceand organisation particularly when these include the transfer of staff frompublic employment by the state to being employees of individual universitiesAs they become formally responsible for personnel decisions and meetingthe wage bill universities under financial pressure have to act as collectiveagencies and take a more active role than when they were primarily adminis-trative agents of the state Where on the other hand cuts in financial sup-port from the state have not been so marked the impact of many NPMreforms has not been so significant as perhaps is the case in Switzerland asdiscussed by Benninghoff et al in their article in this volume
One of the major changes associated with reductions in public financialsupport in many countries has been an increase in competitive relationshipsbetween universities as employment organisations whether for lsquoexcellencersquo(Weingart amp Maasen 2007) or for contributing to state public policy goalsand their concomitant development of separate collective identities as com-peting quasi-corporate entities While this shift is especially marked insocieties where universities were predominantly hollow organisations aspublic administrative agencies under direct ministerial control the
35The Impact of Institutional Reforms on the Nature of Universities
intensification of competition for resources and reputations has alsobecome a significant factor enhancing organisational identities in countrieswhere universities were already separately chartered organisations thatdirectly employed academic staff
Devolving responsibility to cope with resource scarcity and initiatingcompetitive relationships between universities do of course presume thatuniversities are separate organisations able to control some resources them-selves and make their own decisions about what activities to undertake andhow to do so While the extent of such delegation of resource control and(re-)allocation has often been rather less in practice than some governmentofficials have claimed (see eg Wright amp Williams Oslashrberg 2009) manystates in Continental Europe and Japan have granted university managersgreater discretion over resource utilisation by providing public funding as ablock grant
In theory this could mean that managers are able to shift resourcesbetween programmes and fields as strategic choices alter and circumstanceschange However as vice-chancellors in state-chartered universities havefound this is often easier said than done especially in the short term notleast because states often retain control over student numbers and theirassociated public funding as well as considerable influence over the creationand closing of departments in different fields (Estermann Nokkala ampSteinel 2011) Similarly making academics employees of universities mightalso be thought to enhance the strategic autonomy and powers of managersin hollow universities However since most of their salary costs continue tobe met by government ministries and conditions of service tend to remainquite standardised and set by state agencies managers have had little or noability to establish organisationally specific employment policies and sofar have evinced little interest in doing so in many countries
Combining such increases in managerial discretion with negotiatedlsquocontractsrsquo between ministries and universities whereby the lattersrsquo strate-gic choices are agreed with varying degrees of precision and equalityof influence for several years in advance (Dobbins et al 2011) mightbe seen as equivalent to the owners of a holding company setting targetsfor divisional managers of subsidiaries who are then granted enoughoperational autonomy to take responsibility for meeting them Howevergovernments rarely delegate full operational control of all aspects ofresearch and teaching and university managers have rarely had enoughindependence and authority to deliver agreed objectives in practice(Meier amp Schimank 2010) Furthermore most public policy goals foruniversities are not reducible to a single metric such as return on capital
36 RICHARD WHITLEY AND JOCHEN GLASER
employed or profitability and many have divergent if not contradictoryimplications
In most OECD countries cutbacks in the public funding of academicresearch and teaching and intensified competition between universities havebeen accompanied by increased state interest in monitoring their performanceIn some this has taken the form of lsquostrongrsquo research evaluation systems suchas the United Kingdomrsquos Research Assessment Exercise (Martin amp Whitley2010 Whitley 2007b) which directly link peer assessments of research out-puts to levels of financial support while others have established more indir-ect connections between various indicators of scientific and teaching successand funding decisions (Whitley amp Glaser 2007) While varying considerablyin the extent to which they are detailed intrusive and have significant finan-cial consequences all such evaluations presume that states are able to assessthe collective achievements of universities as distinctly bounded and mana-ged organisations against a variety of objectives and standards Anyincreased autonomy granted to universities is thus balanced by the institutio-nalisation of instruments and procedures for steering their activities towardspublic policy purposes
Such steering of academic research priorities by state agencies has ofcourse become much more feasible as the ability of universities to supportall the costs of scientific research has declined and scientists have had toobtain resources from public foundations and research councils to conducttheir research Over recent decades an increasing number of governmentshave developed formal science and technology policies that established stra-tegic goals for public agencies supporting university science and oriented at least in principle their allocation of resources In many state-chartereduniversities the combination of funding cuts strong evaluation systemsand steering of research support has probably reduced their strategicautonomy as they have to compete more intensively for more limited statesupport
As well as such dependence on public research councils increasing the fea-sibility of state steering of universitiesrsquo research goals it also increases theauthority of scientific elites or at least those chosen to advise these coun-cils on the merits of proposals over the direction of academic prioritiesInsofar as universities wish to establish distinctive research goals as strategicpriorities then they are constrained by both research funding agenciesrsquochoices and elite preferences in different disciplines
Another aspect of changing university identities and activities concernsthe increasing commercialisation of research results and capabilities espe-cially through the establishment of new firms with university support and
37The Impact of Institutional Reforms on the Nature of Universities
active researcher participation and intensified commitment to obtaininguniversity patents In principle the more universities are able to attractextra-mural funding from commercialisation activities the more their man-agers should be able to exercise strategic autonomy which is presumablywhy many of them have been active in seeking legitimacy for such develop-ments and establishing administrative units for exploiting new knowledgeproduced by their staff especially in the United States (Berman 2012)This reflects of course their search for new funding sources as state sup-port declines but also their wish to gain revenues that increased their dis-cretion over resource allocation which traditional academic consultingactivities rarely did
However the extent to which any increase in extra-mural revenues fromproject support and commercialisation efforts does in fact create substan-tially greater organisational independence for universities depends on theamounts thus raised and the terms on which they are made availableWhere project-based funding is accompanied by significant declines in statesupport for universities any competitive success is unlikely to enhance theirautonomy since such income including overhead costs are usually closelytied to specific research activities and is rarely available to university man-agers to use at their discretion Such limitation of their autonomy is evenmore restrictive if the state actively steers the allocation of public researchgrants towards specific purposes as in Australia (Glaser amp Laudel 2007)
Furthermore as block grants decline as a proportion of total universityincome and university dependence on extra-mural project grants growsmanagersrsquo influence over research priorities and ability to support particu-lar programmes of work declines compared to that exercised by individualresearch teams and funding councils The more important scientific reputa-tions become for the perceived success of universities and hence for thepublic performance of their managers and the more they are tied to theprovision of external research support that is largely controlled by principalinvestigators the less authority university leaders are able to exerciseauthority over academicsrsquo choices and how they contribute to universitysuccess
Similar points apply to research commercialisation revenues Most uni-versity services intended to increase these depend on the goodwill and com-mitment of researchers to notify them of possible opportunities andsupport their efforts Typically any resultant income streams have had tobe shared with research teams and their parent departments so that theamounts of discretionary resources accruing to the university as a wholehave tended to be less than is often expected (Colyvas amp Powell 2006
38 RICHARD WHITLEY AND JOCHEN GLASER
Krimsky 1999 Stuart amp Ding 2006) As long as universities are concernedto compete for scientific reputations based on the contributions of theiremployees to collective intellectual goals as determined by scientific elitesrather than by private interests or university managers it is difficult to seehow universities as employers could monopolise the control of commercia-lisation revenues or insist on making academicsrsquo salaries partly dependenton success in such ventures
Overall then funding and governance changes in many OECD coun-tries have probably encouraged universities to establish and reinforce dis-tinctive organisational identities as competing collectivities able to exercisesome discretion over their strategic priorities and allocation of resourcesThis has been particularly noticeable where they have previously been moresimilar to the hollow organisational ideal type However the extent of suchdiscretion remains quite severely limited in many European countries bycontinued state oversight of universitiesrsquo profiles restrictions on ministriesrsquoability to provide resources to fulfil strategic choices and political restric-tions on student numbers fee levels including zero and the ways thatuniversity qualifications are connected to access to elite labour marketsFurthermore universitiesrsquo competitive strategies in the pseudo-markets cre-ated by various state initiatives have been quite strongly constrained by themechanisms established to implement public policy goals and monitor uni-versity performance especially in countries where funding has been severelyreduced
This has been particularly the case in many of the Anglophone societieswhere state-chartered universities have previously exercised greater auton-omy from the government than those in much of mainland Europe andJapan Strong research and teaching evaluation systems state steering offunding council priorities and strict targeting of student numbers and asso-ciated public support amongst other developments have probably reduceduniversity managersrsquo strategic autonomy from the state in such countries
THE IMPACT OF INSTITUTIONAL REFORMS ON
UNIVERSITY ORGANISATIONAL CAPABILITIES
Turning next to consider how these reforms and changes to fundingarrangements are affecting the ability of universities as organisations todevelop distinctive strategic capabilities many were expected to enhancemanagersrsquo ability to develop distinctive organisational goals and to
39The Impact of Institutional Reforms on the Nature of Universities
organise research and teaching activities accordingly In particular thecombination of increased competition for resources and reputations grow-ing managerial influence over the terms and conditions governing employ-ment relations and intensified efforts by the state and other agencies toassess their collective performance and contributions to public policy objec-tives has increased the formal authority of university managers as legiti-mate controllers of resources and priorities (Estermann et al 2011 Regini2011)
In both hollow and state-chartered universities traditionally collegiatepatterns of decision making especially about academic and personnel mat-ters are being replaced by a more top-down managerial exercise of author-ity as universities have been forced to compete more overtly with eachother and become what some have seen as lsquoentrepreneurialrsquo organisations(Clark 1998 Laudel amp Weyer this volume) The role of academic senatesin approving let alone contributing to strategic choices has become con-siderably weakened in countries like Australia (Glaser amp Laudel 2007Marginson amp Considine 2000) while senior administrative posts havebecome more obviously full time and highly paid positions Demands forgreater accountability and performance monitoring on the part of the statehave encouraged many managers to institutionalise formal procedures formeasuring the success of their employees in meeting research and teachingtargets that are summarised as key performance indicators (Glaser LangeLaudel amp Schimank 2010)
Although the combination of state policies focused on making universi-ties more effective as organisations has enhanced the authority of thoserepresenting the interests of each university and formally responsible for itsactivities managersrsquo ability to organise and control research and teachingactivities in distinctive ways to create organisation-specific competencesremains quite limited Since the lsquobaseline authority sharingrsquo in publicresearch organisations reflects properties of the key performance processesrather than managerial strategies it changed far less then could be expectedfrom the growth of managersrsquo formal authority Even in Anglophone socie-ties where university presidents and vice-chancellors have been able toboost their own salaries and perquisites and establish administrative hierar-chies mimicking those of corporate bureaucracies the management anddirection of core academic tasks remain substantially in the hands andbrains of small groups of scientists owing primary allegiance to their intel-lectual communities rather than to their current employer In this respectof course universities are not that dissimilar to many professional serviceorganisations whose central work processes are designed executed and
40 RICHARD WHITLEY AND JOCHEN GLASER
certified by qualified professionals and coordinated on the basis of theirstandardised skills as summarised by Mintzberg (1983) in his account ofprofessional bureaucracies
However universities differ from such professional service organisationsin their commitment to create new knowledge educate students in this newknowledge and in the much greater intrinsic technical uncertainty of theircore activities The commitment to contributing intellectual innovations inthe public sciences from incremental puzzle solving to radical reshaping ofcurrent orthodoxies means that the problems to which academics applytheir professional skills the skills themselves and the results of applyingtheir professional skills to particular problems are much lesspredictable and standardised than would be expected in most professionalservice organisations or craft-based firms (Whitley 2006)
Since the choice of problems to investigate formulation of approachesand evaluation of results is primarily decided by academics and their scien-tific communities these activities require much more continuous negotia-tion and immediate adjustment with academic colleagues throughout theworld typically through formal communication media workshops andconferences and informal communication than is the case in most profes-sional services As a result not only do university managers have to sharesubstantive control over most work processes with professional employeesthey additionally surrender the coordination and integration of researchactivities and their outputs to extra-organisational communities and theirintellectual elites (Whitley 2011)
The embeddedness of research and much advanced teaching in extra-organisational contexts that provide the dominant frames of reference foracademicsrsquo work and the fluid and constantly shifting nature of theseframes of reference render some of the tools for controlling professionalwork ineffective Thus the formalisation and standardisation of skills(Freidson 1984 pp 1318) is limited here as scientists constantly reinventthem in conducting their research Supervision by colleagues is also largelyineffectual due to different specialisations in most cases and occupationalcontrol by clients (Child amp Fulk 1982 pp 167175 Freidson 1984pp 78 Simpson 1985 pp 425426) is difficult because in many casesthe only lsquoclientsrsquo are other members of the scientific communitiesPerformance measurements based on citation counts and similar indicatorscan be interpreted as an attempt to exploit scientific communities in thisparticular role
While then managers may be able to influence the selection of scientificstaff and rely on peer judgements of the merits of their work in making
41The Impact of Institutional Reforms on the Nature of Universities
promotion decisions and allocating discretionary resources once universi-ties become direct employers of academics their ability to control thechoice of work goals the allocation of professional labour time and skillsand the evaluation of task performance is inherently limited and subsidiaryto continuing peer group assessments and research activities Essentiallytheir construction of organisation-specific competences based on the collec-tive commitment of expert staff to organisational objectives and the devel-opment of organisation-specific knowledge remains highly uncertainindirect and reactive to othersrsquo judgements
The transition from recurrent to project-based funding that can beobserved in all countries has further and ambivalent effects on the authoritysharing in universities Since peer review of competing applications forresearch funding remains the dominant means of allocating increasinglyscarce resources researchers have become even more dependent on theircommunities As a result it has become more difficult for most researchersto investigate problems that are regarded by senior colleagues as technicallyinfeasible as some physicists considered the experimental realisation ofBose-Einstein condensates (Laudel et al this volume) This means thatresearchersrsquo authority over their work is reduced in favour of scientific com-munities which thereby assume even more authority vis-a-vis universitiesAt the same time researchers who acquire grants have often the authority toindependently control them which gives them an additional base of author-ity and further reduces the authority of managers who control the shrinkingrecurrent funding
This is exemplified by the impact of major research grants from theEuropean Research Council on some universities as discussed by Edleret al in their article in this volume Because these are unusually largegrants often over five years that carry substantial prestige they are muchsought after by both individuals and universities The amount durationand prestige of such a grant produce a substantial influence over universi-tiesrsquo allocation of resources including posts and potentially enabling themto affect the long-term development of departments and faculties particu-larly the balance of activities and staff in different specialisms andtechnologies
In order to meet increasing performance expectations under conditionsof limited control over research processes some university managers areattempting to construct portfolios of lsquohigh performingrsquo research teams andinstitutes They select fields of research for which they provide an interna-tionally competitive infrastructure and attempt to hire leading researchersof these fields in the hope of achieving improved results as a lsquoreturn on
42 RICHARD WHITLEY AND JOCHEN GLASER
investmentrsquo The reshaping of research profiles of universities is similarlyoften concerned with creating lsquocritical massrsquo which is increasingly seen as aprerequisite for high performance (Schiene amp Schimank 2007 Laudel ampWeyer this volume) The resources for these exercises are reallocated fromfields deemed less promising in terms of research performance which oftenatrophy as a result as in the case of evolutionary developmental biology inthe Netherlands (Laudel et al this volume) However as the contributionby Laudel and Weyer to this volume demonstrates neither the effects ofinvestments nor the consequences of closing departments are easy to predictand plan and managersrsquo lack of detailed knowledge of the wide range ofsciences covered by most leading universities means that academic sciencecan end up being driven by short-term fashions and resource availability
CONCLUDING REMARKS
This discussion has highlighted a number of points about the impact ofrecent institutional reforms on the nature and characteristics of researchuniversities that are perhaps worth summarising in conclusion Firstdespite the popularity of the corporate model as an appropriate governancestructure for universities in the United States and more recently elsewherethere remain substantial barriers to universities becoming strategic actorscompeting in decentralised markets in a comparable manner to privatecompanies There are good reasons to believe that even with continuinghigher education reforms neither their governance and strategic autonomynor their ability to organise work and develop strategic capabilities willachieve the same levels as those common in private companies
Second although many OECD governments have delegated some opera-tional autonomy to universities and taken some steps to increase thepowers and accountability of senior managers relative to institute headsand professors this has rarely amounted to granting them substantial self-steering capabilities particularly where they were similar to hollow organi-sations (Estermann et al 2011) As long as the state remains the primaryprovider of funds for education and research in universities whetherdirectly through supporting undergraduate teaching or indirectly throughstate-backed loans and project support governments are unlikely to giveup responsibility for and influence on the operation of the higher educa-tion system the financing of students and how academic qualificationsaffect entry to elite labour markets Indeed the more it becomes widely
43The Impact of Institutional Reforms on the Nature of Universities
accepted that universities are core institutions of the so-called knowledgeeconomy the less likely states are to grant them high levels of strategicautonomy
Rather it seems more probable that third as universities do becomemore formally separate employment organisations with their own manage-rial hierarchies and governance arrangements states will increasinglyenmesh them in elaborate multifaceted performance monitoring regimesintended to ensure that they continue to contribute to public policy goalswhile at the same time competing intensively for resources and reputationsAs is happening in some Anglophone jurisdictions reduced public supportfor higher education has enabled governments to steer many state-chartered universities more towards policy priorities reducing both thediversity of their strategic choices and their ability to follow distinctiveorganisationally specific patterns of development While managers maysuperficially appear to have gained greater independence from state minis-tries tight government organisation of the rules of the competitive gamefor resources can so constrain their options as to restrict severely their stra-tegic actorhood
Fourth the externally guided nature and fluidity of research goals inher-ent uncertainty of work processes and contestable assessments of the valueof results in these organisations make it difficult if not impossible for man-agers to direct and control core activities effectively Not only do intellec-tual communities exert more authority over research and much teachingthan do employment organisations they also organise the coordination andthe integration of results with certified knowledge on a continuing basis
This also means that granting universities and their managers moreautonomy from state control on the presumption that they will be able tomeet demanding performance targets greatly overestimates their powersand capabilities Even where states actually do delegate substantial opera-tional autonomy to university managers and enhance their internal powersrelative to senior professors this in no way leads inevitably to their effec-tive control of research and teaching activities such they can realistically beheld responsible for each universitiesrsquo success or failure in achieving agreedobjectives Strategic independence need not and in universities usuallydoes not imply effective managerial determination of who carries outwhich activities in what ways that collectively contribute to overall organi-sational purposes
Thus while higher education reforms increase universitiesrsquo strategicautonomy and managerial control of university structures and resourcesuniversitiesrsquo ability to compete as cohesive organisations producing
44 RICHARD WHITLEY AND JOCHEN GLASER
distinctive contributions to knowledge pseudo-markets is severely limitedUniversity managers are unable to exercise formal authority over theiremployeesrsquo activities on the basis of delegated powers from owners in asimilar manner to most formally constituted private companies in marketeconomies Furthermore insofar as such reforms do actually promote stra-tegic autonomy but are unable to generate organisation-specific strategiccapabilities they appear to initiate and promote a bifurcation of the twoconstitutive elements of organisational actorhood
REFERENCES
Adam S B (2009) Follow the money Engineering at Stanford and UC Berkeley during the
rise of Silicon Valley Minerva 47 367390
Aguilera R V amp Jackson G (2010) Comparative and international corporate governance
The Academy of Management Annals 4(1) 485556
Aoki M (1994) The Japanese as a system of attributes A survey and research agenda
In M Aoki amp R P Dore (Eds) The Japanese firm the sources of competitive strength
(pp 1140) Oxford Oxford University Press
Aoki M (2001) Toward a comparative institutional analysis Cambridge MA MIT Press
Ballarino G amp Perotti L (2011) Italy Gradual changes and an uncertain autonomy
In M Regini (Ed) European universities and the challenge of the market (pp 168182)
Cheltenham Edward Elgar
Berman E P (2012) Creating the market university How academic science became an eco-
nomic engine Princeton Princeton University Press
Blair M (1995) Ownership and control Washington DC Brookings Institution
Casper S (2007) Creating silicon valley in Europe Oxford Oxford University Press
Child J amp Fulk J (1982) Maintenance of occupational control The case of professions
Work and Occupations 9 155192
Clark B (1983) The higher education system Academic organization in cross-national perspec-
tive Berkeley CA University of California Press
Clark B (1998) Creating entrepreneurial universities Organizational pathways of transforma-
tion Oxford Pergamon Press
Cole R amp Whittaker H (2006) Introduction In H Whittaker amp R Cole (Eds) Recovering
from success Innovation and technology management in Japan (pp 128) Oxford
Oxford University Press
Coleman S (1999) Japanese science View from the inside London Routledge
Colyvas J amp Powell W (2006) Roads to institutionalization The remaking of boundaries
between public and private science Research in Organizational Behavior 27 305353
Crouch C (2005) Capitalist diversity and change Recombinant governance and institutional
entrepreneurs Oxford Oxford University Press
de Boer H Enders J amp Leisyte L (2007) Public sector reform in Dutch higher education
The organizational transformation of the university Public Administration 85 2746
DiMaggio P (Eds) (2001) The twenty-first-century firm Princeton NJ Princeton University
Press
45The Impact of Institutional Reforms on the Nature of Universities
Dobbin F (1994) Forging industrial policy The United States Britain and France in the rail-
way age Cambridge Cambridge University Press
Dobbins M Knill C amp Vogtle E A (2011) An analytical framework for the cross-country
comparison of higher education governance Higher Education 62 665683
Estermann T Nokkala T amp Steinel M (2011) University autonomy in Europe II The score-
card Brussels European University Association
Ferlie E Musselin C amp Andresani G (2009) The lsquosteeringrsquo of higher education systems A
public management perspective In C Paradeise E Reale I Bleiklie amp E Ferlie
(Eds) University governance Western European comparative perspectives (pp 120)
Dordrecht Springer
Freidson E (1984) The changing nature of professional control Annual Review of Sociology
10 120
Geiger R (1986) To advance knowledge The growth of American research universities
19001940 Oxford Oxford University Press
Gerlach M (1992) Alliance capitalism Berkeley CA University of California Press
Glaser J (2010) From governance to authority relations In R Whitley J Glaser amp
L Engwall (Eds) Reconfiguring knowledge production Changing authority relationships
in the sciences and their consequences for intellectual innovation (pp 357370) Oxford
Oxford University Press
Glaser J Lange S Laudel G amp Schimank U (2010) The limits of universality How field-
specific epistemic conditions affect authority relations and their consequences
In R Whitley J Glaser amp L Engwall (Eds) Reconfiguring knowledge production
Changing authority relationships in the sciences and their consequences for intellectual
innovation (pp 291324) Oxford Oxford University Press
Glaser J amp Laudel G (2007) Evaluation without evaluators The impact of funding formu-
lae on Australian university research In R Whitley amp J Glaser (Eds) The changing
governance of the sciences The advent of research evaluation systems Sociology of the
sciences yearbook (Vol 26 pp 127152) Dordrecht Springer
Gourevitch P amp Shinn J (2005) Political power and corporate control The new global poli-
tics of corporate governance Princeton NJ Princeton University Press
Goyer M (2011) Contingent capital Short-term investors and the evolution of corporate gov-
ernance in France and Germany Oxford Oxford University Press
Graham H D amp Diamond N (1997) The rise of the American research universities
Baltimore MD John Hopkins University Press
Greenberg D (1966) Grant swinger Reflections on six years of progress Science 154
14241425
Hallam A (1973) A revolution in the earth sciences From continental drift to plate tectonics
Oxford Oxford University Press
Hamilton G amp Feenstra R C (1995) Varieties of hierarchies and markets An introduction
Industrial and Corporate Change 4 5191
Huisman J (Ed) (2009) International perspectives on the governance of higher education
London Routledge
Kogut B amp Zander U (1992) Knowledge of the firm combinative capabilities and the
replication of technology Organization Science 3 383397
Koike K (1994) Learning and incentive systems in Japanese industry In M Aoki amp R Dore
(Eds) The Japanese firm Sources of competitive strength (pp 4165) Oxford Oxford
University Press
Krimsky S (1999) The profit of scientific discovery and its normative implications Chicago
Kent Law Review 75(1) 1539
46 RICHARD WHITLEY AND JOCHEN GLASER
Kristensen P H (1996) On the constitution of economic actors in Denmark Interacting skill
container and project coordinators In R Whitley amp P H Kristensen (Eds) The chan-
ging European firm Limits to convergence (pp 118158) London Routledge
Kristensen P H Lotz M amp Rocha R (2011) Denmark Tailoring flexicurity for changing
roles in global games In P H Kristensen amp K Lilja (Eds) Nordic capitalisms and glo-
balization New forms of economic organization and welfare institutions (pp 86140)
Oxford Oxford University Press
Kushida K amp Shimizu K (2013) Syncretism The politics of Japanrsquos financial reforms
Socio-Economic Review 11 337369
Laurence H (2001) Money rules The new politics of finance in Britain and Japan Ithaca NY
Cornell University Press
Lazonick W (1991) Business organization and the myth of the market economy Cambridge
Cambridge University Press
Lazonick W amp West J (1998) Organizational integration and competitive advantage
Explaining strategy and performance in American industry In G Dosi D J Teece amp
J Chytry (Eds) Technology organization and competitiveness (pp 247288) Oxford
Oxford University Press
Marginson S amp Considine M (2000) The enterprise university Power governance and rein-
vention in Australia Cambridge Cambridge University Press
Martin B amp Whitley R (2010) The UK research assessment exercise A case of regulatory
capture In R Whitley J Glaser amp L Engwall (Eds) Reconfiguring knowledge produc-
tion Changing authority relationships in the sciences and their consequences for intellec-
tual innovation (pp 5180) Oxford Oxford University Press
Meier F amp Schimank U (2010) Mission now possible Profile building and leadership in
German universities In R Whitley J Glaser amp L Engwall (Eds) Reconfiguring
knowledge production Changing authority relationships in the sciences and their conse-
quences for intellectual innovation (pp 211236) Oxford Oxford University Press
Metcalfe J S amp James A (2000) Knowledge and capabilities A new view of the firm In
N Foss amp P Robertson (Eds) Resources technology and strategy Explorations in the
resource based perspective (pp 3152) London Routledge
Milthaupt C J (Ed) (2003) Global markets domestic institutions Corporate law and govern-
ance in a new era of cross-border deals New York NY Columbia University Press
Mintzberg H (1979) The structuring of organizations Englewood Cliffs NJ Prentice Hall
Mintzberg H (1983) Structure in fives Designing effective organizations Englewood Cliffs
NJ Prentice-Hall
Morck R (Ed) (2007) A history of corporate governance around the world Chicago IL
University of Chicago Press
Musselin C (2007) Are universities specific organizations In G Krucken A Kosmutzky amp
M Torka (Eds) Towards a multiversity Universities between global trends and national
traditions (pp 6384) Bielefeld transcript Verlag
Paradeise C Reale E Bleiklie I amp Ferlie E (Eds) (2009) University governance Western
European comparative perspectives Dordrecht Springer
Paradeise C Reale E amp Goastellec G (2009) A comparative approach to higher education
reforms in Western European countries In C Paradeise E Reale I Bleiklie amp
E Ferlie (Eds) University governance Western European comparative perspectives
(pp 197226) Dordrecht Springer
Penrose E T (1959) The theory of the growth of the firm New York NY Wiley
Pinch T (1986) Confronting nature The sociology of solar neutrino detection Dordrecht
Reidel
47The Impact of Institutional Reforms on the Nature of Universities
Regini M (Ed) (2011) European universities and the challenge of the market Cheltenham
Edward Elgar
Richardson G (1998) Some principles of economic organisation In N Foss amp B Loasby (Eds)
Economic organisation capabilities and coordination (pp 4462) London Routledge
Rothblatt S amp Wittrock B (Eds) (1993) The European and American University since 1800
Cambridge Cambridge University Press
Roy W G (1997) Socializing capital The rise of the large industrial corporation in America
Princeton N J Princeton University Press
Schiene C amp Schimank U (2007) Research evaluation as organisational development The
work of the academic advisory council in Lower Saxony (FRG) In R Whitley amp
J Glaser (Eds) The changing governance of the sciences The advent of research evalua-
tion systems (pp 171190) Dordrecht Springer
Schimank U (2005) New public management and the academic profession Reflections on
the German situation Minerva 43 361376
Schmidt V A (2002) The futures of European capitalism Oxford Oxford University Press
Simpson R L (1985) Social control of occupations and work Annual Review of Sociology
11 415436
Slaughter S amp Cantwell B (2012) Transatlantic moves to the market The United States
and the European union Higher Education 63 583606
Soskice D (1999) Divergent production regimes Coordinated and uncoordinated market
economies in the 1980s and 1990s In H Kitschelt P Lange G Marks amp J Stephens
(Eds) Continuity and change in contemporary capitalism (pp 101134) Cambridge
Cambridge University Press
Stuart T amp Ding W (2006) When do scientists become entrepreneurs The social structural
antecedents of commercial activity in the academic life sciences American Journal of
Sociology 112 97144
Teece D Pisano G amp Shuen A (2000) Dynamic capabilities and strategic management In
G Dosi R R Nelson amp S G Winter (Eds) The nature and dynamics of organiza-
tional capabilities (pp 334362) Oxford Oxford University Press
Thelin J R (2004) A history of American higher education Baltimore MD Johns Hopkins
University Press
Trow M (1993) Comparative perspective on British and American higher education In
S Rothblatt amp B Wittrock (Eds) The European and American University since 1800
(pp 280299) Cambridge Cambridge University Press
Weingart P amp Maasen S (2007) Elite through Rankings The Emergence of the enterpris-
ing university In R Whitley amp J Glaser (Eds) The changing governance of the
sciences The advent of research evaluation systems Sociology of the sciences yearbook
(Vol 26 pp 75100) Dordrecht Springer
Whitley R (1999) Divergent capitalisms Oxford Oxford University Press
Whitley R (2000) The intellectual and social organization of the sciences (2nd ed) Oxford
Oxford University Press (first edition 1984)
Whitley R (2006) Project-based firms New organisational form or variations on a theme
Industrial and Corporate Change 15 7799
Whitley R (2007a) Business systems and organisational capabilities Oxford Oxford
University Press
Whitley R (2007b) Changing governance of the public sciences In R Whitley amp J Glaser
(Eds) The changing governance of the sciences The advent of research evaluation sys-
tems Sociology of the sciences yearbook (Vol 26 pp 327) Dordrecht Springer
48 RICHARD WHITLEY AND JOCHEN GLASER
Whitley R (2008) Universities as strategic actors Limitations and variations In
L Engwall amp D Weaire (Eds) The university in the market (pp 2337) London
Portland Press
Whitley R (2011) Changing governance and authority relations in the public sciences
Minerva 49 359385
Whitley R (2012) Transforming universities National conditions of their varied organisa-
tional actorhood Minerva 50 493510
Whitley R amp Glaser J (Eds) (2007) The changing governance of the sciences The advent of
research evaluation systems Sociology of the sciences yearbook (Vol 26) Dordrecht
Springer
Wright S amp Williams Oslashrberg J (2009) Prometheus (on the) rebound freedom and the
Danish steering system In J Huisman (Ed) International perspectives on the govern-
ance of higher education (pp 6987) London Routledge
49The Impact of Institutional Reforms on the Nature of Universities
EMPOWERMENT OF FRENCH
UNIVERSITIES BY FUNDING
AND EVALUATION AGENCIES
Christine Musselin
ABSTRACT
This article questions how institutional change influences actorsrsquo behaviorwithin organizations affected by the evolution of their institutional envir-onment This issue is addressed by looking at how university leaders areempowered by the external reviews led by evaluation agencies andresearch councils and how they use these reviews as managerial tools andto make decisions It is argued that this process is complementary to thereforms in university governance and structures and amplifies theireffects because it is more legitimate favors organizational coupling andthe appropriation of new norms It draws on a study led in three Frenchuniversities in 2011
Keywords Empowerment external peer review funding and evalua-tion agencies institutional change organizational coupling universitygovernance
Organizational Transformation and Scientific Change The Impact of Institutional
Restructuring on Universities and Intellectual Innovation
Research in the Sociology of Organizations Volume 42 5176
Copyright r 2014 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN 0733-558Xdoi101108S0733-558X20140000042002
51
INTRODUCTION
Change has become a major issue for neo-institutionalists both sociologi-cal (Padget amp Powell 2012) and historical (Streeck amp Thelen 2005) buttheir reflections focus more on the transformation of fields than on howinstitutional change influences actorsrsquo behavior within the organizationsaffected by the evolution of their institutional environment This articleaddresses this lacuna in analyzing universities as examples of professionalbureaucracies Universities in many countries have undergone rather strongreforms aimed at strengthening their managerial capacities and transform-ing them into rationalized organizations (Ramirez 2006) but it remainsunclear how far and under which conditions such changes have been effec-tive rather than symbolic or hypocritical (Brunsson 1989) Building on thecase of French university academic managers I will argue that the intro-duction of managerial norms in universities faced a legitimacy deficit whichcould be overcome if they were able to use professional norms to makedecisions and I will show that these professional norms are produced bythe new state evaluation and funding agencies that construct external peer-based reviews for use as management tools by university leaders
In a paper published in 2007 Richard Whitley (2007) stressed the role ofthe academic elite in charge of assessing research or teaching activities orresearch projects in evaluation agencies or research councils Following thisperspective I looked (Musselin 2013) at the impact of this evolution forthe academic profession I concluded that instead of a dismissal of the aca-demic profession there is a reinforcement of the academic control throughthe role of these academic elite and the peer-reviews it produced But thereis also more differentiation The ldquohavesrdquo that is academics rewarded bythe elite or being part of it have resources enabling them to negotiate withtheir institution and are therefore stronger today while the ldquohave notsrdquo arein a worse situation than before as their failures in terms of professionalreview leave them without resources vis-a-vis managerial control In otherwords the role of the academic profession is stronger because peer-review-based judgment becomes more important but this can conflict with theneeds of institutionsrsquo management by increasing differentiation amongacademics
This article will elaborate on these points by looking more closely athow university leaders are empowered by the external reviews led byevaluation agencies and research councils I will argue that this process iscomplementary to the reforms in university governance and structures andamplifies their effects because it is more legitimate favors organizational
52 CHRISTINE MUSSELIN
coupling and the appropriation of new norms In other words the institu-tional empowerment of universities relies as much on professionals (aca-demics) as on managers and builds more on external pressures than oninternal reshaping of structures because the reinforced role played by exter-nal research councils and evaluation agencies is a powerful driver for inter-nal evolution
It draws on a 2011 study of three French universities1 (Musselin 2012)Students of a the master of sociology program at Sciences Po conductedinterviews in three universities One located in Paris and heavily specializedin health and sciences and two in the French regions one focused on socialsciences and humanities and one that is pluridisciplinary2 I will call themrespectively UniSciences UniSSH and UniPluri About 100 interviewswere held with central university managers (presidents vice-presidentsregistrars and their main collaborators) deans heads of department andheads of labs as well as with academics and administrative staff elected tosit in one of the three deliberative bodies (university council academiccouncil and council for training and students affairs) and are listed inTable 1 Because of the voluntary focus of this study on academic andadministrative staff directly involved in the management of their institu-tion this paper focuses on change in behaviors for this specific populationbut does not consider their effects on academic practices
I will start with a discussion of the literature on organizational changein general and in higher education studies in particular (1) I will go
Table 1 Breakdown of Interviews Held with University Managers
UniSciences UniSSH UniPluri
Academic university managers (presidential team) 4 7 5
Administrative managers (central administration) 12 5 7
Deans heads of departments or labs 15 11 16
Administrative managers in faculties 4 4 3
Members of deliberative bodies with no other managerial
responsibilitiesa2 5 2
Total 37 32 33
aSome of the interviewees of the other categories were also members of these bodies and inter-
viewed simultaneously for their function as administrative or academic leaders and elected
members of bodies
53Empowerment of French Universities
on with an overview on the main recent French reforms and payspecial attention to the creation of a national agency for the evaluation ofteaching and research (AERES Agence drsquoevaluation de la Recherche et delrsquoEnseignement Superieur) as well as a national research council (ANRAgence Nationale de la Recherche) after the 2006 act The next three sec-tions will be organized around the transformations induced by these twoagencies I will show that the assessment process that simultaneouslyinvolves the evaluation of teaching programs the evaluation of researchunits and the evaluation of the governance of universities favors morecentralization and more coordination within universities These processesare furthermore reinforced by the fact that resources and reputations arelinked to them
I will then argue that the results of this peer-review-based assessment like the allocation of grants by the ANR or the budgets linked to theperformance assessed by the AERES enabled universities to allocateresources on a less egalitarian basis and to close down or reconfigureresearch units and teaching programs They also encouraged the introduc-tion of incentives or other new devices in order to improve their resultsI will finally show that these are not only game playing and tactics inorder to address external demands the norms and expectations devel-oped by the peers sitting in the evaluation agency and the research coun-cil are appropriated by many of the interviewees even when they criticizethem this in turn facilitates the use of external reviews as a managementtool by university managers In a concluding section lessons will bedrawn from this case about the role of external bodies on the transfor-mation of universities and the complex links between organization andprofession
ORGANIZATIONAL CHANGE AND
TRANSFORMATION OF UNIVERSITY GOVERNANCE
Looking at Change in Organizations from Within
As stressed by Greenwood and Miller (2010) many recent developments inorganization theory have not only moved away from the study of organiza-tional design but also from the study of organizations from withinAnalysts primarily focused on the institutional environment of organiza-tions (or fields) and looked at how it exercises pressures for homogeneity
54 CHRISTINE MUSSELIN
and pushes organizations in the same field to adopt the same forms(DiMaggio amp Powell 1983) They also developed explanations for changeby looking at the emergence and spread of new institutional logics(Thornton amp Ocasio 2008 Thornton Ocasio amp Lounsbury 2012) withina field leading to the transformation of the organizations belonging to thisfield The crucial role of institutional entrepreneurs in such processes wasfurthermore highlighted by many studies (Battilana Leca amp Boxenbaum2009 DiMaggio 1988 Greenwood amp Hinings 1996) The idea that com-peting logics may exist in a field and that actors are fighting one againstanother to impose their logic rather than another has also been discussedand led to the notion of strategic fields (Fligstein amp McAdam 2012) thatstresses the importance of power relations and skilled actors (Fligstein2001) for change to occur
While very effective in understanding how organizations developing simi-lar activities evolve (for instance how accounting firms became transnationalprofessional service firms as studied by Greenwood amp Suddaby 2006) neo-institutionalist approaches were less interested in looking at the impactor even the effectiveness of these transformations within organizationsFor some of them (Meyer amp Rowan 1977) this is legitimized by the factthat organizations follow institutional scripts working as rationalized mythsorganizations conform to the scripts but practices are not affected becausescripts and practices are loosely coupled Others like Fligstein (1985 1990)are more attentive to the power relations within organizations this authorfor instance considered the relationships between the main functions offirms and observed a shift of weight from production to marketing andmore recently to financial departments overtime but he was neverthelessmore interested in the impact of this shift in power for the conception ofcontrol dominating the field than in the concrete interplays and effects itmay have within the firms
Nevertheless more attention has been recently paid to what happenswithin organizations in a given field Greenwood and Miller (2010) forinstance explain how the evolution of the field of professional service firmslead to change in their organizational design and describe this shift as theresult of neo-institutionalist processes but also for more functionalistreasons Nevertheless they limit their analysis to this change in formalstructures and therefore take for granted the effectiveness of this organiza-tional design they do not consider that it might be diverted by the agencyof the actors working in these formal structures In contrast some authorssuggest putting more light on how actors confronted with a new organiza-tional design coming from an institutional change react to these evolutions
55Empowerment of French Universities
They developed the notion of ldquoinhabited institutionsrdquo (Hallett amp Ventresca2006) to describe how agency affects institutional transformations Insteadof loose coupling or passive acceptation of change they show processeswhere individuals are described as resisting adopting or interpreting thenew structures imposed on them
Conflicts around institutional logics the role of institutional entrepre-neurs the notion of strategic fields or the idea of inhabited institutions allbring some agency back into institutions and highlight the need to have acloser look at the concrete impact of institutional change on actors andhow they adopt and appropriate change but also adapt and react to itwithin the organizations of the concerned field
Following this perspective I would like to show how and under whatconditions institutional change affects practices and norms within theorganizations impacted by such change I will focus on the specific case ofuniversities because they experienced reforms aimed at transforming therelationships between professional and organizational powers by strength-ening the organizational capacities of higher education institutions andincreasing their control over professionals Did the development of a man-agerial logic in universities affect the way university managers work andmake decision
Change in University Governance
The huge literature about university governance converges on one point allauthors agree that universities are experiencing major changes and areurged to become more governed introduce managerial tools and practicesissued from firms reinforce their strategic and decisional capacities andinteract more closely with their stakeholders in order to take societaldemands better into account Such trends are exogenous they are imposedon universities and on academics For some authors this result from theimplementation of public policies inspired by new public managementandor network governance (Braun amp Merrien 1999 Paradeise RealeBleiklie amp Ferlie 2009) and for others from the diffusion of transnationalscripts introducing the model of a rationalized university (Ramirez 2006)all over the world
Whatever their origins and local implementations these evolutions sharein common the objective of building more autonomous universities andrely on quite similar strategies to reach this aim transferring to universitiesresponsibilities and domains of decision previously held by public
56 CHRISTINE MUSSELIN
authorities reinforcing the executive power to the detriment of deliberativeand collegial bodies transforming the relations between universities andthe academic staff into an employeremployee relationship introducingstakeholders in the university governance reinforcing hierarchical powerdeveloping rules and procedures in order to manage the decision-makingprocesses and make them more effective etc (Braun amp Merrien 1999 DeBoer Enders amp Schimank 2007 Musselin 2005 Paradeise et al 2009)
As a result many of the reforms of the last decades aim at modifyingthe structures and decision-making bodies of universities the Qualityreform in Norway (Oslashstergren amp Stensaker 2007) the MUB in theNetherlands (De Boer 2002) the LRU in France the Law n240 ofDecember 30 2010 in Italy (Reale amp Primeri 2013) etc All modified thecompetencies of the university president the composition and attributionsof the deliberative bodies and increased the scope of responsibilities of uni-versity managers Building on the stimulating paper written by Brunssonand Sahlin-Andersson (2000) on the construction of public services intoorganizations one can argue that universities experienced a process of con-struction boundaries hierarchy and rationality that is were constructedinto organizations (De Boer et al 2007 Krucken amp Meier 2006 Musselin2007 Whitley 2008)
Nevertheless the impact of the reforms aimed at transforming universi-ties into organizations by modifying their structures and governance isoften less effective than expected One explanation lies in the fact that theyface and are slowed down by the organizational specificities that character-ize higher education institutions (Musselin 2007)
The loose coupling nature of teaching and research that is the possibi-lity to teach a class (or to work on a research project) without stronglycooperating with onersquos direct colleagues is a key organizational specificityof universities As a result top-down hierarchical forms of management arerather ineffective because the loosely coupled nature of academic activitiesweakens the robustness and connectiveness of the ties between the differentunits and between individuals This lack in hierarchical authority is evenstronger in countries where university presidents cannot appoint their inter-mediary managers in France deans3 heads of department and head ofresearch units are elected for two to five years As shown by StephanieMignot-Gerard (Mignot-Gerard 2006 Musselin amp Mignot-Gerard 2002)for France there is a rather strong gap between the ldquocentrerdquo (the directorsof the central administration and the presidential team) on the one handand the rest of the university (periphery) on the other The deans and theiradministrative staff do not feel (or rarely feel) solidarity with the center
57Empowerment of French Universities
and there is no relation of subordination neither between the president andthe deans nor between the registrar and the administrative staff located inthe faculties In order to face this situation presidents often built parallelstructures mobilizing individuals they trust (Barrier and Mignot-Gerard2013) and thus favor and develop top-down collegiality (Lazega ampWaterbled 2010[2009]) parallel to the formal structures but such informaland interpersonal ties lack legitimacy
A second organizational specificity lies in the fact that teaching andresearch are unclear technologies difficult to describe to prescribe and toreplicate Furthermore for such technologies the relationships betweencauses and effects are complex it is for instance hard to demonstrate thatstudents are successful because they have been taught that or that subjectand this way or another rather than because they were rightly selectedThis unclear nature of the technologies involved in teaching and researchhinders the intervention of university leaders on those activities as one doesnot precisely know how they work or how to improve them or make themmore efficient The power of decision therefore lies in the hands of each dis-cipline characterized by its own norms practices and values (Becher1989) university leaders are considered as incompetent in deciding forchange in a training program (or a research agenda)
These organizational specificities can be found in other organizationsbut they are rarely (if not never) simultaneously present From this point ofview they make universities distinctive from other organizations but alsodistinguish them among other professional bureaucracies In hospitals forinstance professionals are more tightly coupled to achieve their activitiesand technologies of production (surgery treatments anesthetics etc) areeasier to grasp (describe prescribe and reproduce) The simultaneous pre-sence of these two organizational characteristics is crucial in higher educa-tion institutions because they strongly impact on the ways universitymanagers can develop their role and exercise their leadership as they can-not use the same ways as in other productive organizations One centralproblem for university managers to exercise authority and use hierarchicalpower is therefore their lack of legitimacy even when they are elected
In such a situation it is not surprising to observe that university presi-dents rarely extensively use the decisional power they have on paper In thisvolume Richard Whitley and Jochen Glaser also stress the difficulties facedby university managers to control and direct core activities Personal char-isma is often a better source of influence than the formal prerogativesprovided by the law In the study she led on French universities Mignot-Gerard (2006) showed that the strength recognized in some university
58 CHRISTINE MUSSELIN
presidents was always very personalized and first linked to their capacity indeveloping a vision and convincing their staff by repeating this vision onany occasion She furthermore showed that instead of selecting or differen-tiating these presidents would first of all enroll and integrate key academicsbecause of their own lack of legitimacy in imposing principles of selection
The increase in formal authority and decisional capacity attributed touniversity presidents by the reforms are therefore often not sufficient forthem to exercise more control on professionals and develop more manage-rial forms of steering because they lack legitimacy to do so
Building on the case of France I will now argue that they are able toovercome this lack of legitimacy when external peer-reviews are availableand can be used to support their decisions and increase their governancecapacity In the French case the concomitant introduction of an evaluationagency and a funding council and the passing of an act increasing the for-mal prerogatives of university leaders facilitated the effective use of theirnew margins of maneuver by these university leaders After a presentationof these two reforms I will show how academic and managerial powers arecombined rather than opposed in order to govern French universitiesmore strongly
FRENCH HIGHER EDUCATION REFORMS THE
LOPRI (2006) AND THE LRU (2007)
Within the last years two acts aimed at transforming the French higher edu-cation system were passed First in 2006 the LOPRI (act for research andinnovation) led among other measures to the creation of a nationalresearch council (ANR) and a national evaluation agency (AERES)Second in August 2007 three months after the election of Nicolas SarkozyValerie Pecresse the new Minister of higher education and research passedthe LRU act (Loi relative aux Libertes et Responsabilites des UniversitesFreedom and responsibility for universities act) that aimed at transformingthe governance of French universities in two ways by increasing the decisio-nal span of university presidents and by delegating new competencies toFrench universities including the management of their payroll and positionsthat were until then run by the ministry
In the newspapers and in debates the LRU is more often mentionedand discussed than the first one because the 2007 act was one of the firstdecisions made under the new presidency of Nicolas Sarkozy The
59Empowerment of French Universities
Minister Valerie Pecresse presented her law as a decisive rupture building anew area for the French higher education system But looking more closelyat this act it appears that it accelerated and extended but did not greatlymodify the direction set by previous acts (the 1968 Faure Act and the 1984Savary Act) or previous public policies (such as the introduction offour-year contracts between each university and the ministry since thebeginning of the 1990s Musselin 2004[2001]4) as the previous acts theLRU aimed at strengthening and increasing the institutional and manage-rial autonomy of French universities
The LOPRI is in many ways more controversial and disruptive Amongthe different new devices and instruments set by this law the more contestedwere and still are the ANR and the AERES
The ANR was created in 2005 (one year before the LOPRI that officiallygave it birth) The idea was to create an institution close to the GermanDFG (Deutsche Forschungsgemeinschaft) or the American NSF (NationalScience Foundation) able to selectively fund research projects that areeither blue sky research (the programmes blancs) or are selected throughthematic calls for proposals The agency covers all disciplines and is incharge of running the different calls Before the ANR this function wasspread between different institutions and among them between nationalresearch institutions such as the CNRS5 (National Center For ScientificResearch) and the Ministry for higher education and research The loss ofthis funding function was of course not welcome by research institutionsthe budgets they managed before 2005 were allocated to the ANR TheANR in fact centralized within one single institution different project-basedforms of funding that previously existed but the ANR is also more thanthis sum First of all the ANR received more resources than its differentpredecessors and thus accelerated the already on-going process towardmore project-based research Second the ANR became the main operatorof the different highly competitive calls for proposals (Labex Equipexetc) launched in the framework of the Grand Emprunt6 (httpwwwemprunt-national-2010fr) A large public fund was initiated to promotefuture investments and much of it (22 billions of 35 as a whole) was dedi-cated to research and universities The idea was to generously fund a smallnumber of excellent labs (Labex) excellent research equipment (Equipex)excellent higher education institutions (Idex) etc In order to run all theseprograms the ANR furthermore developed new procedures to make deci-sions but they all quite heavily rely on peer review
The AERES agency for evaluation was also created by the LOPRI andstarted to work at the beginning of 2007 It is also the product of the
60 CHRISTINE MUSSELIN
centralization of former institutions and of the introduction of new devicesIt is in charge of evaluating every four years (five now) the activity of allresearch units all training programs and all higher education and researchinstitutions Its missions cover what was previously achieved respectivelyby the scientific councils of the national research institutions academicsappointed by the ministry as experts and the national council for the eva-luation of universities (CNE Conseil national drsquoevaluation des universites etdes etablissements) But like the ANR the AERES is not just the sum ofwhat previously existed It also introduces some new practices All evalua-tions are available on a website (which was only the case for the evaluationof universities before) and until 2011 research labs received a global graderanging from A+ to C7 Last but not least the results of these evaluationsare at least partly used as performance-based funding mechanisms
The staff of the AERES consists of administrative personnel who orga-nize the visits of the experts on campus design templates etc and of aca-demics nominated by the management of the AERES who are in charge ofselecting the experts discipline by discipline but also decide the criteria andprocesses of evaluation Some unions and associations of academics suchas SLU (Save universities) and SLR (Save research) developed strong criti-cisms of the AERES especially against the fact that the experts evaluatingthe research units are no longer designated through elections (and listsoften controlled by the unions) as was the case before when the CNRS orthe INSERM evaluated their labs for instance but nominated by theAERES itself
The creation of these two agencies and the LRU can easily be seen as aproof of the increasing influence of NPM in France after a long period ofresistance to this doctrine and its recipes until the 2000s (Bezes 2008) Theministry is supposed to focus on global strategies and delegate moreoperational tasks to some agencies or directly to more operational units(universities) that receive more autonomy to achieve their tasks More com-petition selective processes and performance-based mechanisms wereintroduced with the new formula for budget allocation called SYMPAthat was developed by the French Ministry about 20 of the budget isperformance-based (one part depending on four-year contracts and theother on automatic allocation based on the number of masters degreesawarded or the AERES grades obtained by the research labs etc)
But the impact of the new agencies is not restricted to their activities andto the missions they received from the state Attention should also be paidto their more indirect effects on the academic profession on the one handand on the management of universities on the other (Musselin 2013)
61Empowerment of French Universities
In this article I will in particular insist on the use of external peer reviewas a management tool by academic leaders Building on the study led inthree French universities in May 2011 I will identify more precisely howthey use external evaluation to manage their institution and how profes-sional and managerial powers are combined
EXTERNAL PEER REVIEW ENHANCES
CENTRALIZATION AND COORDINATION
WITHIN UNIVERSITIES
In French universities centralization of decision-making has been a majorprocess in the recent years This is in line with the objectives of the LRUthat sought to concentrate decision-making and responsibilities at the topof universities The radical increase in university budgets after they receivedthe management of their payroll8 and became responsible for their employ-ees is a further reason for centralization As in many countries that alsoincreased the managerial autonomy of their universities (KruckenBlumel amp Kloke 2013) the number of highly qualified staff grew in finan-cial services and human resources departments thus concentrating at thetop level more qualified individuals able to deal with the decisions and pro-blems raised by the new responsibilities The simultaneous implementationof new budgetary software (called SIFAC) amplified this trend The admin-istrative managers of universities used this opportunity to restrict the num-ber of staff able to enter and work with data in this software by trainingonly some of them in this new instrument They thus better control thequality of the data managed in SIFAC but also control the level of infor-mation accessible to different categories of staff
If not the cause for more centralization the ANR and the AERESfacilitated and increased this trend They first of all technically reinforcedcentralization As a matter of fact no application can be sent to the ANRwithout the approval of the institution that will run the money (the univer-sity or a national research institution if the research units is also recog-nized9 by one (or more) of these and prefers the money to be run by one ofthem) even if no approval is refused it provides the university with infor-mation it did not systematically get before (especially on those who appliedand failed) Similarly the procedures developed by the AERES are alsofavoring centralization The activity reports and projects prepared by theresearch units and training programs must be uploaded on the website of
62 CHRISTINE MUSSELIN
the AERES by the university administration and thus the latter can bettercentralize and control what is sent
But the more important driver for centralization is that getting ANRgrants and being positively evaluated by the AERES have become impor-tant signals of the quality of the institution Communicating about successwith the ANR calls has become crucial and universities are more and moreconcerned with diffusing information about their success Directions forresearch have been created at the university level in order to help academicsapplying for the ANR (or the European Research Council) to pushresearch units to answer to call for proposals and to concentrate informa-tion about who answers and who does not who is successful and who isnot etc
The attention paid to the ANR even increased when it became the mainoperator of the calls for proposals (Labex Equipex etc) launched for theGrand Emprunt (public fund) The time constraints of the process but alsothe high stakes it represented for many institutions led university presidentsto play a major role in identifying the projects that will be submitted to thejuries of the ANR (Mignot-Gerard 2012) instead of letting bottom-upprocesses work for the emergence of research projects University leaderstook the lead and detected the projects to support They of course did nottake the risk to promote obscure research units and relied on those with astrong scientific reputation according to the AERES evaluation and thosewho were successful with the ANR But they chose among them which tosupport and thus by-passed the usual decision-making processes to keepit in their hands even if they built on scientific reputations to legitimatetheir choice
I alerted the president because he is a friend of mine I told him ldquoif you further support
this Labex you take a riskrdquo I reminded him that the deliberative bodies have not been
informed He told me he did but after checking I saw it was not the case It was not
So there is a lack of democracy hellip [] There was a kind of deal between the president
and the vice-president who talked together and decided one with another that the uni-
versity needs this Labex and they opted for it The choice of the Labex is the choice of
the vice-president because he rightly saw there was a scientific opportunity there It
was not discussed by the university bureau until last Monday it was not discussed at
the scientific council and a fortiori not by the university council but by contrast ()
all the labs concerned by the Labex were already involved in the discussion You see
we are walking on the headrdquo (Director of a lab UniSSH)
Reputation and resources are also crucial drivers for centralization inthe case of evaluation processes The introduction of performance-basedfunding even if quite limited in scope leads universities to improve their
63Empowerment of French Universities
assessment records both in order to get more resources and also to bemore attractive Because the three processes (assessment of research unitsassessment of training programs and assessment of university governance)are now led at the same moment for each university they are easier tointerconnect and coordination is also easier to achieve
The central administration and the academic leadership of the three uni-versities under study intervened in different ways First they were involvedin the preparation of the reports and projects labs and training programssent to the AERES At UniSSH and UniScience the presidential team clo-sely supervised the documents prepared for the training programs in orderto be sure that they would have a chance to get a good evaluation and thenbe accredited by the ministry
We knew that some universities sent rubbish to the AERES () The AERES did not
appreciate receiving rubbish () In 2006 when I saw the documents on my desk that
were to be sent to the AERES they were really like rubbish The training programs are
good and the professors are great and at the end this is what counts But a decent uni-
versity cannot send rubbish And it was rubbish They did not understand it was neces-
sary to provide figures () With the vice-president we quickly came back in August
and we worked on the documents (Central administration UniScience)
The first impact of the evaluation was anticipation Otherwise it is difficult to speak of
any concrete impact But the offer in training programs was built by anticipation We
tried to send the best possible projects We worked a lot beforehand and I am not sure
all this work would have been done without all these procedures (Member of the coun-
cil for teaching and student affairs UniSSH)
In some cases it became an opportunity to imagine more innovativeprojects and think of new collaborations But again presidential teams usedthe assessment led by the AERES to justify their position internally whileusing the result of the assessment as an argument in the discussion with theministry In the example related in the following quotation we see that thevice-president refused taking the risk of sending a last minute new versionof the project and suggested finalizing a new version to be tested with theAERES during the on-campus visit Because the new version had been wellreceived on that occasion the president was then legitimate to support itinternally and externally
In history they proposed two masters One rather traditional and another focused on
contemporary history with some history of art and social sciences It was very interest-
ing but the organization of the program was not very clear Nevertheless after a few
months of discussions they did not accept to move on that I finally said ldquook I do not
think it will work like that but letrsquos try and send itrdquo One week before the deadline set
by the AERES they came back to me and said ldquowe thought about it again and would
64 CHRISTINE MUSSELIN
like to change our projectrdquo I told them ldquowait it is already in the pipeline there is no
way to change everything a week before the deadline you should have thought of it
beforerdquo (hellip) I told them ldquosend it as it is and use the time before the visit on campus of
the AERES to work on an alternative project and letrsquos speak of it againrdquo Of course the
AERES asked questions about the organization I was sure they would but we could
answer ldquoabsolutely you are completely right but we already have a new proposalrdquo
And I already sent it to the Ministry to start the discussion (Member of the council for
teaching and student affairs UniSSH)
University presidents furthermore used the opportunity of the AERESevaluation and the preparation of the reports to impose common rules to allIn the following example we see how a vice-president pushed for the samerequirement in the study of foreign languages in all programs of the univer-sity including those of the small department in law hosted by UniScience(despite its specialization in sciences and medicine)
In law the colleagues did not respect the requirements We are required to provide
information for the RNCP national registry on professional competencies All
this for them is hellip The university had also decided that we should help our
students to pass the TOEIC Getting competencies in a foreign language seems
unavoidable A professor in law told me ldquoforeign languages do not count We
should not carerdquo Aware of that the vice-president decided to nevertheless include
in their project ldquowe encourage students to pass the TOEICrdquo So my direction
entirely revised their document for the AERES This is so true that they could not
recognize their document when it came back from the AERES (Central administra-
tion UniScience)
Second some universities prepared the evaluation of the AERES byorganizing a mock assessment before the visit of the AERES UniSciencewas one of the first institutions10 assessed by the AERES in 2007 and wasin a way ldquostruck by surpriserdquo but the two others belong to later waves ofevaluation and had time to learn from the first experiences In both casesthey asked colleagues from other places to evaluate them in order to tryand improve their reports and projects beforehand At UniSSH this con-cerned teaching as well as research while in UniMulti only the researchunits were concerned by the mock exercise In each case the aim is toimprove the quality of the activity report that will be sent to the AERESbut it also makes one and each sensitive to what is expected from them andto the evaluation criteria they should be aware of
We tried to do our best to meet the requirements of the AERES We did it in two steps
We had a pre-evaluation organized by the university where they hellip brought hellip what
they called in English a laquo visiting committee raquo including a former member of the
AERES Mister X (Director of a research unit UniMulti)
65Empowerment of French Universities
Third the three universities exercised control on the answers written bythe evaluated after they received their AERES reviews and used the ldquorightto replyrdquo the AERES provides to each director of lab training programand institution these answers are published on the website of the AEREStogether with the reviews At UniSSH one of the university vice-presidentrewrote these replies in order to make them more positive than defensiveand very nicely explained what he tried to do
None of the reviews was a scandal But we intervened on the answers prepared by the
research units For instance one of the colleagues did not understand anything The
AERES was saying that this unit welcomes international visiting scholars for too
short periods of time And the colleague answers ldquowe welcome international scholars
for short periods because our location is not suitable for longer stayrdquo So I told him
ldquodo you think you will get international PhD candidates if you say that your building
is not suitable Say that you are developing a solid dynamics of international rela-
tionships and that you already collaborate with this and this country and that you
aim at welcoming post-docs and researchersrdquo He was developing a narrow answer
while it was easy to reverse the critics and to say he was about to expand the interna-
tional collaborations for his lab That is the work I have done for almost all answers
with more or less reformulation Because this is public this will be read looked at so
the answers to the AERES must become a way to attract people I told them ldquoyou
will apply for an ANR call and what will the experts do They will look at the evalua-
tion you got from the AERES So be careful and do not worsen your case in your
answer to the AERESrdquo So we somewhat subverted the process (Vice-President
UniSSH)
During the preparation of the evaluation and after it the universityleaders are therefore very active in trying to avoid bad assessments toharmonize some practices and provide the best image they can of theirinstitutions This occurs through more centralization and control overinformation this accentuates and builds on the centralization processalready favored by the LRU
EXTERNAL PEER REVIEW ALLOWS LESS
EGALITARIAN ALLOCATION OF RESOURCES
AND RESTRUCTURATIONS
Centralization and coordination of information in the preparation of theevaluation and application processes are only two of the effects of theAERES and ANR on institutional management The results of these pro-cesses are also used to make decisions within universities They provide uni-versity managers with the legitimacy they lack and become references forallocating resources and conducting restructurings In these ways the
66 CHRISTINE MUSSELIN
assessments and decisions issued by AERES and the ANR are used asmanagement tools by the presidents and empower them they allow themto mobilize and put into action the increased decisional capacities offeredby the LRU This is particularly clear when one look at decisions about theresearch policy of the three universities under study
First because these results are produced by peer review they are legiti-mate Of course they might be contested and there exist many protestsagainst the AERES and the type of evaluations it produces as well asagainst the increased differentiation favored by the ANR but these pro-tests are limited enough for internal decisions to be made on the basis ofthese results at least in the three universities under study although one ofthem is known for its strong participation in social movements and demon-strations The allocation of the university operational research budget tothe different research units of the university is a good example of that
Since the LRU universities are responsible for the distribution of thisbudget They could have decided to attribute to each lab the amount itpreviously received from the ministry but in the three universities consid-ered here mechanisms of allocation have been discussed and accepted bythe scientific councils In all cases they weighted the amount of budgetaccording to the AERES grade (A+ A B or C) and developed a for-mula The former budget of the lab was multiplied by more than 1 inresearch units getting a A or a A+ and less than 1 for those with a CDifferent universities have though adopted some different strategies AtUniMulti (see below) they rewarded quite clearly the best and sanctionedthe ldquoCrdquo while at UniSSH they rather tried to help the ldquoCrdquo to improvethemselves but in the three cases and in most other places in Francethe rules of allocation of the research budget took the grade intoaccount
The research in this university is now organized in four research institutes They are
managerial instruments and are thematic-based To allocate funding the university uses
the same algorithm as the ministry and applies it to each institute taking into account
the number of academics etc (Question what are precisely the criteria) I do not know
precisely but the number of academics the number of research active individuals the
number of people with a second thesis (habilitation a diriger des recherches) are taken
into account I do not know precisely On top of that we introduced a supplementary
parameter taking into account the grade given by the AERES to each lab Last year
this multiplying parameter was of 14 if a lab got a A and 16 if they got a A+
(Director of a research lab UniMulti)
In all cases too this differentiation was pushed by the presidential staffbut was agreed upon by their respective academic council11 Their membersvoted in favor of the new principles of budget allocation
67Empowerment of French Universities
Second we observed that evaluation and research grants were also usedto justify restructuration In two of the three universities (UniScience andUniMulti) research units getting a C were reorganized some were sup-pressed and their members had to integrate new teams or develop new pro-jects At UniScience even a B could lead to a restructuration a researchgroup that received a B was ldquoaskedrdquo to join another lab despite the protestof the research group leader who felt they will lose the benefits of the goodrelationships they had built with the former lab Such difficult decisions arenot limited to research but may also concern teaching The creation of newtraining programs receiving a critical evaluation was also abandoned
Some colleagues wanted to create a professional master with a potential on the non
academic job market for historians Some colleagues thought it would be booming
But the evaluation of the AERES was not very positive We did not go further The
AERES was not positive the project was not well enough prepared to convince them
(hellip) If we couldnrsquot convince the AERES how could we convince students employers
professionals Maybe we will come back to it later on We considered it was probably
not ripe (Head of department UniSSH)
The perspective of getting highly selective grants may also justify restruc-turing As mentioned above the ANR was in charge of the differentcompetitive call for proposals (Labex Equipex etc) launched in the frame-work of the Grand Emprunt University presidents were very much involvedin the process and centralized the preparation of the answers they used thisoccasion to suggest mergers or justify reorganizations Meeting the criteriafixed by the ANR was a justification for some internal adaptations andarrangements Even if they remain marginal such interventions are new onthe French scene Furthermore in France and by contrast with the profil-ing policy imposed on German universities restructurings are not onlyinternal to universities but also involve strengthened relationships betweenuniversities grandes ecoles and local units of the national research institu-tions that are all invited to if not merge at least coordinate into jointregional structures
Last but not least the external pressures led by the evaluations and thecompetition for grants justified the development of internal tools aimed atimproving the performance of the university or its conformity with externalrequirements The notion of research active staff used12 by the AERESmade visible the low level of publications of some academics or their pre-ference for publication in rather peripheral journals In some cases this ledto avoidance strategies co-authorship with less research active academicswas encouraged while some labs created a category of ldquoassociated staffrdquofor research passive academics not to be counted as staff of their research
68 CHRISTINE MUSSELIN
unit This also led some university presidents to imagine incentives aimedat transforming research passive staff into research active membersUniMulti for instance developed an internal call for proposals explicitlyaimed at research passive academics selected projects received someresearch money and benefited from a decrease in teaching duties
They developed a fellowship on a competitive basis to allocate 3000 euros to faculty
members who do not publish enough and who were offered 50 less teaching So they
had to write a research project and each lab could present its candidates (Director of
the department for research UniMulti)
Many evidences and decisions made therefore confirm the attention pro-vided by university leaders to the decisions and judgments produced by theANR and the AERES but also the use of these external reviews as manage-ment tools supporting internal decisions They legitimate the choices andorientations university leaders are formally empowered to make since theLRU passed
A RATHER LARGE APPROPRIATION OF THE
NORMS AND EXPECTATIONS DEVELOPED
BY THE EVALUATION AGENCY AND
THE RESEARCH COUNCIL
One should not too quickly conclude that peer-review-based decisionsmade by the AERES and the ANR only influence and empower the presi-dential team To put it differently they are not only used as managementtools The reason for that is that the AERES and the ANR do not onlyproduce decisions (on grants) and evaluations they are also producers ofnorms They contribute to the definition of what is a good research projectwhat is a good research unit or training program what are the journalswhere academics of this discipline should publish (Pontille amp Torny 2010)the right activities a research unit should be involved in the indicators atraining program should concentrate on how a research project should bemanaged and split into deliverables etc The ldquoreference documentrdquo pro-duced by the AERES to explain how the evaluation of research units willbe led is a typical example for the normative work produced by such agen-cies to ldquotellrdquo what is right and what is expected
This makes the audience and the reception of the activities of the ANRand of the AERES much larger than the presidential team and their influence
69Empowerment of French Universities
goes beyond the rather instrumental use of these results to make difficultdecisions The interviews with the deans the heads of departments and thedirectors of research units showed that they might be critical of the currenttrends but have integrated them into their own priorities and preoccupationsAs observed by Camerati (2014) in UK departments the critics are not (orrarely) oriented toward the promotion of scientific meritocracy or evaluationin itself but much more toward the way it is led the constraints it creates orthe time it takes But overall with a few exceptions the prevalence ofresearch over teaching and the increased attention paid to publications israther well accepted and even promoted When asked about their role asdirectors or deans they often spontaneously mention they have to take careof the scientific production This of course facilitates the use made by theuniversity leaders of the external reviews and recognized norms
I do not push my colleagues into more teaching because I want them to have time to
develop their research agenda I canrsquot impose them not to invest in teaching but I can
dissuade them and say be careful for your career Managing degrees will not help your
career (Dean UniScience)
They also develop a discourse about the inevitable transformation of theacademic world and the inevitable adaptation that their colleagues shouldgo through They for instance consider that their role is to encourage theircolleagues to publish in visible and reputed journals
You might still publish a very important paper in a third tier journal I say it again we
have amateurs We have colleagues deliberately publishing good papers in third tier
journals They publish in journals diffusing 50 exemplars and run by a friend of them I
do my best Slowly they come to respect what is asked We canrsquot ask for resources if we
do not evolve Twenty years ago publishing in a top journal was not a must Everyone
knew about what was out There was a specific form of evaluation based on the fact
that everybody knew everybody But it is over We canrsquot count on that anymore
Colleagues must say in which journals they publish (Director of lab UniSSH)
These interviewees all the more accepted the on-going transformationin university governance when their success in getting grants or goodevaluations was high On the one hand and on the basis of their goodresults they can negotiate more support from the university leaders but aresimultaneously less dependent on them because they get external resourcesthrough the ANR grants what they get from their own institutions is lesscrucial and they can also get further external support (from the Region forinstance or from firms looking for partners) thanks to their good reviewsThere are thus empowered twice vis-a-vis their university and vis-a-vistheir external environment
70 CHRISTINE MUSSELIN
The more critical interviewees were those that were not positively evalu-ated andor were not able to get grants They generally developed concep-tual arguments against project-based research the evaluations led by theAERES and performance-based funding but their critical stand was notonly ideological It is also positional and linked to their increased depen-dence on the resources provided by their own university while at the sametime their institution more and more conditions the allocation of resourcesto their adaptation to the new ldquorules of the gamerdquo There is from this pointof view a strong contrast between the social sciences and the sciences atUniMulti or between the humanities and the social sciences at UniSSHmany of the former trying to resist while the latter were more prompt tocomply even when they disagree with the recent evolutions
CONCLUSION
In professional bureaucracies managers lack legitimacy to introduce man-agerial practices because professional norms prevail and resist The capacityof universities in lowering the impact of the organizational reforms they areimposed is linked to this deficit in legitimacy but also to the organizationalcharacteristics specific to higher education institutions that is teaching andresearch rely on loosely coupled ties and unclear technologies This reducesthe possibility for presidents to rely on classical ways of governance (hierar-chy rationalization of processes setting of boundaries etc)
In this article I argued that the increasing margins of maneuver used byFrench university leaders and managers are not directly (or only) linked tothe new capacities offered by the LRU act but to the legitimacy they gainby relying on the external reviews produced by evaluation and fundingagencies In other words French university presidents are not strongerbecause the LRU gave them more hierarchical power they are strongerbecause they can justify their decisions with the external peer-based reviewsproduced by the ANR and the AERES These evaluation and fundingagencies thus play an important role in empowering the management ofuniversities as their reviews are used as management tools by university lea-ders who built on them to make decisions allocate budgets restructureThey legitimate the use of the margins of maneuver and decisional capaci-ties provided by the reforms in university governance They make suchreforms more effective somewhat increasing the level of coupling and coor-dination within universities without nevertheless changing the fundamental
71Empowerment of French Universities
loose-coupled nature of academic activities and the unclearness of the tech-nologies they rely on
This relies on a complex but effective collusion between university lea-ders managing universities and an academic elite created by public authori-ties in order to achieve reviews and evaluations that these public authoritieswill use to make their own decisions (whom to fund with research moneywhich research projects should be funded what training program should befurther supported)
The deal relies on the legitimacy university leaders gain vis-a-vis theacademic staff by using the external reviews produced by this elite Thisallows the university leaders to mobilize and have an effective use of the for-mal organizational capacities they received from structural reforms It pro-motes the diffusion and appropriation of the norms and preferences set bythe academic elite This process does not only empower university managersbut also impacts the definition formalization and spread of academic norms
The combination (rather than the opposition) of academic and manage-rial powers seems therefore to be a condition for the effectiveness ofstructural reforms in the governance of universities that is for institutionalpressures for change to be coupled with change in governance practiceswithin organizations This finally sheds lights on the interplay between thefield of higher education institutions and the field of peer-review-basedagencies and suggests that such interplay between fields might be moreimportant for change than isomorphic processes within one field
NOTES
1 This study was then completed with a questionnaire sent to the same kind ofsample in all French universities This survey confirmed the trends toward centrali-zation the development of indicators and the increasing importance of researchactivities and productivity (Chatelain Mignot-Gerard Musselin amp Sponem 2012)in French universities but does not tell as much on behaviors and practices linkedto the reforms under study here and will therefore not be used
2 In France most universities are not pluridisciplinary After the 1968 act thatfollowed the studentsrsquo movements French universities were reorganized and in bigcities they split into different institutions that often specialized into disciplinarydomains (Sciences for some sciences and medicine for others or Humanities andsocial scienceshellip) Cf Aust (2005)
3 Deans for instance are elected for a five-year mandate It is renewable but asin France this function is neither well recognized nor powerful it is rare to accept asecond mandate rather than to go back to usual academic work University presi-dents are elected for a four-year mandate renewable once
72 CHRISTINE MUSSELIN
4 In France since the Revolution the ldquofacultiesrdquo (facultes) not the universitieswere the main pillars of the French university system until 1968 From this point ofview the French universities as we know them today are very recent and were cre-ated after 1968 This explains their rather low institutional consistency as it is onlyrecently (with the introduction of the four-year contracts at the beginning of the1990s) that the Ministry recognized universities (and not the faculties) as their maininterlocutors The LRU is reflecting this evolution toward stronger institutions Toread more about the (recent) emergence of French universities as institutions seeMusselin (2004 [2001])
5 In France national research institutions have been created in order to com-pensate for the lack of research in French universities The CNRS was the first tobe created in 1936 More than 11000 researchers in all disciplines are currentlyemployed by the CNRS
6 This is very similar to the German Exzellenzinitiative that also aimed at mas-sively funding some highly selected projects (Leibfried 2010) but there are majordifferences between the two processes Letrsquos point at only three In Germany supple-mentary budgets are allocated for a defined period of time there are only three calls(one for scientific clusters one for graduate schools and one for institutional pro-jects) universities are the main level at which answers are developed In France theGrand Emprunt initiative relies on the allocation of an endowment and the avail-able budget is derived from the interest gained from this capital there are morethan half a dozen of different calls and answers are joint projects proposed bygroups of higher education institutions
7 Since 2012 research units are evaluated along six criteria (publicationsattractiveness valorization and technology transfer governance doctoral programand project) For each of them a grade is attributed (A+ A B or C) and a globalassessment of the unit is written at the end of the report The global grade issuppressed
8 Until recently the ministry run the payroll of French universities and the lat-ter only managed an operating budget
9 In France some research units are called ldquomixedrdquo because they are simulta-neously located in universities and recognized by a national research institution(CNRS INSERM INRA etc) They receive funding and staff from both sides andmay decide whether a specific research project will be run by the university or bythe research institution to which they are associated10 France is divided into four (now five) territorial zones and the AERES evalu-
ates one zone each year Each campaign is called a wave11 In French universities the academic council (conseil scientifique) is an elected
body of 2040 members composed by 6080 of permanent staff (915 of thesebeing academics) by 1015 of representatives of the doctoral candidates and1030 members external to the university It decides on the scientific policy ofthe university and on the allocation of research budgets It is also involved in careerissues for the academic staff (promotion for instance)12 This notion of research active and its definition were often criticized by those
opposed to the AERES It was first introduced in the late 1990s by the ldquoscientificmissionrdquo a pool of academic experts appointed by the ministry before the creationof the AERES but the results were not publicly used and made visible
73Empowerment of French Universities
REFERENCES
Aust J (2005) La loi Faure une rupture avortee Effets de policy feedback et creation des
universites a Lyon Politiques et management public 23(1) 5369
Barrier J amp Mignot-Gerard S (2013) Leadership et changement dans une organisation plur-
aliste Le cas des transformations drsquoune universite In I Vandangeon-Derumez amp D
Autissier (Eds) Le changement organisationnel 10 etudes de cas commentees (pp
121140) Paris Eyrolles
Battilana J Leca B amp Boxenbaum E (2009) Agency and institutions A review of institu-
tional entrepreneurship Academy of Management Annals 3 65107
Becher T (1989) Academic tribes and territories Philadelphia PA Open University Press
Bezes P (2008) The reform of the state The French bureaucracy in the age of new public
management In A Cole P Le Gales amp J Levy (Eds) Developments in French politics
four (pp 172190) London Palgrave MacMillan
Braun D amp Merrien F-X (Eds) (1999) Governance of universities in a comparative perspec-
tive London Jessica Kingsley Publishers
Brunsson N (1989) The organization of hypocrisy Talk decisions and actions in organiza-
tions New York NY Wiley
Brunsson N amp Sahlin-Andersson K (2000) Constructing organizations The example of the
public sector reform Organization Studies 21 721746
Camerati F (2014) Les transformations du systeme de financement de la recherche dans les uni-
versites britanniques PhD in sociology Paris Sciences Po
Chatelain S Mignot-Gerard S Musselin C amp Sponem S (2012) La gouvernance des uni-
versites francaises Pouvoir evaluation et identite Paris CSO Monograph Retrieved
from httpcsoeduuploaddossiersRapport_GouvernanceUniversitesFrance_2012pdf
De Boer H (2002) On nails coffins and councils European Journal of Education 37(1)
720
De Boer H Enders J amp Leisyte L (2007) Public sector reform in Dutch higher
education The organizational transformation of the university Public Administration
85 2746
De Boer H Enders J amp Schimank U (2007) On the way towards new public management
The governance of university systems in England the Netherlands Austria and
Germany In D Jansen (Ed) New forms of governance in research organizations
Disciplinary approaches interfaces and integration (pp 137152) Dordrecht the
Netherlands Springer
DiMaggio P (1988) Interest and agency in institutional theory In L Zucker (Ed)
Institutional patterns and organizations (pp 322) Cambridge MA Ballinger
DiMaggio P amp Powell W (1983) The iron cage revisited Institutional isomorphism and
collective rationality in organizational fields American Sociological Review 48(1)
147160
Fligstein N (1985) The spread of the multidivisional form among large firms 19191979
American Sociological Review 50(3) 377391
Fligstein N (1990) The transformation of corporate control Cambridge Harvard University
Press
Fligstein N (2001) Social skill and the theory of fields Sociological Theory 19(2) 105125
Fligstein N amp McAdam D (2012) A theory of fields Oxford Oxford University Press
74 CHRISTINE MUSSELIN
Greenwood R amp Hinings C R (1996) Understanding radical organizational change
Bringing together the old and the new institutionalism The Academy of Management
Review 21(4) 10221054
Greenwood R amp Miller D (2010) Tackling design anew Getting back to the heart of orga-
nizational theory Academy of management perspectives November 7888
Greenwood R amp Suddaby R (2006) Institutional entrepreneurship in mature fields The big
five accounting firms Academy of Management Journal 49(1) 2748
Hallett T amp Ventresca M (2006) Inhabited institutions Social interactions and organiza-
tional forms in Gouldnerrsquos patterns of industrial bureaucracy Theory and Society
35(2) 213236
Krucken G Blumel A amp Kloke K (2013) The managerial turn in higher education On
the interplay of organizational and occupational change in German academia Minerva
51(4) 417442
Krucken G amp Meier F (2006) Turning the university into an organizational actor In
G Drori J Meyer amp H Hwang (Eds) Globalization and organization (pp 241257)
Oxford Oxford University Press
Lazega E amp Wattebled O (2010[2009]) Two definitions of collegiality and their
inter-relation The case of a Roman Catholic diocese Sociologie du travail in English
53(1) e57e77 [originally published in French in 2010 Deux definitions de la collegia-
lite et leur articulation le cas drsquoun diocese catholique Sociologie du travail 52
481502]
Leibfried S (Ed) (2010) Die Exzellenzinitiative Zwischenbilanz und Perspektiven Frankfurt
Germany Campus
Meyer J amp Rowan B (1977) Institutionalized organizations Formal structure as myth and
ceremony American Journal of Sociology 83 340363
Mignot-Gerard S (2006) Echanger et argumenter Les dimensions politiques du gouvernement
des universites francaises PhD in sociology Paris Sciences Po
Mignot-Gerard S (2012) Le gouvernement drsquoune universite face aux investissements
drsquoavenir Entre reactivite et resistances Politiques et Management Public 29(3)
519539
Mignot-Gerard S amp Musselin C (2002) More leadership for French universities but also
more divergences between the presidents and the deans In M Dewatripont
F Thys-Clement amp L Wilkin (Eds) European universities Change and convergence
(pp 123146) Bruxelles Belgium Editions de lrsquoUniversite de Bruxelles
Musselin C (2004[2001]) The long march of French universities New York NY Routledge
(first published in French by Presses de Sciences Po Paris 2001)
Musselin C (2005) Change and continuity in higher education governance Lessons
drawn from twenty years of national reforms in European countries In I Bleiklie amp
M Henkel (Eds) Governing knowledge A study of continuity and change in higher
education a festschrift in honour of Maurice Kogan (pp 6580) Dordrecht the
Netherlands Springer
Musselin C (2007) Are universities specific organisations In G Krucken A Kosmutzky amp
M Torka (Eds) Towards a multiversity Universities between global trends and national
traditions (pp 6384) Bielefeld Germany Transcript Verlag
Musselin C (2012) Liberte responsabilites centralisation monograph Paris CSO and ESEN
Retrieved from (httpcsoeduuploaddossiersRapport_LiberteResponsabiliteCentra
lisationUniversites_2012pdf) Accessed on February 13 2014
75Empowerment of French Universities
Musselin C (2013) How peer-review simultaneously empowers the academic profession and
university managers Evolution of the relationships between the state the universities
and the professiorate Research Policy 42(2) 11651173
Oslashstergren K amp Stensaker I (2007) Strategic responses to the quality reform A comparative
study of changes in Norwegian higher education Manuscript prepared for the biannual
NFF conference Bergen Norway
Padget J amp Powell W (2012) The emergence of organizations and markets Princeton NJ
Princeton University Press
Paradeise C Reale E Bleiklie I amp Ferlie E (2009) University governance Western
European comparative perspectives Dordrecht the Netherlands Springer
Pontille C amp Torny (2010) The controversial policies of journal ratings Evaluating social
sciences and humanities Research Evaluation 19(5) 347360
Ramirez F (2006) The rationalization of universities In M-L Djelic amp K Shalin-Andersson
(Eds) Transnational governance Institutional dynamics of regulation (pp 224245)
Cambridge Cambridge University Press
Reale E amp Primeri E (2013) Reforming universities in Italy Towards a new paradigm In
C Musselin amp P Teixeira (Eds) Reforming higher education Public policy design and
implementation Dordrecht the Netherlands Springer
Streeck W amp Thelen K (Eds) (2005) Beyond continuity Institutional change in advanced
political economies Oxford Oxford University Press
Thornton P H amp Ocasio W (2008) Institutional logics In R Greenwood C Oliver
K Sahlin amp R Sudaby (Eds) The Sage handbook of organizational institutionalism
(pp 99119) London Sage Publications
Thornton P H Ocasio W amp Lounsbury M (2012) The institutional Logics perspective a
new approach to culture structure and process New York NY Oxford University
Press
Whitley R (2007) Changing governance of the public sciences The consequences of establish-
ing diverse research evaluation systems In R Whitley amp J Glaser (Eds) The changing
governance of the sciences The advent of research evaluation systems (pp 327)
Dordrecht the Netherlands Springer
Whitley R (2008) Constructing universities as strategic actors Limitations and variations In
L Engwall amp D Weaire (Eds) The university in the market (pp 2337) Colchester
UK Portland press limited
76 CHRISTINE MUSSELIN
FUNDING INDIVIDUALS CHANGING ORGANISATIONS
THE IMPACT OF THE ERC ON
UNIVERSITIES
Jakob Edler Daniela Frischer Michaela Glanz and
Michael Stampfer
ABSTRACT
University governance is constantly challenged by changing expectationsand contexts New prestigious and well-endowed funding schemes areone possible source of pressure for change of university governance Thisarticle analyses the impact of one such scheme the grants of theEuropean Research Council (ERC) on the governance of European uni-versities After outlining a model of how this impact on universities canbe expected to occur we present the results of an exploratory study at avery early stage of the ERCrsquos existence (20102011) The empiricalanalysis is based on an investigation of 11 universities in eight countrieswhich shows that different kinds of universities are affected in varied and
Organizational Transformation and Scientific Change The Impact of Institutional
Restructuring on Universities and Intellectual Innovation
Research in the Sociology of Organizations Volume 42 77109
Copyright r 2014 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN 0733-558Xdoi101108S0733-558X20140000042003
77
often unexpected ways with particular differences arising at differentlevels within the universities
Keywords European Research Council project funding impact uni-versity strategies multi-level impact
INTRODUCTION
The European Research Council (ERC) was founded in 2007 as a science-led funding body intended to support lsquofrontier researchrsquo carried out byindividuals and their teams on the basis of an open competition acrossEurope The sole selection criterion for funding so far has been scientificexcellence Within its overall mission to improve scientific excellence andthe conditions under which it is produced within Europe the ERC alsoexplicitly wishes to lsquohelp universities and other research institutions togauge their performance and encourage them to develop better strategies toestablish themselves as more effective global playersrsquo (ERC MissionStatement)1 mainly through its procedures benchmarks and transparencyof results As the ERC schemes are an attractive income source and carrierof considerable prestige and reputation organisations have an obviousincentive to react to the new European instrument
The starting assumption of our analysis2 is that there are two distinctways in which organisations can respond to the ERC First they are likelyto react strategically to the fact that there is a new European excellencedriven instrument on offer that changes the overall incentive structures oforganisations and provides a pan-European benchmark Second once uni-versities have a grantee they are likely to adjust to the fact that they haveERC grantees endowed with prestigious and generous grants It is impor-tant though to note that universities and similar public research organisa-tions are limited in their strategic capabilities and ability to exercise centralmanagerial control over work processes in a comparable manner to manyprivate firms in market economies (Musselin 2007 Whitley 2008 Whitley ampGlaser this volume)
Against this background the objectives of this article are twofold It firstoffers a conceptualisation of how the ERC could affect universities as orga-nisations developing an impact model and operationalising its major inter-vening and independent variables Second it presents the results of aninitial exploratory application of this model analysing impact at a very
78 JAKOB EDLER ET AL
early stage of the ERC existence (20102011) We base our analysis on aqualitative research programme investigating eleven universities in eightcountries based on documentary analysis and an interview programme
Starting with the core question of how to conceptualise the impact ofthe ERC on universities we further address a number of empirical ques-tions What changes are induced due to the relative value universities attri-bute to the ERCrsquos new funding instruments and to the fact that they havegrantees What is the awareness and relative meaning of the ERC grantsfor strategic leadership How are the different levels within the organisa-tions affected What context and organisational conditions and transfor-mations mediate the effect of the ERC grants on organisations
The paper is structured in three parts The first develops the conceptualmodel the second summarises the methods used in our empirical studywhile the third presents our exploratory empirical findings structuredaccording to three different stylised types of organisations and three typesof impact The conclusion discusses these findings in terms of the maindimensions of the previously developed impact model
CONCEPTUAL MODEL
In this section we present the various elements of the conceptual model tobe used to operationalise and measure the likely impact of the ERC on uni-versities We begin with some reflections on the nature of universities asparticular kinds of organisations that have limited strategic capabilitiesespecially in their core functions of research and teaching
Universities as Strategic Actors
Universities today are facing a complex set of challenges and requirementsthat is well documented in the literature The New Public Management(NPM) influenced shifts of the past 20 years have slowly but steadily chan-ged the way many universities see themselves and are seen by others (egAghion Dewatripont Hoxby Mas-Colell amp Sapir 2008 Clark 1998Ferlie Musselin amp Andresani 2008) They are increasingly subjected todemanding performance requirements broadened functions such as lsquothirdmissionrsquo (Laredo 2007 Nedeva 2007) and efficiency pressures and arehaving to compete nationally and internationally against each other for
79Funding Individuals Changing Organisations
prestige resources and staff As a consequence many universities attemptto manage research activities that is to define and support effectiveresearch portfolios and establish structures and processes that are attrac-tive for excellent researchers of all career stages (eg Schmoch amp Schubert2010) It is this understanding of universities as organisational actors thatunderpins the ERCrsquos wish to improve the effectiveness of universities
However when analysing the impact of an external impulse such as thenew funding schemes of the ERC on the strategic action of universities weneed to understand the inherent limitations of universities as actors Theorganisational actorhood of universities can be understood lsquoto imply uni-fied central authority over the design of work processes the co-ordinationof their outputs and the development of collective capabilities for dealingwith problems adapting to change and seizing entrepreneurial opportu-nities through mobilizing the commitment of skilled staffrsquo (Whitley 2008p 26 see also Whitley amp Glaser this volume) Whitley argues that thedominant authority of scientific global communities in assessing the rele-vance and excellence of science and quality of scientists implies thatresearch strategies and their outcomes cannot be fully planned and mana-ged by university managers Moreover the frameworks for assessment andattribution of excellence and rewards are different in different areas ofknowledge production and thus extremely heterogeneous across universi-ties This leads to lsquohellip limitations on the ability of universities as employersto organize and direct particular divisions of scientific labour and integrateresulting outputs around specific organizational purposesrsquo (ibid) and thusto restrictions of centralised authority and collective actorhood As a resultthe management of research and advanced teaching generally occurs at thelevel of sub-units within the organisation
This overall assessment of strategic capabilities and limitations needs tobe differentiated for different types of organisations (Van Vught et al2005) that vary according to the overall level of autonomy granted to uni-versities by different states their established governance structures socialand scientific prestige the culturally embedded inclination of academicsand researchers to function within lsquomanagedrsquo systems and the wealth oforganisations This last the level of resources at the disposal of the organi-sational leadership considerably determines its ability to steer as largeshares of institutional funding or large endowments can be used for settinginternal incentives while on the other hand a transition to project basedfunding shifts the logic of resource acquisition to the level of research unitsand individuals All this implies that the actorhood of universities is deeplyshaped by the regulatory and political frameworks of national science
80 JAKOB EDLER ET AL
systems (Bleiklie amp Kogan 2007 Estermann amp Nokkala 2009 Geuna ampMartin 2003 Jongbloed 2009 Observatory of the European University2006 Oquist amp Benner 2012 Paradeise Reale amp Goastellec 2009Whitley 2008) With these qualifications for strategic actorhood in mindwe can now turn to the development of our impact model for the ERCgrants on universities We start with the independent variable the ERC itsobjectives in wishing to affect universities and its instruments
The Independent Variable The ERC its Mission for ResearchOrganisations and its Instruments
In its official mission statement the ERC is defined as the first Europeanfunding body set-up to support bottom-up investigator-driven frontierresearch which is intended to substantially strengthen and shape theEuropean research system There are three major ways in which it aims toaffect universities to help them lsquogaugersquo their performance as research orga-nisations to improve their strategies and to stimulate investment in sup-porting promising new researchers
Three General Dimensions of ERC ImpactThe ERC affects organisations by offering them three kinds of value mate-rial symbolic and normative3 The material value of grants can make a sig-nificant difference not only for the individual but also for the organisationand its sub-units At the time of this investigation the ERC offered twodistinct instruments to fund investigator-driven frontier research namelythe Advanced Grants scheme and the Starting Grants scheme StartingGrants aimed at boosting the career of the next generation of research lea-ders by promoting early scientific independence and required a respectivecommitment by the host institution that allowed potential grantees to inde-pendently direct their research and administer their resources AdvancedGrants were intended to support researchers who had already establishedthemselves as exceptional leaders in their fields with a track record of sig-nificant achievements over the past ten years ERC grantees receive up toh20 million for Starting Grants StG and up to h35 million forAdvanced Grants AdG both for a period up to five years Budgets inboth cases can be handled flexibly giving the grantees a powerful andindependent position Moreover ERC grants are portable between organi-sations and across countries4
81Funding Individuals Changing Organisations
On the symbolic level ERC grants have already developed high prestigeand reputation both for the grantee and the hosting organisation To get orto host ERC grants is considered a measure of excellence and academiccompetitiveness Lastly the ERC exerts normative influence on individualsepistemic communities research organisations and funding organisationsin Europe by providing a definition of risky and excellent research as wellas an understanding of the role and autonomy of (young) researcherswithin organisations all of which underpin the justification of the ERCand its instruments
Two Impact LogicsThe potential impact of the ERC and its new funding instruments on orga-nisations can be understood on two levels (1) the consequences of havinggrantees and more generally (2) the consequences of the sheer existence ofERC and its offer of European excellence grants on the competitive contextof organisations and thus their strategic positioning
The Consequences of Having Grantees First we assume that granteeswithin a university will make a difference due to the ERCrsquos material sym-bolic and normative value Universities may react at different levels inorder to create the legal financial career related or infrastructural environ-ment that is most favourable for the grantee to be successful ERC offersfunds of considerable size and at favourable conditions While within theresearch budgets of large organisations the individual grant may seem rela-tively small a number of awards actually can make a difference Evenmore when we focus on sub-units within organisations as this additionalsource of income may change their situation considerably both in a sup-porting and in a turbulence-stirring way
The Consequences of Having the ERC and its Instruments as such Secondwe assume that the very existence of the ERC and its offer of excellencegrants on a European level makes universities react even if they do not(yet) have grantees The logic here is that the ERC through its symbolic(ie high prestige coming with getting a grant) and normative value(through contribution to the definition of excellent research and position ofyoung researchers linked to the grants) changes the context and incentivestructures of organisations potentially leading to strategic reactions andorganisational change
82 JAKOB EDLER ET AL
Capturing Impact Operationalising the Dependent Variables
We can now further differentiate the dependent variables and define cate-gories of impact while sticking to the three explicit objectives of the ERCas regards research organisations As mentioned above the ERC seeks tohelp universities to improve their performance enhance their strategic cap-abilities and attract (and nurture) scientific researchers In this context
(1) improving (perceived and expected5) performance relates to quality andquantity of research output (impact level of journals setting newresearch agendas etc) to increased visibility of the university or indivi-dual units to extraordinary growth of research groups and to theincreased capability in raising further research income (high-levelgrants)
(2) enhancing strategic capability involves the development and implemen-tation of strategies to succeed in a global competition for staffresources and reputation This includes attempts to raise the profile andcoherence of organisations and to increase regional and global recogni-tion as well as strengthening or building-up of organisational capabil-ities to define goals and develop structures and processes to achievethem
(3) supporting and attracting talent means to support lsquopromisingrsquo research-ers and to increase the capability to successfully incorporate them intothe universityrsquos research environment This also translates morebroadly into the creation of contextual conditions and supportmechanisms within organisations that best suit (current or future) gran-tees in their drive to set up and implement challenging frontier researchprojects which might need new levels of autonomy and co-operation toallow for different kinds of research This furthermore includes(intended or sometimes unintended) effects on staff retention that isstaying attractive for both grantees and other already existing staff notdirectly profiting from an ERC grant
This operationalisation of the dependent variables constitutes the basisfor the empirical programme
Determinants of Change Defining Three Bundles of Intervening Variables
We now turn to the broad range of intervening variables that can influencehow the ERC can be expected to affect universities In this study these are
83Funding Individuals Changing Organisations
derived from the major dimensions of change elaborated in the compara-tive analysis on national reform programmes as regards governance of uni-versities in Europe in the book by Paradeise Reale Bleiklie et al (2009)(2009) and Paradeise Reale and Goastellec (2009) This comparative analy-sis has confirmed the overriding importance of power and competence struc-tures to adapt (autonomy of universities) endowment with both financialresources and human resources and ways in which those financial and humanresources can be allocated Consequently we focus on three intervening fac-tors in the unfolding of the ERCrsquos impact on universities autonomy leveland allocation mechanisms of funding and human resource policy
First the level of organisational autonomy This refers to the ability ofuniversities to determine their goals and procedures independently fromthe state as granted by the national framework and established in a his-torical trajectory This also includes the autonomy of sub-units withinorganisations which may limit the development of organisational capabil-ities at the centre but may also enhance their ability to (re-)act to theexternal impulse of the ERC National frameworks if granting formalautonomy do not automatically translate into actual autonomy At thesame time external pressures to develop organisational strategies are ris-ing both through regulations and increased competition (Fumasoli ampLepori 2011 Krucken Kosmutzky amp Torka 2007) The ERC as a newactor in this context adds to the opportunity environment by providingsupport for promising individuals and frontier scientific research(Nedeva amp Stampfer 2012)
The second set of variables has to do with research funding namely theoverall endowment with funds and the conditions under which these fundsare allocated to research We can see a considerable shift to conditionalprogramme funding that alters the rationale and degrees of freedom forresearch within organisations Universities need to adapt their strategiestowards higher funding income and diversification of activities and fundingsources External sources of various kinds probably have gained a higherinfluence on the distribution of remaining internal funds (eg Granberg ampJacobsson 2006) Actor strategies vary across organisations and can beseen as a kind of indicator for the intended positioning of different actorsincluding organisations and countries and their readiness to change(Braun 2011) As regards the ERC organisationsrsquo record with other fund-ing sources experiences as well as options might be responsible forimportant differences in ERC success ambition and ability to incorporatethe changes induced by ERC funding
84 JAKOB EDLER ET AL
Third the last bundle of variables is clustered around human resourceissues the people performing research This includes not only the quality ofpeople per se but also the way that universities can and actually do orga-nise recruitment and reward procedures and develop personnel strategiesto best serve their overall goals Inward mobility has increased though dif-ferently in different countries (Franzoni Scellato amp Stephan 2012) andunder certain constraints universities are becoming more and more fullyresponsible for their (scientific) staff As part of national autonomy reformsall over Europe the management of human resources at universities hasbecome one of the central tasks performed at the organisational level(Bleiklie 2011 Enders 2001 Musselin 2005) In this context the ERCadds new opportunities for younger researchers raises the autonomy ofresearchers and might allow for catalytic change
We do though need to stress that individual organisational contexts mat-ter strongly as changes in overall governance and fundinginstruments meandifferent things to different organisations Universities implement externallyinduced reforms in different ways as they are facing unequal starting condi-tions Adaptation pressures may be enforced blocked or re-directed throughvaried organisational settings endowments and existing prestige and excel-lence While path dependency may account for many differences it alsodemonstrates that universities are quite stable organisations where change israther incrementally implemented and mostly resource driven
The Model
Our resulting conceptual and analytical model of the ERCrsquos impact onresearch organisations is summarised and illustrated in Fig 1 We considerthe ERC and the material symbolic and normative value it offers as anexternal impulse acting on organisations Here we focus on two basicimpact logics namely the consequences of having grantees and the conse-quences of having ERCrsquos offer of European excellence grants as suchHowever such impact is shaped catalysed or hindered by additional inter-vening factors Here we consider three intervening aspects highly impor-tant (i) autonomy (ii) level and allocation mechanism of funding and(iii) human resource policies and practices
Following the ERCrsquos mission to improve the performance of universitiesenhance their strategic capabilities and attract (and nurture) researcherswe conceptualise impact along the following three dimensions (i) improving
85Funding Individuals Changing Organisations
(perceived) performance (ii) enhancing strategic capability of universitiesand (iii) attracting (and retaining) promising talent6 Further all the impactsare to be differentiated for the overall organisation and the organisationalsub-levels (faculty department institutehellip)
A few further specifications to this model First there is a certain inter-play of variables and dynamics Thus note that arrows in Fig 1 only indi-cate the direction of overall impact logics without illustrating the moredetailed feedback loops and relationships between variables (in a certainpoint of time and over time) Second we assume that the discussed aspectsof autonomy funding and human resource are strongly related to the over-all pre-existing capability of universities to act and to successfully competeas to academic performance and reputation as well as for additional fund-ing and outstanding talent It is thus obvious that our intervening anddependent variables do overlap to a certain extent Organisations have ahistory and history of course does matter This aspect is particularlyinteresting but also tricky in the context of ERCrsquos impact on universitiesThe Matthew principle seems strong but can also more generally beapplied to the attracting talent dimension of our model a considerablenumber of high-level people do attract more high-level people etc
Third as often with research designs we do face an attribution problemWe need to consider that universities are facing additional pressures toadapt and transform themselves While we assume that the ERC indeed canmake an additional difference in this transformation process the attributionof effects is quite difficult (as with all impact assessments) This means that
via the
allocated
grants
via the offer
of European
excellence
grants
perceived
PERFORMANCE
REPUTATION
Intervening variables Dependent variablesIndependent variable
At d
iffere
nt le
vels
of o
rgan
isatio
n
ER
C
ENDOWMENT FUNDING
country characteristicshellip organisational charact
HUMAN RESOURCES
country characteristicshellip organisational charact
AUTONOMY
country characteristicshellip organisational charact
STRATEGIC
CAPABILITY
other conditions
Mate
rial s
ym
bo
lic n
orm
ativ
e v
alu
e
SUPPORTING
ATTRACTING
TALENT
Fig 1 The Basic Analytical Model
86 JAKOB EDLER ET AL
a solid understanding of the framework conditions organisations areencountering is thus crucial to finally assess the relative importance of theERC and the additional effect it has in relation to other pressures andopportunities We are aware that there might be impacts on organisationsthat are not covered in those three dependent variables and we are awarethat other developments might have effects in the same direction and ERCmight simply be one force out of many With respect to the results of theempirical investigation which is largely qualitative and based on a casestudy approach we emphasise that the ERC is one impulse for change oradaptation among many others
METHODOLOGY
The empirical sample comprises 16 organisations in 8 countries 11 univer-sities and 5 public research organisations This article however focuses onuniversities only Cases have been selected to cover a range of differentfunding environments size of organisations as well as a certain variety intheir strategic foci and actor capabilities as well as could be establishedex ante Although our focus is on universities hosting grantees to capturethe changes induced by having a grantee we included two universities with-out grants (referring to the first cohort of ERC grantees) Countries repre-sented in the sample establish a balanced mix of ambitious scientificallywell-performing countries moderately ambitious medium-performingcountries and countries with low ambitions and rather low performance(Braun 2011)7 Furthermore they show a range of governance patternsand HR frameworks that impinge on the way the ERC instruments canexert influence The different national conditions represented in these coun-tries and the role they play as intervening variable have been taken intoconsiderations in the university case studies We are aware that all cross-national university comparison needs to be seen in the light of the interven-ing national conditions and thus treat cases always as the combination ofnational and organisational framework conditions It is however beyondthe remit of this paper and its underlying research to deliver a systematicanalysis of the different national science systems and framework conditionswithin them
For our university sample in total 46 informed interviews8 have beenconducted on the basis of an interview guide building upon the operationa-lised variables as introduced above During interviews our major research
87Funding Individuals Changing Organisations
interest focused on perceptions of actors within respective organisations onhow the ERC has influencedinfluences organisations within the above-mentioned dimensions To abstract from individual opinions and to differ-entiate between formal and actual capability to act interviews have beenperformed on different levels within organisations (top intermediaryfaculty functionaldepartment administrative level) Finally we differenti-ate between four representations of lsquochangersquo (i) statements about reactionsthat actually represent real change (ie learning) (ii) responses claimingchange but rather representing window-dressing (iii) intentional non-change and (iv) change that is induced not by the ERC but through otherpressures and challenges
Analysis of empirical material was based on the methodological assump-tions of qualitative content analysis (according to Mayring 2008) allowingfor systematic analysis of empirical data which at the same time isinterpreted by taking into consideration the respective context (organisa-tional and country-specific intervening variables) Based on the previouslydefined research questions a preliminary set of analytical categories hasbeen defined In contrast to closed category-systems as proposed byMayring this category-system was refined in the course of our analysisuntil all relevant text passages (related to the impact of the ERC on organi-sations) were adequately recorded this was intended to allow for amechanism-oriented investigation of causal mechanisms (Laudel amp Glaser2010) The extraction of information relevant to answer our research ques-tions rests upon this iteratively refined category-system
We must stress that our empirical analysis is exploratory it serves thepurpose to develop and test the impact model and develop ideas for itsfuture improvement and application Moreover interviews were performedat a time too early to fully assess the performance difference the ERCmakes on universities in terms of scientific output prestige and furthergrant getting However the analysis could identify early signs of changeand gives indications that the symbolic and normative value of ERC doessignificantly matter
EMPIRICAL FINDINGS ON IMPACT PATTERNS
A Systematic Link Pre-Existing Organisational Capacity and ERC Impact
Our conceptual model assumed that the impact the ERC can have onresearch organisations is strongly moderated by a set of intervening
88 JAKOB EDLER ET AL
variables namely the organisational and country frameworks that deter-mine the organisational capabilities and positioning of the organisationOverall our exploratory empirical work has confirmed this assumption Itappears that there is a systematic connection between the pre-existing cap-abilities of organisations on the one hand and the nature scale and scope ofimpact the ERC has on the other hand
In particular it is possible to reduce complexity for the impact analysisby grouping the universities in our sample on the basis of our ex post analy-sis into three categories that vary in their resource endowment and co-ordination capabilities powerful in-between and powerless The assignmentof the organisations to one of these three categories was based on ourassessment of the three key bundles of both national and organisationalcharacteristics we consider important as intervening variables shapingERCrsquos impact that is (1) autonomyability to act as an organisation (atorganisational or sub-unit level) (2) pre-existing research fundingendowmentand (3) recruitmenthuman resources policies In our view these threedimensions altogether are assumed to define a universityrsquos overall organisa-tional capacity9 For the initial sample selection a simple strategy had beenapplied in order to cover a sufficient range of organisations with respect toorganisational types (universities non-university research organisations)coverage of fields (full coverage specialised into certain fieldsdisciplines)and number of hosted ERC grantees within the first cohorts of Startingand Advanced Grants (hosting more than 5 grantees up to 5 grantees no grantees) A first empirical analysis showed that certain kinds of impactare correlated with a specific set of manifestations of our intervening vari-ables This finding then was the basis of our definition of the stylised orga-nisational baskets and a preliminary allocation of organisations to thosebaskets This allocation has been further refined during the later stages ofthe analysis Because of the heterogeneity of the sample organisations inthe same category still differ considerably as regards individualorganisational characteristics10
More concretely for each intervening variable an ordinal scale wasdeveloped which allowed us not only to consider the statistical facts ofeach organisation (gathered by desk research) in more detail but to furtherinclude the more subjective impressions and perceptions of our intervieweeson their universityrsquos organisational capacities The scales were then com-bined into a summary scale and translated into our three categories ofintervening parameter sets as described above This categorisation has beeniteratively developed during the course of the qualitative content analysisof the collected interview material where we further decided to differentiate
89Funding Individuals Changing Organisations
with respect to essential supportive and hindering intervening parametersAs a result each of the three categories contains at least three organisationsfrom at least two different countries whereas several countries are repre-sented in more than one basket This is far from constituting a comprehen-sive sample but goes beyond a single-organisation case study approach andjustifies a certain approximation of results referring to the organisationaltypes
On this basis a typical stylised powerful organisation is well endowedand can draw on considerable funds available to perform research11
Furthermore such organisations are characterised by a high level of orga-nisational autonomy allowing them to actively respond and act to chan-ging environments Finally these organisations do have a clear and wellimplemented recruitment and human resources development strategy foryears that has already resulted into hiring of a larger pool of highly reputedresearchers by offering favourable highly competitive employment condi-tions (including salary levels) as well as incentive oriented human resourcesdevelopment activities This of course is only possible if organisations facea considerable degree of autonomy and can dispose of important researchendowments Overall those organisations regard themselves and areregarded by others as delivering excellent research
A typical stylised in-between organisation is characterised by a substantialoverall endowment for research that is however constantly under pressureIt shows some capability to realise and act upon emerging opportunities toimprove its own status but its strategic actor capability is limited (for his-torical institutional andor legal reasons) While those organisations oftenmay not have an elaborate formal recruitment or human resources develop-ment policy at the top organisational level on the level of operating unitsrecruitment can be quite systematic Furthermore within these organisa-tions we find acknowledged fields of excellence but often these are limitedto specific areas andor not yet fully visible In sum those organisationsboth have some basis for research excellence and for strategic actor capabil-ity to respond to the external lsquoshockrsquo of a new high level and highlydemanding ERC funding scheme
A typical stylised powerless organisation is characterised by a low levelof research funding available beyond that available funding is probablyearmarked for other tasks or cost categories than research activities Thissituation is accompanied by low levels of organisational autonomy andstrategic leadership further translating into a lack of capability to ade-quately handle human resource policies at the organisational levelHowever this lack of overall strategic capability to act as an organisation
90 JAKOB EDLER ET AL
does not necessarily imply that weak organisations are not able to performhigh-quality research and host top researchers in certain individual areas orpockets of excellence However these mainly individual pockets of excel-lence do not put an excellence mark on the organisation as a whole Theallocation of organisations to those categories is a supporting mechanismfor the analysis and generalisation as a result of the analysis as we havefound some very generally speaking systematic relation between thosebaskets and ERC impact12
Impact on Powerful Universities
(Perceived) Performance and VisibilityIn some of these organisations the ERC has led to a new discourse on whatconstitutes quality and how people should be assessed as well as on howgovernance structures should be set up to best support scientific qualityFurther many interviewees see ERC grants due to funding conditions andduration that are more generous than most national grants as potentiallyenhancing the quality of the research more generally The ERC is here per-ceived as a visible indicator of a catalyst for quality This is a unanimousassessment All interviewees at all levels confirmed that ERC grants areseen as a new and more telling indicator for research excellence for somethe ERC is even regarded as a means to overcome shortcomings they per-ceive with traditional organisational rankings or assessments Thus theseorganisations aim at confirming their position within the Europeanresearch landscape through the number of grants they receive Elite organi-sations with a high number of grantees strive to keep this level whereasthose with lower number of grants than comparable organisations see astrong need to catch up in this indicator and they have the means and theorganisational capabilities to do so
Interviewees consider the pre-existing quality as most important pre-condition for successful grant getting As top organisations are inter alia defined through having high reputation and adequate support structuresalready many of the interviewees do not see a special need for an additionalinvestment in ERC activities Further the difference the ERC makes isdependent on the existence of similar types of funding instruments in theorganisationrsquos home country If there has been such an instrument beforetop organisations have already been proficient in integrating these externalimpulses in their internal processes
91Funding Individuals Changing Organisations
Strategic CapabilitiesThe impact on strategic capabilities at the leadership level in powerful orga-nisations is not obvious While leadership has and exerts the capability torecognise and strategically incorporate emerging research fields and newlydeveloped research lines the major decisions and conditions with respect toresearch are set at a lower organisational level the operational effects arethus at the lower (and sometimes intermediate) level Even for elite organi-sations research strategy is often developed within constraints of fundingconditions and assessment exercises and their specific output-oriented anddiscipline-oriented criteria and it is often highly path-dependent The well-endowed and highly prestigious ERC grant allows a change of researchdirection which often necessitates investment of time and patience that isnot possible in more traditional funding environments With ERC lsquowe canbe much bolderrsquo and do things lsquowe could not do otherwisersquo An importanteffect here even in well-endowed organisations is lsquomaterialrsquo as ERC contri-butes to the funding of sub-units that are competing for institutional andproject grants This allows for more flexibility in one case even securingthe sustainability of a centre This overall impact of the ERC is especiallyimportant for top performing organisations that are growing quicklywhere a number of grants can be catalytically and accelerate the process oftransformation and profiling in this growth period
In terms of thematic bundling and profiling contrasting impacts are to beobserved On the one hand ERC grants have helped to mobilise and bene-fit from platforms with long term risky trajectories This was possiblethrough multiple grants in related areas On the other hand thematicbundling through ERC grants is structurally hampered as it becomes onlypossible through ex post bundling but not through the submission of jointgrants Thus it has limited impact on thematically integrative research stra-tegies of larger units faculties or large divisions which are planning to buildnew horizontal teams along clearly defined thematic lines
In powerful organisations the ERC is increasingly used as a benchmarkof units within the organisation and thus is a tool for internal competitionfor prestige and grants In two examples serial ERC successes haveseverely enhanced and strongly confirmed the reputation of a specific areawithin a top organisation In other cases specific care is taken that the ERCsuccess is reported not only externally but within the organisations expect-ing an improved bargaining position at higher levels and vis-a-vis otherunits
All of those aspects have already led to the common perception in sub-units of powerful top organisations that efforts should be put towards
92 JAKOB EDLER ET AL
getting those grants Some interviewees stressed that ERC grant gettingsupport within their units is more pronounced than support for national orEU Framework Programme funding Thus we see some investment in orderto be visible as ERC attracting unit in these organisations While in thesecases we did not observe fundamental changes in the way internal supportof research quality and income generation is organised they take advan-tage of existing structures and have strengthened and tailored them wherenecessary
Human Resource Policy Support Attraction and Retention of PromisingResearchersIn general terms the ERC as a means to increase attractiveness for organi-sations is perceived as very positive it fits into the general claim of eliteorganisations to recruit support and retain high performers However asregards the toolbox of human resource management most intervieweesreported no major changes The existing structures and incentives to attractand support talent are merely extended to ERC grantees as existing prac-tices are considered efficient enough to actually perform the task of gettingand lsquoservicingrsquo high potentials For these organisations the ERC providesfor an additional (symbolic and normative) currency in the market amutual signalling device for the individual researcher and for employingorganisations but the key criterion currently is and will remain output andpublications In some cases the larger grant is used to pay an additionalbonus for the researcher who succeeded in the ERC competition
In terms of actual recruitment the ERC has not yet had a significantimpact In our interviews cases of recruitments of grantees are not particu-larly common instead most grants (AdG and StG) were submitted byresearchers who have already been employed by the respective home insti-tution for some time At the unit level however there are additional effortsundertaken as some see a need to recruit ERC grantees or at least potentialERC grantees At times the ERC grant was even regarded as a sufficientquality stamp reducing the number of formal interviews in an ongoingrecruitment process In addition for some organisations the ERC repre-sents a welcomed opportunity to retain high-performing staff as by meansof ERC (and ERC money) generous prolongation of existing contracts canbe handled more easily
Even in elite organisations individual strategic leaders at the lower levelsof the university hierarchy perceive a potential downside of ERC grants Firstit puts normative pressure as to the conditions granted to high-performingindividuals as it potentially enhances the autonomy of individuals and thus
93Funding Individuals Changing Organisations
limits the autonomy of their strategic managers Leaders within top organisa-tions see a second downside caused by the fact that large ERC grants actu-ally can endanger the balance within units and groups and distort establishedlevels of cohesion This is specifically true when the grant allows for multiplerecruitments by the PI and thus leads to the development of research groupsthat is not planned for strategically and may not spill over to the unit moregenerally (leading to a lsquotwo class systemrsquo) In these cases unintended effectson retention of existing staff might be an issue however this needs to beinvestigated further Moreover in a comparable situation one organisationhas started to be restrictive with grant applications taking care of a balancedgrowth rather than an accidental growth within their portfolio Internalpre-selection procedures do play a decisive role here The above-mentionedchallenges of large grants especially if units have attracted multiple grantstrigger new efforts and set-up of structures to plan handle and support largeproposals and grants in sub-units by simplifying internal rules and proce-dures Even in organisations that have been reluctant to apply for externalfunds as not many schemes meet the criteria of their research staff arenow requested to consider ERC applications
Impact on in-between Universities
(Perceived) Performance and VisibilityAll interviewees agree on the validity of ERCrsquos quality brand that arisesfrom its symbolic and normative value ERC is expected to impact on qualityof research as it allows researchers to focus on basic research withoutfurther conditions However for in-between organisations ERCrsquos impacton the perceived performance of organisations or sub-organisational levels(divisions departments institutes labs groups) is closely linked to whetherthese are hosting ERC grantees or not Having (the right number of) ERCgrantees is perceived as bringing visibility and reputation for the respectiveorganisation or sub-division internally as well as externally (nationallyand internationally) With a growing number of ERC grantees acrossEurope the pressure on those organisations and sub-units within themincreases to succeed in order to further assure their visibility a commonlyused statement in this respect lsquoWe cannot afford not to succeed (in thenext round again)rsquo
Further the ERC allows certain units or research groups within anorganisation to grow extraordinarily or faster than comparable units with-out any grantees While more generally in-between organisations see this as
94 JAKOB EDLER ET AL
an opportunity for a more competitive research performance in the comingyears some of them are challenged by an inherent risk of uncontrolled andimbalanced growth triggered by multiple ERC successes pushing organisa-tions towards their limits as regards allocation of necessary space supportor infrastructure
Strategic CapabilitiesEvidence of ERCrsquos impact on the overall strategic capability is mixed Inour sample of in-between organisations there has been no evidence that amajor re-orientation of research strategies or completely new research prio-rities have come up due to ERC However often through ERC grantsalready ongoing policies and research priorities have been (i) made visibleor justified (ii) considerably strengthened in their research capacities and(iii) in some cases used as seed money directed into new more risky lines ofresearch within the field For the in-between organisations this chance ofincreased visibility also serves as a driver to adapt internal incentive struc-tures to motivate existing staff to apply for ERC grants Furthermore dueto the considerable size and the favourable funding conditions of ERCgrants research may be re-organised in a quite flexible and non-bureaucratic manner ERC grants are considered an opportunity to catch upwith excellence strategies and helping to raise a unitrsquos research capabilitiesbe it via new infrastructure via building-up critical mass via extraordinarygrowth of research groups via more incentive-driven overall financialendowment but also via new and different research methodologies orapproaches to be applied As regards the latter ERC money primarily inthe case of Advanced Grants also catalyses new more risky lines ofresearch (within already existing priority research areas) particularly as itcomes with a favourable time span In many cases ERC money acts as seedmoney and allows units or research groups within an organisation to growextraordinarily or faster than comparable units without any grantees
As ERC success is considered an easily measureable and valid qualitycriterion for research it is increasingly used for strategic decisions also at thetop level from the human resources domain to the internal distribution ofmoney Organisations allow faster promotion and leap-frogging of careersteps for ERC grantees or provide additional money or infrastructure forhisher research group or department This is not only handled ad hoc butin some cases integrated as additional momentum into existing evaluationstrategies for people as well as for organisational units Furthermore orga-nisations do not only integrate the actual ERC success into strategic con-siderations but also reflect on ERCrsquos funding criteria and thus the
95Funding Individuals Changing Organisations
perceived likelihood for winning an ERC grant in already established qual-ity assurance mechanisms
All in all while the strategic merit of the ERC more likely unfolds at thesub-organisational levels of organisations (faculties departments instituteslabs) the perceived material and symbolic value of the ERC has triggeredstrategic reaction at leadership level as well There is a stronger awarenessthat the ERC is a means to catch up as well as to signal this catch up morebroadly The ERC is used as a trigger for reinforcing quality strategies andinternal allocation for resources in some cases the university allocates addi-tional money to the groups or sub-units of grantees in others ERC grantsare considered a useful indicator for quality based allocation of competitiveuniversity budgets The main challenge for the university leadership lies inbalancing these conflicting strategies of sub-levels a challenge that is notyet met by all organisations
Human Resource Policy Support Attraction and Retention of PromisingResearchersIn the realm of organisationrsquos employment and HR policy the existence ofERC and of having ERC grantees triggers diverging promotion structuresERC grantees are by means of being branded as ERC-excellent oftenentitled to leap-frog certain career steps that is are promoted to positionswith more autonomy than colleagues of comparable academic age arealready tenured at an earlier stage or entitled to a full professorshipHowever the ways in which grantees are governed differs ranging frommere expansion of already existing exceptional promotion rules for com-parable honours to ERC grantees to newly creating staff categories andpositions for ERC grantees that substantially differ from pre-existing staffprofiles and tasks (eg research only positions in a teaching-heavyenvironment)
Even more so than in powerful organisations in-between organisationsthat seek to catch up are facing considerable challenges in this respect TheERC related reward system puts a lot of pressure onto existing staff struc-tures and organisational capabilities especially if multiple grantees have tobe integrated within a certain unit This at the same time challenges existingcohesion among the unitrsquos staff members This is not only a question ofpromotion policy on the formal level (including national legal regulationsand governance frameworks) but also closely related to the question ofmotivating and retaining existing staff and thus closely linked to the attrac-tion of talent dimension Empirical evidence in this group of organisationsshows that ERC related reputation and prestige will likely lead to further
96 JAKOB EDLER ET AL
internal differentiation among staff members Generally ERC grantees are given the amount of funding they bring and the reputation and autonomythey enjoy in a privileged bargaining position towards their host univer-sity Success in the ERC grant competition is therefore often related withmore freedom and flexibility in doing research in most cases they are addi-tionally released from teaching tasks and benefit from lsquoresearch onlyrsquo-posi-tions lsquoThe other side of the coinrsquo is growing evidence of internal imbalancesand pressures within organisations given that other staff members do notenjoy the kind of special treatment provided to ERC grantees The internaldiscrimination also increases pressure on (so far) non-successful staff mem-bers For in-between organisations those pressures are critical as large ERCgrants are potentially more disruptive than in top organisations that havebetter endowments and are already more accustomed to success in compar-able high-level funding schemes
Relating human resource strategy to research strategy our empirical evi-dence shows that a excellence-oriented strategy prevails over a (top-downpre-defined) research priorities strategy This means in essence that theurge to attract and retain top class researchers is the highest priority thedefinition of a research strategy follows from that Catching up in-betweenorganisations basically employ bottom-up strategies and allow strong fieldsto grow and become more important However in parallel they start todevelop the strategic capability to incorporate these into the overallresearch strategy
In terms of overall professionalisation of HR the ERC appears to be acatalyst but not the main driver for ongoing trends in terms of HR man-agement In a number of organisations we found strong previous experi-ence with individual grant schemes or prizes of comparable size andcomparable selection criteria on the national level Applying for ERCgrants or having attracted ERC grants only strengthens and extends exist-ing mechanisms Examples for this re-enforcement are internal peer com-mittees for developing proposals further in terms of content as animportant way of supporting for potential ERC grantees These peer com-mittees play an additional role in terms of assuring internal quality controlover research conducted by potential grantees
This use of internal reviewers as a filter for ERC applications can be con-sidered rather ambiguous On the one hand the ERC instrument targetsindividuals and seeks to fund risky unconventional research On the otherhand it wants to support research organisations in becoming more effectivestrategic actors and thus defines an organisational role in supporting ERCgrants and grant getting In the attempt to use the ERC strategically by
97Funding Individuals Changing Organisations
establishing supportive and quality assuring structures (and thus to enlargetheir management capability) organisations and their units could counter-act the spirit of the ERC especially as they are challenged to provide aninternal peer review procedure that is in line with and can match therequirements of the ERC procedure
Impact on Powerless Universities
(Perceived) Performance and VisibilityThe impact of ERC on powerless organisations that host grantees is two-fold On the one hand grantees are rare and therefore the overall impacton the organisation can be regarded as low On the other hand the numberof general success stories within these organisations is by definition againrather low Therefore the visibility of the ERC successes tends to be highalso when it comes to the issue of (perceived) performance and translatesinto enhanced international reputation Due to the small number of casesin these organisations the impact of ERC grants is still limited to indivi-dual sub-fields On this level however ERC grants seem to make a real dif-ference and allow institutes and labs to internationally better positionthemselves and to gain reputation more generally also for the very goodresearch being done before or beside the ERC grant
Further ERC grants bring the opportunity to establish new fields orresearch avenues In view of weak performance and organisational rigid-ities the funding volume flexibility and freedom coming with ERC grantsoften allow these organisations to bypass traditional hierarchies and estab-lished disciplinary boundaries In this respect ERC grants seem to stronglycontribute to the creation of additional pockets of excellence within weakerorganisations Impact thus can be considerable in individual places withinthese organisations due to individual ERC successes
Strategic CapabilitiesThe organisations covered in our sample face a number of equally seriousproblems limiting their strategic options These limitations are about mea-ger financial endowment and low flexibility of spending rigid recruitmentstructures problems with recruitment more generally and a lack of top-down quality policies steering and priority setting processes Thereforeimpacts on strategic capabilities appear only at the level of individualresearch groups labs or institutes In powerless organisations individualERC grantees with considerable funds flexibility and freedom are truly
98 JAKOB EDLER ET AL
outliers as they are able to change the geometry habits and perceptions sig-nificantly within their immediate surroundings
On the level of individual grants some first indications show that someeffects including new research avenues and higher degrees of flexibility interms of research organisation or administration can be traced to the fewERC grantees Grantees serve as role models and generally become highlyvisible they stand out due to the dearth and immobility reported for theirgeneral situation The fact that ERC grants are regarded as lsquoseed moneyrsquoand contribute to lsquosetting prioritiesrsquo in weak organisations suggests a lowlevel of strategic research planning at all levels
Further as some powerless organisations saw the ERC grant(s) as animportant source of income they developed considerable efforts to rushscientists into application processes The subsequent lack of success led todisillusion and emphasised existing weaknesses In powerless organisationsthe goals of strategic leaders at all levels focused on other national andEuropean instruments to build up state-of-the-art research infrastructuresIn some of our interviews the importance of the Structural Funds and itsbureaucratic burdens were even used as arguments why there was notenough time and attention available for developing stronger ERCapproaches
For powerless organisations both with and without grantees a generalobservation can be made First there is no impact on the overall organisa-tional setup The organisations investigated share (i) a high degree ofdecentralisation and (ii) lack of vigour as regards priority setting recruit-ment provision of incentives or quality policy These practices have notgone through considerable changes due to ERC as such or through itsgrantees Overall in such organisations the symbolic and material value ofthe ERC alone is by far not sufficient to induce the desired change
Human Resource Policy Support Attraction and Retention of PromisingResearchersPowerless organisations are in general not in a position to offer favourableconditions for employment and usually lack systematic recruitment policiesAs a consequence also the perception of potential future key employeesabout this working environment tends to be rather low A powerless organi-sation is no place to be while attractive organisations can choose amongmany strong applicants or successfully approach top scientists
Change can be and is being introduced on a lower level but mainlyerratically and unplanned These organisations try to attract researchersfrom their own countries who have started or already successfully
99Funding Individuals Changing Organisations
mastered their career abroad One strategy to do so is to focus on pocketsof excellence in certain labs or parts of the organisation which show somepotential to become internationally attractive at least in a few sub-fieldsBoth strategies are in use as regards ERC and ERC grants Excellentexpatriates are lured back with the possibility of a (Starting) Grant andERC grants help in building-up pockets of excellence However evidenceis blurred due to the small number of grants and the competitive nature ofthe ERC
For powerless organisations that do host grantees they provide valuablerole models within and beyond the organisation The impact on organisa-tions is twofold First they have something to offer (in case of success paidby ERC) they could not offer on their own Second ERC grants allow forthe establishment of specific areas offering different dynamics More free-dom more flexibility more focus on competition and quality The down-sides again are small numbers and the largely passive role of theorganisation The effective decision is made first by ERC then by theapplying researchers and only to a minor degree by the organisation itselfHowever the reported irritations within organisations caused by thosegrantees may trigger a broader re-thinking of the organisation of researchand the conditions to be provided to top researchers within those organisa-tions more generally
For these kinds of organisations without grantees there is no impact tobe found for them the record of not hosting any grantee merely confirmsunfavourable conditions in the specific case under review The performancegap is just too high the requirements of ERC are too demanding the ERCis too far away to send any signals that could trigger any response to beinterpreted as impact One potential future impact here however notdetected in our interviews could be that certain retention policies need tobe introduced or strengthened in reaction to pressures to retain people thatmight be tempted to go to more attractive organisations in order toincrease their chances for ERC success
DISCUSSION
The following research questions have been asked during our exploratoryempirical study What changes do we see because of the ERC what is theawareness of strategic leaders within universities and how are the differentlevels within universities affected How do organisational and country
100 JAKOB EDLER ET AL
characteristics determine the impact of ERC and how do ERC inducedchanges relate to ongoing transformations and other pressures towardschange The exploration outlined in this article has provided a set ofanswers to these questions that are summarised below structured alongthe key dimensions of our conceptual model
Material Symbolic and Normative Value
Our first important finding is that the symbolic value of the ERC is alreadyimmense as the sheer existence of the ERC has made a difference ERCgrants assign prestige and symbolise excellence of individuals and of hostorganisations There is a strong awareness of strategic leaders as to thiseffect and the ERC has already acquired a high level of credibility ERCgrantees are used as a benchmark for quality between organisations andwithin organisations By and large this impact could be found at all orga-nisational levels Although our sample is biased towards grant getting orga-nisations it is obvious that the symbolic value is relevant to allorganisations across the European Research Area This symbolic effect isstrongest for powerful and in-between organisations that claim to competefor research excellence Powerless organisations see less symbolic value inthe ERC as for them the prestige excellence gap seems far too big Itwill be interesting to see if this effect winds down over time as the numberof ERC grants and grantees grows and the threshold of what is perceivedto be excellent might shift In any case the high normative value of ERCinstruments puts pressure on the strategic actorhood of universities whohave no choice but to support staff to get those grants
The material value is of different importance to different organisationsIn general the more funding pressure universities have the more the mate-rial value drives their reactions This is especially true for individual unitswithin organisations for which a grant can make all the differenceHowever we have also seen counter-productive efforts of organisationsthat are poorly endowed with funds to push their staff in a rather unpre-pared and erratic manner to apply for ERC schemes Though for very well-endowed organisations even considerable ERC funding only accounts for arelatively small share of their overall institutional budget they at the sametime recognise and emphasise the value of the grants beyond both thematerial and the symbolic value mainly providing additional capacity atlower levels
101Funding Individuals Changing Organisations
The impact related to the normative value is ambiguous The ERCthrough its generous grant conditions and its explicit aims propagates a cer-tain model of researchersrsquo freedom and empowerment within organisationsConsequently we found all kinds of adjustments to create favourable con-ditions for applicants and grantees as universities compete for grantees andpotential grantees These adjustments involve more autonomy for (young)researchers better contractual conditions and the build-up of teams aroundthem Those adjustments are more or less far-reaching depending on thepre-existing ambitions of organisations and the national funding environ-ment as organisations in countries with grants similar to ERC grantspartly have made respective adjustments already prior to ERCrsquos existenceWhile the changes are in line with researchersrsquo expectations in many casesthe change of HR policies and the increase in autonomy has meant disrup-tion of an existing balance between research staff within units (facultiesschools labs) and actually less authority of strategic leaders who have nochoice but to offer greater autonomy to grantees At the unit level this hasled to bottom-up growth and research foci whereas depending on theorganisational capability and ambition this may either be enhancing orendangering systematic research planning13 This observation is related tothe tension for the strategic leadership of universities between quality andprestige on the one hand (ie getting a grant and providing the conditionsfor its successful completion) and the aspiration of strategic control ofresearch profiles and staff equity on the other hand
Impact on Performance Strategic Capability and HR
Finally across all impact dimensions our findings show that the ERC is acatalyst in most cases reinforcing certain transformations rather than trig-gering them When the ERC actually triggers strategic organisational reac-tions it is on the lower organisational levels which are more often locallyisolated and opportunity driven rather than systematically organised andspreading across the organisation
Generally it is too early to assess any impact on performance yetHowever all indications and the assessment of interviewees point towardsan effect on performance as the ERC helps to create conditions for excel-lent researcher that are conducive to producing excellent results that will bemeasured in output and impact indicators in the future The competitionbetween organisations and units in organisations to get grants will contri-bute to enhance further the likelihood of excellent research being produced
102 JAKOB EDLER ET AL
The impact on strategic capability is mixed In many cases the impact isweak on the highest organisational level The ERC does not fundamentallychange the limitations of top-level leadership to manage research asresearch direction and field specific performance criteria cannot be influ-enced very strongly Further the ERC as such does not offer bundling ofresearch projects or lines and thus horizontal management to define newcooperative areas of research in organisations is not supported by theERC However particularly in the case of several grantees in one unitbundling happens and some organisations are able to strategically incorpo-rate this into overall research strategies
At the unit level within organisations the ERC grant has been used tomake strategic adjustments towards more risky and daring researchHowever in many cases especially in powerless organisations (radical)changes in research portfolios have been un-coordinated and entirelybottom-up whereby the offer of re-direction challenges top-down manage-ment of research Strategic capabilities are most influenced in the organisa-tions that are already committed to defining an excellence-based strategyand where there is still room for learning and improvement In these casesorganisational willingness and capacity to learn between units and betweenunits and the centre can be found For more powerless organisations ERCgrants create pockets of excellence (and disruption) that are not fosteredorganisationally
The impact on attracting and managing researchers is not straightfor-ward In powerful organisations incentive structures and support mechan-isms are often already in place and the ERC does not make muchdifference In in-between or powerless organisations the ERC grant canmake a big difference in terms of professionalisation of support structureand in terms of autonomy of and reward for staff Where there is not astrong culture of competition-based allocation of reward and autonomy tostaff yet the ERC can be perceived as a challenge to HR management dis-rupting established and often cherished procedures and team balance
The Importance of the Intervening Organisational andCountry Conditions
Our empirical analysis so far has demonstrated the importance of pre-existing organisational capabilities and prevalent conditions of national fund-ing and science systems We have found a connection between the existingstrengths and development of organisational capabilities (at various levels)
103Funding Individuals Changing Organisations
on the one hand and impact of the ERC on the other hand We havebundled the various conditions within organisations into three types thatshow sufficient similarities as to the intervening variables that we havedefined For the overall organisation (ie aggregated organisational impacts)we can summarise the impact of the ERC in the simplified stylised Fig 2
This figure is an over-simplification but it is meant to highlight thataggregate effects on organisations (ie the combination of lower and higherlevel effects) in general differ for our three types For different reasonsboth powerful and powerless organisations are less affected by the creationand operation of the ERC than in-between organisations All in all power-ful organisations already show ambitions structures and processes that areconducive to getting ERC grants and despite some adaptations on lowerlevels learning and adaptations in these organisations remain limited Forthe powerless organisations the performance gap and the lack of support-ing structures are prohibitive for a broad ERC impact to unfold Strongimpact on sub-unit level is not translated into organisational lessons at ahigher level By and large within in-between organisations which oftenhave developed a conscious strategy to improve as organisations there is agreater likelihood that an impact on lower levels leads to organisationallearning at higher organisational levels Furthermore as leaders in thoseorganisations recognise the leverage of the ERC for re-positioning theirorganisation on the European landscape the ERC functions as a strongcatalyst to set up new structures and processes
Obviously as the analysis has shown this relation has to be further qua-lified for individual sub-units Here the individual grant unfolds a strongermaterial symbolic and normative value and the management of research
Overall impact onorganisations
powerfull organisationspowerless organisations
Fig 2 Simplified Illustration of Role of Pre-Existing Organisational Conditions
on Overall Impact
104 JAKOB EDLER ET AL
is less distant from the rationalities of the epistemic community allocatingthe grant in the first place What matters here for the impact on the organi-sation as such in the medium and long term it seems is the extent to whichchanges at lower levels are transmitted to the upper level or to other unitsand furthermore incorporated into the overall organisational strategy
Regarding the intervening organisational and country variables it mightappear that organisational features are often more important than countrycharacteristics in that there are strong sub-units or organisations in countrieswith a somewhat weaker overall funding arena We have experienced a num-ber of cases where individual sub-units or outlier organisations follow differ-ent rules and mechanisms and thus employ different policies The evidence isstill mixed however and further application of the approach is neededIt seems that powerful organisations can flourish also in less supportivelandscapes and the uphill struggle for powerless and even in-between orga-nisations in a weak funding landscape is a bigger one However a certainminimum financial endowment and a minimum degree of legal freedom toact have to be in place Further we concede that this claim is based on astrong simplification In reality organisational development history and cul-ture legal framework and other factors are deeply entangled (and the ERCcould become one factor over time to untie some of the historical knots)
CONCLUSION THE IRONY OF ORGANISATIONAL
ACTORHOOD AND ERC SUCCESS
Universities have to search for money that is not bound to specific review-based activities if they are to function as independent organisations asthose are beyond their own reach to steer The ERC while providing repu-tation and prestige as well as money does not add to the organisationalauthority over resources It often strengthens the position of excellentresearchers and binds the granted money to a specific activity in someinstances even leading to further claims of the grantee This binds resourceseven further or leads to the establishment of groups that would not havebeen built within the strategic planning of the organisation or even of theunit otherwise There is thus an inherent tension an irony the more theprestige of the ERC is sought the less autonomy organisations actuallymight have The normative pressure (ie ways to organise research andgrant freedom to the grantees even the younger ones) rather appears to
105Funding Individuals Changing Organisations
limit and not to enhance organisational actor capability and the additionalresources granted bind other resources and trajectories are built that werenot planned This could potentially be used in a re-active adaptive strategybut essentially limits the degrees of freedom of the organisation as a whole
On the other hand some organisations have started to interfere withERC grantees and applicants For the good of the organisation they try tomaximise the success thus putting pressure on their staff and designingsupporting structures that are entirely re-active and at worst foil the veryidea of risky frontier research Internal peer reviews may be set up in orderto assure quality of ERC applications however these may not always becapable to assess risky ERC-type approaches Further the organisationalinterference exemplifies the tension between the organisational interest andthe individual logic of the ERC
NOTES
1 The official mission statement of the ERC stimulated our interest in how theERC might be affecting universities httperceuropaeumission
2 We are indebted for manifold feedback and suggestions for improvementsfrom colleagues in the EURECIA project led by Maria Nedeva and the projectrsquosAdvisory Committee (see wwweurecia-ercorg for a list of those colleagues and thefinal report Nedeva et al 2012)
3 We owe this idea of different ways the ERC exerts impact to discussions withMaria Nedeva Jochen Glaser and Dietmar Braun from the EURECIA project Theelaboration on those three ways of impact and thus any misconception is ours
4 For more details see for example ERC Grant Schemes Guide forApplicants for the Advanced Grant Call 2011 Guide for ERC Grant Holders (Part IPart II) all available at httperceuropaeudocument-library
5 The performance of organisations in terms of research output cannot yet bedetermined a proxy here will be the perception and expectation of performance andits justification in interviews
6 We acknowledge that there might be further impacts spillover effects in otherareas or unintended impacts but focus in our research on the three target dimen-sions defined by the ERCrsquos mission
7 Braunrsquos classification helps to understand the funding situation of universi-ties not their human resources policies or general governance structures Braunuses a combination of proxies that is research expenditures as a percentage ofGDP scientific articles per million population number of researchers per 1000employees and ERC success has been used (Braun 2011) While we fully agree onthe first three indicators we are a bit more sceptical about the use of ERC successrates in this context However his classification of countries to these categories washelpful for our purpose
106 JAKOB EDLER ET AL
8 We are grateful to our colleagues in the EURECIA project for supporting uswith a considerable number of interviews and related material for the cases in oursample
9 Note that our assignment of universities to the three baskets is based on amix of qualitative and quantitative variables including both country and individualorganisational characteristics which we gathered by means of upfront deskresearch literature review secondary data and our interview analysis Followingour iteratively developed research approach the baskets thus establish an integralpart of our results We are aware that there are a range of other classificationsdeveloped meanwhile most recently by Paradeise and Thoenig (2013) who use atypology based on a two by two matrix of excellence and reputation and thus deriveat four different types10 Further the allocation is based on an aggregate assessment for example
assigning University X to the top category can mean that it was relatively highranked for all three intervening variables but not necessary top ranked in all ofthem11 In this first approach we did not distinguish the different financing models
that is we did not systematically distinguish between organisations that rely moreon grants and those that can heavily or exclusively draw on institutional funding Inthe actual analysis however we did find differences and report on them12 While the baskets have worked as simplifying device in communicating broad
lines of impact the borders between the baskets are blurred This has to do with thevariety of capabilities within an organisation at the level of sub-units Thus it didnot prevent analysis of idiosyncrasies and special cases13 Note that within our empirical sample and at this early stage the type of
ERC grant (StG or AdG) made no systematic difference yet However there aresome first empirical indications that the different types of grants address differentdimensions of the researcher organisation nexus in terms of both the symbolicand normative values for example StG funding allows for new degrees of freedomfor young people that cause organisational turbulences that are not caused by AdG(as grantees most often are well established and integrated before the ERC grant)However this has to be further explored in the course of future investigation
ACKNOWLEDGEMENTS
This paper has its origins in the EURECIA project funded by the ERC asa CSA action within FP7rsquos IDEAS Programme (Grant No 229286)EURECIA set out to develop and apply a novel conceptual frameworkand methodology to investigate measure attribute and assess the impactand outcomes of the ERC and its funding schemes We are indebted formanifold feedback and suggestions for improvements to the leader of thatproject Maria Nedeva to the project team and the projectrsquos Advisory
107Funding Individuals Changing Organisations
Committee (see wwweurecia-ercorg for a list of those colleagues and thefinal report Nedeva et al 2012)
REFERENCES
Aghion P Dewatripont M Hoxby C Mas-Colell A amp Sapir A (2008) Higher aspira-
tions An agenda for reforming European universities Brussels Belgium Breughel
Blueprint Series
Bleiklie I (2011) New public management network governance and the university as a chan-
ging professional organization In T Christensen amp P Laeliggreid (Eds) Ashgate research
companion to new public management (pp 161176) Aldershot UK Ashgate
Bleiklie I amp Kogan M (2007) Organization and governance of universities Higher
Education Policy 20 477493
Braun D (2011) ERC impact on funding systems Reporting and synthesis deliverable D 93
for the EURECIA project Brussels Belgium Manchester
Clark B R (1998) Creating entrepreneurial universities Organizational pathways to transfor-
mation Oxford IAU PressPergamon
Enders J (Ed) (2001) Academic staff in Europe Changing contexts and conditions Westport
CT Greenwood Press
Estermann T amp Nokkala T (2009) University autonomy in Europe I exploratory study
Brussels Belgium European University Association Publications
Ferlie E Musselin C amp Andresani G (2008) The steering of higher education systems A
public management perspective Higher Education 56 325348
Franzoni C Scellato G amp Stephan P (2012) Foreign born scientists Mobility patterns for
sixteen countries National bureau of economic research Working Paper 18067
Retrieved from wwwnberorgpapersw18067 Accessed on February 15 2014
Fumasoli T amp Lepori B (2011) Patterns of strategies in Swiss higher education institutions
Higher Education 61(2) 157178
Geuna A amp Martin B (2003) University research evaluation and funding An international
Comparison Minerva 41 277304
Granberg A amp Jacobsson S (2006) Myths or reality A scrutiny of dominant beliefs in the
Swedish science policy debate Science and Public Policy 33(5) 321340
Jongbloed B (2009) Higher education funding systems An overview covering five European jur-
isdictions and the Canadian province Ontario Report prepared for the Hong Kong
University Grants Committee Enschede The Netherlands
Krucken G Kosmutzky A amp Torka M (Eds) (2007) Towards a multiversity Universities
between global trends and national traditions Bielefeld Germany Transcript Verlag
Laredo P (2007) Revisiting the third mission of universities Toward a renewed categoriza-
tion of university activities Higher Education Policy 20(4) 441456
Laudel G amp Glaser J (2010) Experteninterviews und qualitative Inhaltsanalyse Wiesbaden
Germany Springer Fachmedien
Mayring P (2008) Qualitative Inhaltsanalyse Grundlagen und Techniken Basel Switzerland
Beltz Verlag
108 JAKOB EDLER ET AL
Musselin C (2005) Le marche des universitaires France Allemagne Etats-Unis Paris
France Presses Sciences-Po
Musselin C (2007) Are universities specific organisations In G Krucken A Kosmutzky amp
M Torka (Eds) Towards a multiversity Universities between global trends and national
traditions (pp 6384) Bielefeld Germany Transcript Verlag
Nedeva M (2007) New tricks and old dogs The lsquothird missionrsquo and the re-production of the
university In D Epstein R Boden R Deem F Rizvi amp S Wright (Eds)
Geographies of knowledge geometries of power Framing the future of higher education
(pp 85103) World Yearbook of Education 2008 New York NY Routledge
Nedeva M amp Stampfer M (2012) From lsquoscience in Europersquo to lsquoEuropean sciencersquo Science
336(6084) 982983
Nedeva M et al (2012) Understanding and assessing the impact and outcomes of the ERC and
its funding schemes (EURECIA) Final Synthesis Report Brussels Belgium
Manchester UK Retrieved from httperceuropaeusitesdefaultfilesdocumentfile
eurecia_final_synthesis_reportpdf Accessed on February 15 2014
OEU (Observatory of the European University) (2006) Methodological guide Prime OEU
Guide Lugano Retrieved from wwwenid-europeorgPRIMEdocumentsOEU_guide
pdf Accessed on February 15 2014
Oquist G amp Benner M (2012) Fostering breakthrough research A comparative study
Stockholm Sweden Akademirapport Kungl Vetenskapsakademien
Paradeise C Reale E Bleiklie I amp Ferlie E (Eds) (2009) University governance Western
European comparative perspectives Dordrecht The Netherlands Springer
Paradeise C Reale E amp Goastellec G (2009) A comparative approach to higher educa-
tion reforms in Western European countries In C Paradeise E Reale I Bleiklie amp
E Ferlie (Eds) University governance Western European comparative perspectives
(pp 197226) Dordrecht The Netherlands Springer
Paradeise C amp Thoenig J C (2013) Academic institutions in search of quality Local orders
and global standards Organization Studies 34(2) 189218
Schmoch U amp Schubert T (2010) Strategic steering of research by new public management
in German universities A looming statescience conflict Research Evaluation 19(3)
209216
Van Vught F et al (2005) Institutional profiles Towards a typology of higher education insti-
tutions in Europe Enschede The Netherlands University of Twente
Whitley R (2008) Universities and strategic actors Limitations and variations In L Engwall amp
D Weaire (Eds) The university in the market (pp 2337) London Portland Press
109Funding Individuals Changing Organisations
WHERE HAVE ALL THE
SCIENTISTS GONE BUILDING
RESEARCH PROFILES AT DUTCH
UNIVERSITIES AND ITS
CONSEQUENCES FOR RESEARCH
Grit Laudel and Elke Weyer
ABSTRACT
This article investigates the links between universitiesrsquo opportunities toshape their research profiles the changing state interest concerning theseprofiles and the impact of profile building on research at university andfield levels While the authority of the Dutch state over research profilesof Dutch universities has increased university management has consider-able operational authority over the inclusion of new research fields andremoval of existing research fields Since all universities have begun tofollow the same external signals prescribing applied research researchthat has easy access to external funding and research in fields prioritised
Organizational Transformation and Scientific Change The Impact of Institutional
Restructuring on Universities and Intellectual Innovation
Research in the Sociology of Organizations Volume 42 111140
Copyright r 2014 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN 0733-558Xdoi101108S0733-558X20140000042004
111
by the state a lsquoquasi-market failurersquo may emerge as is demonstrated forevolutionary developmental biology and Bose-Einstein condensation
Keywords New public management university research profilesresearch fields scientific communities the Netherlands
INTRODUCTION
The new public management (NPM) reforms have been studied by highereducation researchers mainly with regard to their impact on universitieswith a focus on the lattersrsquo autonomy and action capabilities (Clark 1998De Boer Enders amp Leisyte 2007 Marginson amp Considine 2000) Theimpact of these reforms on the conduct and content of research has enjoyedfar less attention with empirical studies focusing on the micro-level that ison researchers and research groups (Glaser amp Laudel 2007 Glaser LangeLaudel amp Schimank 2010 Jansen 2010 Leisyte 2007) Owing to the meth-odological difficulties of causally linking micro-level changes in researchpractices to macro-changes of research fields in a country such micro-levelimpact studies can only speculate about the impact of NPM reforms on theresearch of a country In this article we contribute to closing this gap byaddressing possible macro-level effects of an aspect of NPM that hasnrsquotenjoyed much attention so far namely the building of research profiles byuniversities and its impact on research fields at the national level
These effects are of both theoretical and political interest Theoreticallystudying the effects of changes in organisations on research fields reversesthe perspective that has dominated organisational sociology so far Therecent interest of organisation studies in communities has treated them pri-marily as a context of organisational activities for example as the domi-nant context (OrsquoMahony amp Lakhani 2011) or as a resource fororganisations (Wenger amp Snyder 2000) The example of open source soft-ware production has been used in an attempt to establish communities as adistinct organisational form of production (Seidel amp Stewart 2011) Ourinvestigation takes changes in organisations as the independent variableand asks how changes in organisations in the aggregate can affect commu-nities that are distributed across many organisations
There is also the theoretical question of what universities can actually dowith their research Investigations of the nature of the university as an orga-nisation and of properties of research processes suggest that a universityrsquos
112 GRIT LAUDEL AND ELKE WEYER
capabilities to shape its research can be increased by NPM reforms but stillhave fundamental limitations NPM reforms are supposed to turn universi-ties into corporate actors with full strategic action capability which in-cludes an increased capability to steer their core processes of research andteaching At the same time there are theoretical reasons which make thesuccess of these steering attempts rather unlikely the lsquolack of transparencyrsquoof the universityrsquos core processes (Musselin 2007) the uncertainties inher-ent to research which make scientific communities the lsquocontrol centrersquo ofresearch (Whitley 2008) the inability of universities to define researchtasks to control work flow interdependencies and to control quality anduse of results caused by properties of the researchersrsquo work (Glaser 2012)Building research profiles seems to be as far as a university can go in thesteering of its research (Von Stuckrad amp Glaser 2012 Whitley amp Glaserthis volume) which makes this process a theoretical limit of NPM reforms
Politically the possible effects of NPM reforms on research fields havebeen discussed with regard to threats to the diversity of research Theseeffects have been shown to exist in British economics (Lee 2009) and havebeen hypothesised for the interdisciplinary field of innovation studies(Rafols Leydesdorff OrsquoHare Nightingale amp Stirling 2012) In Germanythe conference of university rectors warned against a possible disappearanceof lsquosmall subjectsrsquo particularly in the humanities due to similar processes ofprofile building in German universities (Hochschulenrektorenkonferenz2007) In the Netherlands the Dutch Academy of Science pointed out thatuniversitiesrsquo responses to profile building could result in blank spots in theresearch portfolio on the national level with modern languages probablybeing the earliest victims (KNAW 2013)
Effects of profile building on research fields can be expected from theinteraction of two distinct trends in NPM reforms namely the pressure bygovernments on their universities to develop distinct teaching and researchprofiles and the strengthening of hierarchical structures in universitieswhich are supposed to enable that profile building (Meier amp Schimank2010 Whitley amp Glaser this volume) These trends represent a departurefrom past practices which of course always included decisions that shapedthe profile of a university Investment decisions decisions on the denomina-tion of chairs and recruitment decisions have consequences for the fieldsrepresented in a university and the topics addressed within these fieldsSimilarly the decisions of professors on their research topics contribute toshaping the research profile of a university However these decisions havenot always been part of a lsquomaster planrsquo They were made locally and ad
113Where Have All the Scientists Gone
hoc and included a lsquolist of disconnected choices made by individual profes-sors with respect to their theoretical and methodological interestsrsquo or otherforms of voluntary bottom-up coordination (Meier amp Schimank 2010p 214) University profiles emerged as an aggregate effect
The creation of research profiles as an intentional strategy of the univer-sity to lsquoset collectively binding rules for the membersrsquo is a relatively newdevelopment triggered by NPM reforms (Meier amp Schimank 2010p 213) Many universities are now able to manage their research and teach-ing profiles by allocating resources between competing components ofteaching and research activities which are treated as investment portfolios(Whitley 2008 p 26) The creation or modification of profiles may stilloccur for a variety of reasons including responses to external expectationsinternal initiatives by academics who want to extend their resource baseand financial considerations that are unrelated to interests concerningresearch content The new intentional building of profiles that has beenadded to this mix is usually directed towards lsquoenhancingrsquo the profile whichincludes a reduction of topics and a redistribution of resources betweenfields with the aim of concentrating resources These processes have beenhardly investigated and their effects are poorly understood
The aim of our article is to explore the links between organisationalsituations of universities and faculties their profile-building strategies andeffects of such strategies on research at universities as well as research fieldson the national level The Netherlands provide an excellent laboratory forthe analysis of such consequences because their relatively small size makesnational fields very sensitive to decisions at individual universities Thissituation enables the causal attribution of changes of research fields toprofile-building activities of universities
Our argument draws together findings from several studies of the Dutchuniversity system and will be developed in the following steps We will firstdescribe the evolution of state expectations concerning the profile buildingof Dutch universities and demonstrate that these expectations have not onlygrown in strength but have been substantially transformed from the expecta-tion that universities may develop any profile to the expectation of specificcontent namely topics that contribute to specific political goals (third sec-tion) We then analyse profile-building activities and their effects at Dutchuniversities and develop an empirical typology of strategies of profile build-ing (fourth section) Moving to the field level we demonstrate consequencesof these decisions for three fields including two that are now developingcontrary to international trends (fifth section) As a conclusion we discussthe field-level effects of profile building as a quasi-market failure of a system
114 GRIT LAUDEL AND ELKE WEYER
in which universities must use profile building to compete for money (sixthsection)
THEORETICAL FRAMEWORK AND METHODOLOGY
The literature is rather opaque in providing definitions of research profilesand profile building that can be operationalised For the purposes of thisstudy we define the research profile of an organisation or organisationalunit as the distribution of research effort across a spectrum of researchtopics and their epistemic characteristics Thus a research profile dependson the research topics addressed by researchers of the organisation and theresources allocated to research on these topics Both aspects of a profilecan vary relatively independently of each other and are changed both unin-tentionally and increasingly by actions that intentionally aim at creating aspecific profile (profile building) Profile building at universities can includea wide range of activities from simply renaming research or defining newheadings for old research to significant investments in research centres fortopics not previously present at the university or the closing down ofresearch on topics that are no longer deemed useful for a universityrsquos pro-file Research profiles can be described by their content (research themes)their diversity (number and breadth of themes) and various epistemic char-acteristics of the research including its basic versus applied nature interdis-ciplinarity and collaborations
So far research profile building by universities has been investigated pri-marily in the German context Meier and Schimank (2010) investigated theprofile building at universities in the German state of Lower Saxony whichwas initiated by an evaluation of the university research of that state Theyfound that traditional bottom-up processes of profile building had becomesupplemented by top-down processes which were enabled by the increasedaction capabilities of German university leadership Owing to the still earlystages of the profile-building activities they observed little could be saidabout effects on the research profiles of universities Probably for the samereason Langersquos observation of attempted top-down profile building byfaculties at one German university showed nearly no effects (Lange 2007pp 164165) Germanyrsquos excellence initiative has also been consideredbecause one of its aims is to stimulate the building of distinct research pro-files (Leibfried 2010 Meier amp Schimank 2010 p 233) Again little can besaid about the effects because there is no conclusive evidence yet The sameholds for a recent study of the building of research profiles at Finnish
115Where Have All the Scientists Gone
universities (Pietila 2014) for a descriptive account of Dutch and Germanprofile-building activities (Klumpp De Boer amp Vossensteyn 2013) and fora bibliometric investigation of a Dutch government initiative termed lsquofocusen massarsquo (focus and mass) by Van den Besselaar and Horlings (2010)
While the knowledge about effects of profile-building activities is stillvery limited the studies quoted above suggest that a variety of considera-tions may inform decisions on research profiles and the profiles emergingfrom these decisions These include at least external expectations towards auniversityrsquos profile strategies formed at different levels of the universityfinancial interests and perceived necessities and interests as well as capabil-ities of researchers Thus the building and shaping of research profilestakes place in a complicated actor constellation involving external actors aswell as actors at several levels of the university hierarchy
Our analysis of decision processes concerning university research profilesuses the authority relations perspective (Whitley 2010) which is a promis-ing way to conduct an integrated assessment of changes in public sciencesystems With this approach we can systematically analyse how authorita-tive agencies (the state research organisations organisational elites exter-nal funding agencies national as well as international scientific elites andthe researchers themselves) exercise authority over specific matters ofgovernance in our case research goals Its basic assumptions are (a) thatthe changes that public science systems go through have implications forthe relationships between actors and the way that they are able to realisetheir interests and (b) that authority relations as regards the selection ofresearch goals are the main channel through which changes in the knowl-edge production system are effectuated (Whitley 2010) We consider theauthority perspective to be particularly useful for our purpose because itenables the integration of multiple overlaying governance processesthrough which a set of actors contributed to the rise and fall of scientificfields in universities
We focus on the situations of universities in which profile-building stra-tegies are developed their decision processes on research profiles and theimpact of these decisions on research at the university as well as on fields atthe national level Concerning the situations of universities the expectationsby external stakeholders (mainly the state) and their authority and the uni-versityrsquos financial situation are of particular importance Decision processeson research profiles vary in their aims as well as the actors involved andtheir relative authority Concerning the outcomes we pay particular atten-tion to the actual changes of the profile including secondary effects acrossthe university Concerning the national level we discuss aggregate effects
116 GRIT LAUDEL AND ELKE WEYER
which can be described as unanticipated consequences for the nationalresearch profile of the country
Our analysis is based on case studies in Dutch universities which albeitconducted for different purposes all included decision processes about uni-versity research profiles or the consequences of such decisions The empiricaldata stem from three research projects The first project seeks to find outwhether the organisational transformation of Dutch and English universitiescan be explained as transition between organisational archetypes of authoritystructures (Weyer 2014) The second project studied the impact of changingauthority relations in four European countries on conditions for intellectualinnovations in the sciences social sciences and humanities (RHESI see theintroduction to this volume) Finally a comparative project on the impact ofnational career systems on opportunities for researchers to change lines ofresearch also contributed data on authority structures in Dutch universitiesand their impact on university research profiles (Laudel 2012)
Case studies in all these projects were based on documentary analysis andsemi-structured interviews (more detail on the methodology of the investiga-tion of evolutionary developmental biology and Bose-Einstein condensationcan be found in the articles by Laudel et al this volume) Documentarysources include university newspapers internet pages minutes of universitydecision-making bodies as well as policy documents Interviews were con-ducted with members of the university top management mid-level managersheads of department and academics at different levels Interviews withresearchers were prepared by reading research-related documents (web sitespopular descriptions of research publications project proposals CVs) andindividual-level structural bibliometrics (Glaser amp Laudel 2009) The inter-views were analysed with qualitative content analysis (Glaser amp Laudel2013) Analyses for the purposes of this article focused on authority relationsconcerning university profiles and consequences of profile building for thegrowth and shrinking of fields at universities
THE EVOLUTION OF DUTCH STATE EXPECTATIONS
CONCERNING UNIVERSITY RESEARCH PROFILES
First Phase Changes in Profiles as Unintended Effects
State expectations of academic research started to rise at the end of the1970s The Dutch government wanted public research to be to a certainextent (nationally) programmed and it wanted to lsquoenhance the quality the
117Where Have All the Scientists Gone
practical value and the accountability of researchrsquo (Blume amp Spaapen1988 p 26) This is why research evaluations were introduced in 1983
The government asked universities to submit so-called lsquoresearch pro-grammesrsquo that is coherent research plans for (initially) five years and ofgroups of at least five researchers in order to receive funding for thisresearch Part of the funding of university research was made conditionalon the approval of the research programmes by peer review committees(Blume amp Spaapen 1988 p 11 Van der Meulen 2007) Although the ori-ginal idea was to conduct ex-post evaluations of the funded programmeslater on in order to link further funding to their success this never hap-pened in practice
While this exercise had some consequences for the research profile ofuniversities they were not intended and still fall under the category oflsquounintended consequences of local decisionsrsquo described by Meier andSchimank (2010) Changes in research profiles might have occurred in uni-versities and at the national level for three reasons
The necessity to provide five-year research programmes for groups of atleast five researchers may have induced local collaborations
The success of programmes in being approved for funding was influ-enced by the fit between a disciplinersquos epistemic practices and the needto have long-term plans for larger groups
The varying success rates of programmes from different disciplines between 36 and 100 in the sample investigated by Blume andSpaapen (1988 p 21) effectively led to a redistribution of fundingbetween disciplines
None of these effects whose strength is difficult to assess due to a lackof data were intended or planned by either the state or universities Thewhole evaluation exercise was still a bottom-up process in which researchgroups chose their own research areas
The first attempts of the state to initiate profile building by specialisationand concentration in universities occurred in 1983 (lsquotaakverdeling en con-centratiersquo division of labour and specialisation) and 1986 (lsquoSelectieveKrimp en Groeirsquo selective shrinkage and growth) These policies led to theclosure of some and amalgamation of other small university departments(Blume amp Spaapen 1988 p 6) The aim of the policies was to increase effi-ciency in higher education (Westerheijden De Boer amp Enders 2009p 109) The state did not prioritise any themes
Since these first initiatives Dutch universities constantly changed thestructures and thematic labels in which research was conducted The
118 GRIT LAUDEL AND ELKE WEYER
resulting structures are lsquovirtualrsquo aggregations of research (lsquoresearch schoolsrsquolsquoresearch institutesrsquo lsquoresearch programmesrsquo etc) that overlay the more per-manent faculties (although mergers of faculties occurred as well) Facultiesare largely considered to be the units responsible for teaching which is whythe various research units create a matrix structure of the Dutch universityOur interviews confirm that until the 2000s the impact of these structuralchanges on the content of university research remained limited (see alsoLeisyte Enders amp De Boer 2010)
Second Phase Political Pressure on Profile Building
It was only in the early 2000s that the Dutch government took concretemeasures to bring about the lsquorightrsquo content in university research It intro-duced its lsquofocus and massrsquo policy The argument is that the Netherlands istoo small to excel in all research areas Therefore resources should be con-centrated in areas where the Netherlands are excellent and which areimportant from a socioeconomic point of view (Van den Besselaar ampHorlings 2010 p 13) The concentration should be supported by the foun-dation of multi-disciplinary networks and consortia (ie build lsquocriticalmassrsquo)
To support its focus and mass policy the government initiated two paral-lel measures first public higher education institutions should develop dis-tinct research (and teaching) profiles Second the government establishedlsquotop sectorsrsquo in which research efforts should be concentrated and researchcollaboration with industry be stimulated Taking the lsquostrongrsquo researchareas of the Dutch economy as a point of departure public and private sta-keholders identified nine top sectors in the following areas Agro-FoodHorticulture and Propagating Stock High-Tech Materials and SystemsEnergy Logistics Creative Industry Life Sciences Chemicals and Water
About h15 billion was earmarked for the nine top sectors mostly byreallocating existing budgets of various ministries that manage an innova-tion budget (Technopolis 2011 pp 7 11) Particularly noteworthy in thisrespect is the decision made by the government to spend roughly half ofthe budget of the Dutch Research Council NWO on research in the top sec-tors This re-allocation raised concerns about worsening conditions for fun-damental research as voiced by the Dutch Academy of Science lsquoin essencethere is room for fundamental research in the economic top sectors butwhether that room will be made available remains to be seenrsquo (KNAW2013 p 13)
119Where Have All the Scientists Gone
Third Phase Using Performance Agreements for Prescribing Content
In 2010 a governmental advisory committee (the Commission Veermanled by the former minister of agriculture Cees Veerman) pointed out theneed for profile building in Dutch higher education By picking up thefocus and mass argument the Commission reiterated that universities couldnot excel at everything and should therefore try to concentrate on theirstrengths in order to remain competitive with other knowledge economiesThe Ministry of Education Culture and Science followed the recommenda-tions of the Veerman report by including the demand for profile building inteaching and research in the Strategic Agenda for Higher EducationResearch and Science (September 2011) Although universities were for-mally free to decide whether to participate in the nine top sectors or nottheir active involvement was nevertheless clearly expected by thegovernment
The process of profile building in Dutch higher education was givenshape in the so-called performance agreements that were agreed uponbetween the Secretary of State and all universities In the framework ofthese performance agreements universities committed themselves to thedevelopment of unique teaching profiles and research profiles
Participation in the performance agreements was voluntary but hadfinancial consequences To ensure that universities would take the call forprofile building seriously the government earmarked h105 million (5 ofthe total higher education budget) as performance-based funding In addi-tion h 38 million (2) were to be distributed among those universities hav-ing handed in the best plans as evaluated by the ministry By means ofcoupling at least some part of the funding to specific performance expecta-tions the government reserved itself the right to withdraw this money againif universities should not reach the targets agreed upon in the performanceagreements
The evolution of profile-building activities in the Netherlands demon-strates interesting shifts in authority relations which in some respects con-tradict the central tenet of NPM reforms namely increasing the autonomyof universities with respect to formulating their own goals The originalintention of Dutch higher education policy back in the 1980s was that uni-versities should build profiles that is should concentrate their resources inselected areas which were supposed to differ between universitiesHowever at this time there were no political expectations concerning thecontent of these priority areas at universities These expectations whichincreasingly took the form of financially supported political directions
120 GRIT LAUDEL AND ELKE WEYER
emerged in the early 2000s Clear expectations concerning the content ofprofiles were formulated and reinforced first by financial incentives in exter-nal funding and later by performance agreements and linked universityblock funding Although the Dutch publicly funded universities are in prin-ciple free to lsquoignorersquo the call for profile building their increasing financialdifficulties create a strong pressure to take the financial rewards It is henceworth investigating how universities respond to these calls when decidingabout the prioritisation of research activities
PROFILE BUILDING AT DUTCH UNIVERSITIES
Strategies of research profile building include all decision processes at theuniversity that are aimed at changing the distribution of a universityrsquosresearch effort across the spectrum of research fields The two outcomes ofprofile-building activities include a description of research topics and anallocation of resources to topics
Devolved Profile Building at University 1
The first case of profile building emerged as a response to the Dutch gov-ernmentrsquos call for creating focus and mass in university research and tocuts in university block funding The process was initiated by the univer-sityrsquos executive board1 as part of the universityrsquos strategic planning Thestrategic plan included the introduction of clear-cut criteria for the develop-ment of new lines of research and a critical review of the current researchportfolio which should be modified if necessary These aims should beachieved by the creation of faculty wide lsquofocus areasrsquo in all faculties Theprocess was supervised by the senior management team (consisting of theexecutive board and the deans) and was supported by start-up funding(conditional on the faculties providing plans for their profile-building pro-cess) In addition several budget tranches were distributed among thefaculties during the subsequent three years to support the process of profilebuilding
The selection of lsquofocus areasrsquo was largely a faculty affair The executiveboard had formulated some general expectations (eg the stipulation ofexcellent research lines) which were discussed with faculty management
121Where Have All the Scientists Gone
The number and content of focus areas was at the discretion of the facul-ties each of which applied its own strategies
Faculty One Further DevolutionOne of the faculties the faculty of the humanities further devolved theprofile-building process by delegating the choice of focus areas to the lea-ders of the existing lsquoresearch programmesrsquo that is the current thematicunits of research The research programme leaders discussed the profilebuilding with the researchers in their programmes After this consultationprocess the research programme leaders decided that the facultiesrsquo researchprogrammes were to become the new focus areas The only actual changewas that within the research programmes (now focus areas) sub-focishould be formulated of which some tended to be slightly narrower in theirfocus than had previously been the case These narrower descriptions didnot affect the research itself Content-wise the facultyrsquos research portfolioremained by and large the same
This left the task of distributing the funds that were allocated for theprofile-building process The faculty management required the researchprogramme leaders to submit budget plans in which they explained howthey intended to use the funds The faculty management thereupon spreadthe amount of profile-building funds almost evenly across all research pro-grammes (now lsquofocus areasrsquo)
Faculty Two Participative Decision-Making at the Faculty LevelAs part of the same university-wide profile-building process anotherfaculty a science faculty applied different strategies All decisionsabout the facultyrsquos new research profile were made by the faculty manage-ment albeit in consultation with the universityrsquos executive board and seniorfaculty staff (research group leaders heads of departments etc)
While the research profile of the faculty was not changed thematicallythe profile building and the money made available for it were used selec-tively In a first phase the decision was made to strengthen the position oftwo of the facultyrsquos research areas which became the new lsquofocirsquo of thefaculty One area was a recent lsquoacquisitionrsquo a group that had moved fromanother faculty and had brought some lsquoprofile-building moneyrsquo with it Theother one was a highly successful research group in need of investmentSince the group leader had offers from other universities the faculty waskeen to provide the funds in order to make him stay The money offeredfor the profile-building process provided a good opportunity to integratethe first group and to strengthen the other which happened by channelling
122 GRIT LAUDEL AND ELKE WEYER
all the funds available for profile building to these areas Although not allstaff members were satisfied with the faculty management decision to usethe funds selectively the faculty management could convince most of themnot least by promising that the regular distribution of funds would be leftuntouched
Bottom-Up Generation of Options and Top-DownConfirmation at University 2
In University 2 too the profile-building process was initiated by the uni-versityrsquos top management which asked the faculties to search for agreater concentration in their research portfolios In particular it wasargued that if the university wanted to maintain a strong position inter-nationally it would have to channel its resources into a number of clus-ters and disinvest from some areas The selection of focus areas wasconducted as a centrally led process in which the faculties were expectedto cooperate with each other in the creation of inter-faculty focus areasThe Executive Board did not provide financial support for the selectionprocess itself but promised financial rewards once the final selection offocus areas had been made
The selection process began as a bottom-up process The deans askedthe leaders of the research institutes (the current organisational units forresearch at university 2) to organise workshops among their research-activeacademics at which focus areas were to be suggested On the basis of theseinitial ideas researchers were expected to collaborate with their colleaguesfrom other faculties in the development of multi-disciplinary proposals thathad the potential to become part of the Universityrsquos new research profileThe proposals were collected by the deans and submitted as a proposal offocus areas to the Executive Board All proposed areas were multi-disciplinary and envisaged research collaboration among a considerablenumber of (senior) researchers from at least two faculties The majority oftopics were application-oriented and focused on issues of socioeconomicrelevance Some also addressed more fundamental questions In contrast touniversity 1 the choice of focus areas was selective in the sense that not allresearchers of a faculty were involved in them their own research activitieshence coexisted with the new focus areas
Having received the proposals the executive board asked an internalcommittee of active and emeritus professors from all faculties to assessthese plans Thereafter the list was presented to an external committee that
123Where Have All the Scientists Gone
provided an assessment of chances and risks of each proposed focus areaNeither committeersquos recommendations led to changes of the proposalwhich was finally approved by the executive board
To support cross-faculty collaborations in the new focus areas theExecutive Board provided a start-up budget for each focus area that wastopped up by the Faculty Researchers of the focus areas were then askedto double this original investment from external grants The resulting bud-get could be used for the acquisition of additional personnel and researchequipment
Closing Research Groups at a Faculty of University 3
A faculty at university 3 responded to a growing budget deficit which hasbeen accumulating since the late 1990s The faculty responded by a decade-long process of restructuring and profile building with the aim of having lessand less costly research The dean appointed an internal committee whichproposed a plan for the restructuring of biological research at the facultyThe committee suggested an organisational structure that was supposed tolead to a more homogenous research profile to enable the creation of lsquocriti-cal massrsquo and to ensure the efficient use of infrastructure Since some groupsdid not contribute to the new profile it was suggested that they should beclosed Among the areas to be closed was a subfield in biology
The original plans were opposed by the facultyrsquos science policy commit-tee especially with regard to the closure of the biology group As a com-promise another group was closed and members of the biology groupwere asked to retire early This cost-saving strategy is possible at Dutchuniversities and is applied occasionally Dutch researchers go into earlyretirement receive pensions and continue to work at the university Thereis an obvious financial advantage for the university and little disadvantagefor the academics if they are close to retirement anyway If a faculty is infinancial difficulties the existence of such an opportunity creates a strongpressure on those who can take it If they do not agree there is the possibi-lity that some other unit or some of their colleagues are made redundant
The senior staff of the biology group agreed to help their departmentwith its financial problems However their early retirement turned out tobe only a temporary solution for keeping the original profile When thefaculty ran into financial problems again the department and faculty feltunable to honour their previous agreements The head of the biologydepartment strongly backed by the dean of faculty and the financial
124 GRIT LAUDEL AND ELKE WEYER
officer imposed the original plans For this purpose the faculty created acommittee with lsquooutside expertsrsquo who again had the task of recommendinga profile and a structure for the department
Several factors indicate that the restructuring was triggered off by finan-cial problems and that it served the purpose of solving them rather thancreating a specific profile Thus an important criterion for including orexcluding research topics in the second round of profile building was theamount of external funding acquired by the different research groupsResearch performance on the other hand did not seem to be a selectioncriterion During the last research assessment the international reviewcommittee had positively assessed the biology grouprsquos research and hadrecommended to keep this group These recommendations were disre-garded in the decision-making process
Removing a Discipline from University 4
This section is based on an extensive description of a profile-building pro-cess published after the closure of Utrecht Universityrsquos institute for astron-omy by its last director (Keller 2012) The account was triangulated byother sources (blogs of researchers press releases of the university) andadditional inquiries
The decision on the facultyrsquos profile was a response to a budget deficitWhen the universityrsquos faculty of science was created by merging severalscience departments some of them brought substantial budget deficits intothe new faculty The department of astronomy and physics did not but wasnow part of a faculty that started with financial problems
Apparently unconvinced by the facultyrsquos own plans for overcoming itsproblems the executive board of the University of Utrecht commissionedtwo lsquooutside expertsrsquo to provide external advice on how to solve the pro-blems of the faculty Their recommendations deviated from the facultyrsquosown perceptions and plans The dean of the faculty resigned and aninterim dean was appointed for seven months The interim dean appointeda committee (whose membership was kept secret) to advise him on the dif-ferent institutes of the faculty This committee advised the dean thatastronomy should remain in Utrecht but that the previously planned addi-tional investments for the institute should not take place The instituteachieved an agreement with the interim and the prospective dean accord-ing to which the existence of the institute of astronomy was secured Theinstitute for astronomy had received a very good assessment in its last
125Where Have All the Scientists Gone
external evaluation which was confirmed by a public internal ad hocevaluation The results of an additional ad hoc evaluation commissionedby the dean were kept secret Keller quotes the deanrsquos answer to his criti-cism of the secret nature of this evaluation
Yes as a scientist I respect the wish of scientists to be evaluated according to a well-
defined objective quantitative procedure In contrast most of the decisions that I will
have to take concerning which units of the faculty we will discontinue do not strictly
depend on such a procedure (Email of the Dean of the Faculty of Science at Utrecht
university to C U Keller 2012 p 4)
For reasons that are unknown to Keller (he suspects an intervention bythe universityrsquos executive board) the previous agreement was overturnedand the dean of the faculty decided to discontinue astronomy Accordingto Keller the dean provided only very vague reasons related to the contentor quality of research and teaching The institute was split into three partsthat were moved to three other Dutch universities
The loss of a strong university institute had consequences for local colla-borations which were epitomised by a large and prestigious theoretical phy-sics group leaving the University of Utrecht after the institute forastronomy was closed The group leader explicitly referred to the closure ofastronomy as a reason for her move
My institute does not operate in an isolated environment the elimination of other excel-
lent groups in the department of Physics and Astronomy affects us considerably
(Digitaal Universiteitsblad 4 April 2012 our translation)
The researchers themselves (here the physicists) exercised authority overtheir research goals in that they moved away from unfavourable researchconditions produced by the university through its profile-building activitiesThe reduction of theoretical physics in quantity and quality within theorganisation was an unintended effect of these activities
Authority Relations and Strategies of Profile Building
A comparison of the processes of profile building shows a pattern inauthority relations concerning research profiles In none of the cases didresearchers lose the authority over their research goals Some of them lostthe opportunity to continue their research at their university due to beingforced to leave or voluntarily leaving in response to worsening conditionsThis is however different from having other authoritative agencies
126 GRIT LAUDEL AND ELKE WEYER
intervening in the formulation of research goals or tying the provision ofresources to specific goals
While profile building does not lsquoreach throughrsquo to the formulation ofresearch goals it decides on the presence of research areas and their accessto resources which are important conditions for the formulation ofresearch goals Three distinct combinations of thematic structuring andresource allocations occurred in the cases presented above
Purely Thematic Re-ContextualisationThis strategy consists of developing a new description of research activitieswithout changing them For example links between research fields can beemphasised by subsuming them under new headings or links to externalexpectations can be emphasised by describing existing research in a newway This strategy can be applied to create a seemingly new profile becauseresearch is a multivalent process and thus can be linked to a number of dif-ferent descriptions without being misrepresented For example a universitycan pretend to give priority to some lsquonewrsquo topics without actually changingthe allocation of resources This strategy was applied by the humanitiesfaculty in university 1
Thematic and Financial PrioritisationThis strategy links new descriptions of research to changing the relativeweight of topics in the new profile which is supported by a matching allo-cation of resources The latter can happen either by mobilising additionalresources or by taking away resources from non-priority areas and allocat-ing them to the new priority topics It implies that only some researchthemes are included into the new profile The changes in the allocation ofresources do not lead however to the appearance of new or the disappear-ance of existing research The university keeps all research efforts butchanges their prospects of development This strategy was applied by thescience faculty in university 1 and by university 2
(Positive or Negative) Selection of TopicsIn contrast to the previous strategy of prioritisation which can be seen asshifting emphasis this strategy leads to a sharp change of profile It includesthe import of new research or the removal of existing research In the firstcase additional resources or resources taken away from non-priority areasare used to create positions and infrastructure for a new area which isusually linked to some existing priority area An example of this strategyhas been observed by Lucas (2006 pp 100104) in a British university
127Where Have All the Scientists Gone
Noticing that all biology departments that were rated highly in the researchassessment exercise had biotechnology in their profile the observed depart-ment decided to import biotechnology in its research profile The oppositestrategy negative selection means entirely removing topics for example bytaking away all vacant positions or by negotiating the redundancy of staffNegative selection is usually motivated by the necessity to free resourceseither for investing them elsewhere or for reducing budget deficits It wasapplied by a faculty in university 3 and by university 4
The occurrence of these strategies in our cases is linked to specific pat-terns in the distribution and exercise of authority summarised in Table 1which we will now further discuss Authority over research profiles that isconcerning decisions about what research areas should be at the universityand how strong these areas should be was held by the university executiveboard in all cases When the modification of whole university profiles wasdecided upon this authority had to be shared with researchers becauseexisting research competences and priorities limit the options a universityhas While researchers and research groups always have a spectrum ofpossible directions for future research this spectrum is limited and for-mulating topics outside the spectrum risks loss of performance This iswhy researchers have authority over the generation of options for a pro-file In some cases authority over options had to be shared with managersat several levels of the hierarchy However the options were generatedby researchers and higher-level managers only reserved the right toselect among them (a right which was not exercised in the cases analysedby us)
The situation was different however when research areas had to beremoved for financial reasons Again these processes have very similarauthority structures The university management shares authority withinternal and external expert committees rather than researchers Veto rightsare exercised if the higher-level management is not satisfied with the out-comes of processes
THE IMPACT OF PROFILE BUILDING ON
RESEARCH FIELDS
In the previous section we discussed characteristic strategies and financialinstruments for profile building at Dutch universities and some of the con-sequences for research in these universities We now turn to the
128 GRIT LAUDEL AND ELKE WEYER
Table 1 Authority Relations in Different Strategies of Profile Building
University Strategy Authority over Options
for Profile
Authority over
Selection of Options
Authority over
Allocation of Funds
Veto Positions Changes in Profiles
1 Thematic
re-contextualisation
Researchers research
programme leaders
Research programme
leaders
Faculty board Faculty board
(not used)
Virtually none
1 Thematic and financial
prioritisation
Faculty management
shared with university
executive board and
senior staff
Faculty management Faculty management Faculty
management (not
used)
Growth of two
research areas
2 Thematic and financial
prioritisation
Researchers research
institute leaders
University executive
board shared with
expert committees
University executive
board shared with
funding agencies and
national elites
University
executive board
(not used)
Growth of selected
research areas
3 Negative selection of
topics
Dean shared with
internal committee
Dean shared with
science policy
committee
Dean shared with
science policy
committee
Dean head of
Department (used
see next row)
see next row
3 Negative selection of
topics
Dean shared with
internal committee
Head of department
shared with dean
Head of department
shared with dean
none Removal of several
research areas
4 Negative selection of
topics
University executive
board shared with
external experts
Dean and university
executive board
Dean and university
executive board
Dean (used) Removal of one
discipline loss
of one field
129
Where
Have
AlltheScien
tistsGone
consequences of these strategies at the national level of fields of researchIn each case the strategies described in the previous section have beenapplied by all universities that host groups from a field thereby creatingaggregate effects at the field level
The Extinction of Dutch Evolutionary Developmental Biology
Evolutionary developmental biology (evo-devo) is an interdisciplinaryresearch field that combines evolutionary biology and developmental biologyIt investigates the evolution of developmental processes aiming for amechanistic understanding of phenotypic change (Sommer 2009 p 416)It can be traced back to the end of the 1970s when it became more andmore obvious that neo-Darwinian theory was unable to account for all theempirical findings of evolutionary biology was fuelled by discoveries ofgenes (HOX genes) regulating embryonic development in the 1980s andgot further impulses through the development of molecular technologytechniques (Muller 2007 Raff 2000) Researchers who want to start evo-devo research may come from a variety of fields mainly from evolutionarybiology and developmental biology By now it is an accepted field world-wide still strongly rooted in its original disciplines but with its own confer-ences and journals and emerging career tracks (for an extendedrepresentation of the evo-devo case see Laudel et al this volume) It isfundamental research that is difficult to link to medical or agriculturalapplications It can be cost intensive if comparative experimental studiesare undertaken and will often fare low on common citation-based indica-tors due to its newness and interdisciplinary nature
Evo-devo was originally deemed an underdeveloped area in theNetherlands by the scientific community which led to the creation of achair for evo-devo at one university This chair was created at the initiativeof a bioscience faculty and was co-funded by the research council and theuniversity Shortly thereafter however evo-devo research suffered fromnegative selection of related fields at several universities and prioritisationof other areas at other universities
At the university at which the chair was created a subsequent longphase of reorganisations eventually led to the closure of an evolutionarybiology institute The top management of the university had put pressureon the faculty to solve their budget problems An lsquoexpert committeersquo wasrecruited for preparing the decision It considered the evolutionary biol-ogy grouprsquos potential to acquire external funding insufficient despite the
130 GRIT LAUDEL AND ELKE WEYER
high quality of the research which was confirmed by a previous regularevaluation The decision to close evolutionary biology had an effect thatnone of the profile builders anticipated the largest and internationallyhighly recognised Dutch evo-devo group left the university shortly afterThe prolonged process of decision-making about the fate of researchgroups had led to an atmosphere that researchers considered as unfavour-able Two leading evo-devo researchers went abroad Although movesbetween universities usually have several reasons including the prospectsof the new post (as was the case with those two researchers) a generalpattern emerges in this case of simultaneous migration Several tenuredsenior researchers moved to posts elsewhere which can be traced back tothe reorganisation
What I have seen now already a couple of times happening in various places where Irsquove
worked people get a bit sort of nervous if theyrsquore getting an opportunity to [establish]
a good research group elsewhere they leave Two or three good people left in that
group
A similar fate was met by one of the first Dutch evo-devo groups whohad just started to move into the direction of evo-devo The group leader(a professor) had secured an external grant and had several PhD studentsand postdocs working in this area In the late 1990s the faculty had toreduce its number of permanent positions due to financial problems As aresult the professor was forced to retire The faculty appointed a newchair in a priority area which would enable large-scale collaborationswith others PhD students and postdocs whose professors retired or wereobliged to retire went abroad or left science As a result the original evo-devo research lines that had only just started to gain momentumstopped
In addition to universities there is one public research institute (theHubrecht Institute funded by the Dutch Academy of Science) that is apotential host for evo-devo research This institute had a long tradition indevelopmental biology it is fully autonomous and the directions of itsresearch are decided by the institutersquos director This director a member ofthe national elite of biological research was not interested in and even hos-tile to evo-devo research Owing to his specific interest the institute hasshifted its research away from developmental biology and has recruitedresearch groups in other areas of biology The second reason for this shiftis the strong dependence of the institute on external funding and fundingagenciesrsquo thematic expectations (see below) As a consequence almost noevo-devo research is conducted in this institute
131Where Have All the Scientists Gone
Apart from effects that can be traced back to universitiesrsquo and the publicresearch institutersquos profile-building activities other conditions had a stronginfluence on the development of the field most notably external fundingopportunities In recent years the most important funding agency for fun-damental biological research in the Netherlands has shifted its fundingmore and more into an applied direction and earmarked thematic pro-grams Success rates for small investigator grants have gone down from30 to 20 in the biology area This led to significant increase in the compe-tition for grants between evo-devo researchers and well established funda-mental biology groups
There is this orientation towards fundamental research in Arabidopsis There excellent
groups exist And they are very competitive in these grants for example Vidi or Vrije
Competitie But with evo-devo I felt we never had much of a chance
Several researchers had difficulties in pursuing their evo-devo researchbecause they could not get grants for it The unfavourable climate contrib-uted to career decisions to leave the Netherlands and to do evo-devoresearch abroad
Different from other countries who have large well-known evo-devolaboratories (eg the United Kingdom Germany Switzerland) research inevo-devo in the Netherlands exists today as scattered activities of indivi-duals What the researchers themselves refer to as lsquohardcorersquo experimentalevo-devo research hardly exists
So I would say I actually donrsquot know if evo-devo has any significant presence in the
Netherlands any more [hellip] I think evo-devo doesnrsquot really have a presence in the
Netherlands unless there may be some plants people that do evo-devo type research
[hellip] Some of its work may be classifiable as evo-devo But I wouldnrsquot know (Dutch
evo-devo researcher currently working abroad)
The combined actions of authorities state expectations funding agen-ciesrsquo priorities managerial decisions of research organisations all wentinto the same direction and thus drove evo-devo in the Netherlands closeto extinction
Chilling out Bose-Einstein Condensation
Bose-Einstein condensate (BEC) is a state of matter that occurs whengases of atoms or subatomic particles are cooled to near absolute zerodegree Kelvin (see Laudel et al this volume on BEC) In 1995 the firstBEC was experimentally produced by two US groups This achievement
132 GRIT LAUDEL AND ELKE WEYER
has first triggered attempts at replication After it became apparent thatBEC can be used for a wide range of fundamental research several sub-fields of physics BEC research grew rapidly Today more than 100experimental groups and a multitude of theoretical groups worldwidework on BEC
The Netherlands had a long tradition in atomic and laser physics andwere particularly well positioned to move in this new direction One Dutchresearch group belonged to the pioneers of BEC research and made impor-tant contributions on the route of experimentally realising a BEC for thefirst time Other groups had already incorporated laser cooling techniquesin their research However the successful production of BECs and interest-ing research with them came rather slowly In the mid-2000s four groupshad succeeded in producing a BEC and established experimental BEC astheir main line of research
When universities began to build research profiles atomic and molecularoptics (AMO) physicists in general and BEC researchers in particularfound themselves in a situation where they did not fit into a priority areaAlthough there was no immediate threat for the existing groups notbelonging to a priority area also meant that AMO physics and BECresearch is not a growth area New professors are more likely to beappointed in the priority areas of the universities
And now the focus is shifting to use these lasers for life science and biomedical optics
So we have some new professors in biomedical optics Thatrsquos where most of the money
goes
I And who made this shift
That is on the highest level of the university At the highest level of university is decided
that the natural sciences should focus their research more on the medical center (thatrsquos
the big thing over there) Actually that does not hold so much for our research So we
are a bit in a different situation But if we keep up the quality we survive
I But why actually It sounded as if the university didnrsquot really hinder your research
but didnrsquot also promote it too much
No Now in recent years universities are struggling And they need to have their profile
And [our university] focuses on medical research
I Yes But you have your lasers and you get your money from [the funding agency]
FOM How can they make you uneasy
They donrsquot throw us out yet But in chemistry now they made plans hellip chemistry had
to make budget cuts And they keep the people who are in the life sciences So if there
is at any point these political decisions are coming that there must be budget cuts made
the [university] would choose to keep the life sciences
133Where Have All the Scientists Gone
But the problem is that BEC research is not the focus of this department [hellip] We are in
another reorganisation at the moment And it is much easier if you can say okay but
Irsquom in the focus
Beside research profile-building activities of universities profile-buildingdecisions of the most important non-university institute in this area hadalso influenced BEC research Strongly backed up by the national physicscommunity the agency funding the institute decided to move researchtowards biophysics Biophysics was considered a neglected field and neces-sary to catch up with international developments Consequently it wasdecided not to invest into BEC research despite the fact that the instituteprovided an ideal infrastructure and hosted the most successful BECresearcher
As a result of these trends the dynamics of BEC research in theNetherlands markedly differs from the international trend While the fieldstill grows worldwide new professorships are devoted to BEC in Germanyand countries like Switzerland initiate research with BEC (see Laudel et althis volume) Dutch BEC research is stagnating rather than growing andits situation may become precarious
CONCLUSIONS CAUSES AND CONSEQUENCES OF
PROFILE BUILDING
Profile building at Dutch universities emerged at the intersection of threemajor shifts in the governance of research over the last decades Scienceand higher education policy considers that the Dutch state is unable tofund research across all fields and topics and has introduced policies tofocus research funding on some areas while by implication abandoningothers This policy includes the demand that universities should not all dothe same research which is implemented rather forcefully The requirementfor focus and (critical) mass is accompanied by directions what to focus onand where to create mass The policy shift to more and stronger demandsfor lsquousefulrsquo research has made the state define research areas that promisecontributions to the solution of societal problems These areas have beenset up as targets for university profiles and have been made attractive bythe redistribution of public funding
Thus the history of profile building of Dutch universities is a history ofuniversities and academics losing authority over research topics Universitieslose authority to the state and researchers lose authority to both the state
134 GRIT LAUDEL AND ELKE WEYER
and their universities While the latter process is in line with the NPM ideol-ogy which promotes increasing action capabilities of universities bystrengthening hierarchical steering the former is not Quite contrary to thepolitical announcements of increasing the autonomy of universities Dutchuniversities have lost autonomy over their research profiles Higher educa-tion reforms appear to have merely changed the channel though whichauthority over universities is exercised which makes the newly grantedautonomy contingent on state interests (see also Enders De Boer amp Weyer2013 on lsquoregulatory autonomyrsquo)
The shift in authority relations concerning the formulation of researchtopics that has been produced by Dutch NPM reforms can be described asfollows Academics at Dutch universities retain the authority over their for-mulation of research goals although they lost some of their authority con-cerning the profile building Their authority is epistemic authority that isauthority that is based on their exclusive access to the knowledge that isnecessary to design research topics and goals (Whitley amp Glaser thisvolume) This is why the university leadership which by now has the ulti-mate formal authority to shape research profiles devolves any processes inwhich a profile needs to be constructed from existing research Only if bud-getary considerations are dominant and consequences for research seemless important university management uses formal authority to overridethe epistemic authority of researchers usually by using outside experts as asource of information on the research and as a source that is supposed tolegitimise their decisions
The consequences of these two profile-building processes are manifoldOnly some of them are anticipated and even fewer are taken into accountin the two processes of profile building Within universities at least someresearchers in fields that are given a lower priority feel unwelcome andthreatened Even if a prioritisation strategy lets them stay with the same orslightly reduced funding they know that their position will become precar-ious whenever budget problems occur Prioritisation strategies andincreased selectivity of resource allocation in the university create a ten-dency to crowd out researchers in non-priority areas These researchersbegin to leave the university whenever the opportunity arises
These shifts and the removal of topics usually sever collaborative tieswithin the university They thus affect the removed or devalued topicrsquos(previous) environment in ways that are difficult to understand and to pre-dict This was the case with the astronomy institute at Utrecht Universitywhose disappearance discouraged theoretical physicists It may also affectthe general climate for research as we have seen for the field of evo-devo
135Where Have All the Scientists Gone
where the closure of one institute from a closely related field led evo-devoresearchers to leave the university The case of BEC is quite similar becausethe development in the Netherlands ground to a halt at a time where BECsbecome a research method that is used in an increasing number of subfieldsof physics
The most severe consequences of profile building occur when a field dis-appears entirely This kind of quasi-market failure has been discussed inrelation to research evaluation systems (Glaser 2007) The creation ofresearch profiles by Dutch universities occurs in a similar situationThrough creating their profiles the universities compete for money boththe additional block funding that is conditional on having the lsquorightrsquo profileand research council funding that is increasingly shaped by research priori-ties set by the Dutch state The signals the universities have to follow arethe same for all universities and include priority areas of research set by thestate the state expectation of utility and the necessity of maximising grantincome which is increasingly shaped by the same state expectations Wehave shown that the management of Dutch universities has the formalauthority to significantly strengthen or to remove research areas from theirprofile and that university and faculty leaders do exercise this authority Ifall universities follow the same signals when shaping their profiles they arelikely to positively and negatively select the same research topics whichleads to the growth of some fields and the disappearance of others at thenational level This has almost happened with Dutch evo-devo researchand may well happen with BEC in the future
Thus the current authority relations in Dutch higher education create apropensity for quasi-market failure because they combine thematic priori-ties prescribed by the state a powerful demand for profile building of uni-versities financial problems of many universities and authority structuresat universities that enable far-reaching changes of profiles The warningabout a possible quasi-market failure in Germany namely the disappear-ance of the so-called lsquosmall disciplinesrsquo (HRK 2007) indicates another pos-sible reason for such failure namely the ubiquitous belief that the creationof lsquocritical massrsquo of research activities in an area creates better research(Schiene amp Schimank 2007) Such a quasi-market failure would have takena very long time to occur due to the much lower authority of universitymanagement in Germany which can only intervene when professorshipsbecome vacant However the example indicates a possible systematic side-effect of NPM Transforming the governance of universities to sendingstrong signals via quasi-markets creates opportunities for quasi-marketfailures
136 GRIT LAUDEL AND ELKE WEYER
The argument that the small size and financial situation of the Dutchhigher education system require the concentration of resources does soundconvincing and the current processes will very likely lead to a strengthen-ing of research areas science policy wants strengthened The loss of univer-sity autonomy in the process and the shrinking or disappearance ofresearch fields are however inevitable side-effects of these processes whoselong-term impact on the Dutch science system is difficult to predict
NOTE
1 The Executive Board is the highest internal governing and administrative bodyin higher education institutions It consists of three persons including the RectorMagnificus All members are appointed for a 4-year term by the UniversityrsquosSupervisory Board after receiving the views of the University Council It assumesthe ultimate responsibility for the policies of the university including the appoint-ment of deans directors and professors The Board is legally entitled to initiate newprogrammes and related research institutes and is responsible for managing the uni-versityrsquos finances and other management tasks The division of labour between themembers of the Executive Board differs from university to university
ACKNOWLEDGEMENTS
We thank Jochen Glaser for numerous discussions and ideas about mergingour two empirical studies into one article We would also like to thank ananonymous reviewer for his or her critical reading of the manuscript Thiswork was supported by two project grants from the NetherlandsOrganisation for Scientific Research (NWO) and a travel grant from theEuropean Science Foundation (ESF)
REFERENCES
Blume S amp Spaapen J (1988) External assessment and ldquoconditional financingrdquo of research
in Dutch universities Minerva 26 130
Clark B R (1998) Creating entrepreneurial universities Organizational pathways of transfor-
mation New York NY Pergamon
De Boer H Enders J amp Leisyte L (2007) Public sector reform in Dutch higher education
The organizational transformation of the university Public Administration 85 2746
Enders J De Boer H amp Weyer E (2013) Regulatory autonomy and performance The
reform of higher education re-visited Higher Education 65 523
137Where Have All the Scientists Gone
Glaser J (2007) The social orders of research evaluation systems In R Whitley amp J Glaser
(Eds) The changing governance of the sciences The advent of research evaluation sys-
tems (pp 245266) Dordrecht the Netherlands Springer
Glaser J (2012) Are universities professional organisations Paper presented at the sub-theme
17 Organizing science The increasingly formal structuring of academic research 28th
EGOS Colloquium 2012 Helsinki Finland
Glaser J Lange S Laudel G amp Schimank U (2010) The limits of universality How field-
specific epistemic conditions affect authority relations and their consequences In
R Whitley J Glaser amp L Engwall (Eds) Reconfiguring knowledge production
Changing authority relationships in the sciences and their consequences for intellectual
innovation (pp 291324) Oxford Oxford University Press
Glaser J amp Laudel G (2007) Evaluation without evaluators The impact of funding formu-
lae on Australian university research In R Whitley amp J Glaser (Eds) The changing
governance of the sciences The advent of research evaluation systems (pp 127151)
Dordrecht the Netherlands Springer
Glaser J amp Laudel G (2009) Identifying individual research trails In B Larsen amp J Leta
(Eds) Proceedings of the 12th international society for scientometrics and informetrics
(ISSIrsquo09) (pp 841845) Rio de Janeiro Brazil
Glaser J amp Laudel G (2013) Life with and without coding Two methods for early-stage
data analysis in qualitative research aiming at causal explanations [96 paragraphs]
Forum Qualitative Sozialforschung Forum Qualitative Social Research 14 Retrieved
from httpnbn-resolvingdeurnnbnde0114-fqs130254 Accessed on February 12
2014
HRK (Hochschulenrektorenkonferenz) (2007) Die Zukunft der kleinen Facher Potenziale -
Herausforderungen - Perspektiven Bonn Germany Hochschulenrektorenkonferenz
Jansen D (Ed) (2010) Governance and performance in the German public research sector
Disciplinary differences Dordrecht the Netherlands Springer
Keller C U (2012 April 25) Sterrekundig Instituut Utrecht The last years 370 years of
astronomy in Utrecht Proceedings of a conference held at Hotel Leeuwenhorst
Noordwijkerhout The Netherlands San Francisco CA Astronomical Society of the
Pacific Retrieved from httparxivorgabs12084095 Accessed on February 12 2014
Klumpp M De Boer H amp Vossensteyn H (2013) Comparing national policies on institu-
tional profiling in Germany and the Netherlands Comparative education published
online Retrived from httpwwwtandfonlinecomdoipdf101080030500682013
834558 Accessed on April 5 2014
KNAW (De Koninklijke Nederlandse Akademie van Wetenschappen) (2013) Effecten van
universitaire profilering en topsectorenbeleid op de wetenschap in Nederland Een eerste
kritische Reflectie Amsterdam the Netherlands KNAW Retrieved from httpsknaw
nlnlactueelpublicatieseffecten-van-universitaire-profilering-en-topsectorenbeleid-op-de-
wetenschap-in-nederland Accessed on February 16 2014
Lange S (2007) The basic state of research in Germany Conditions of knowledge production
pre-evaluation In R Whitley amp J Glaser (Eds) The changing governance of the
sciences (pp 153170) Dordrecht the Netherlands Springer
Laudel G (2012) Doing something new in the Netherlands The impact of research organisa-
tions and funding agencies on the start of new research lines Paper presented at the sub-
theme 17 Organizing science The increasingly formal structuring of academic research
28th EGOS Colloquium 2012 Helsinki Finland
138 GRIT LAUDEL AND ELKE WEYER
Lee F (2009) A history of heterodox economics London Routledge
Leibfried S (Ed) (2010) Die Exellenzintiative - Zwischenbilanz und Perspektiven Frankfurt
Germany Campus Verlag
Leisyte L (2007) University governance and academic research Case studies of research units in
Dutch and English universities Enschede the Netherlands CHEPS University of Twente
Leisyte L Enders J amp De Boer H (2010) Mediating problem choice Academic research-
ersrsquo responses to changes in their institutional environment In R Whitley J Glaser amp
L Engwall (Eds) Reconfiguring knowledge production Changing authority relationships
in the sciences and their consequences for intellectual innovation (pp 266290) Oxford
Oxford University Press
Lucas L (2006) The research game in academic life Maidenhead UK SRHEOpen
University Press
Marginson S amp Considine M (2000) The enterprise university Power governance and rein-
vention in Australia Cambridge UK Cambridge University Press
Meier F amp Schimank U (2010) Mission now possible Profile building and leadership in
German universities In R Whitley J Glaser amp L Engwall (Eds) Reconfiguring
knowledge production Changing authority relationships in the sciences and their conse-
quences for intellectual innovation (pp 211236) Oxford Oxford University Press
Musselin C (2007) Are universities specific organizations In G Krucken A Kosmutzky amp
M Torka (Eds) Towards a multiversity Universities between global trends and national
traditions (pp 6384) Bielefeld Germany Transcript Verlag
Muller G B (2007) Six memos for Evo-Devo In M D Laubichler amp J Maienschein (Eds)
Embryology to evo-devo A history of developmental evolution (pp 499524)
Cambridge MA MIT Press
OrsquoMahony S amp Lakhani K R (2011) Organizations in the shadow of communities In C
Marquis M Lounsbury amp R Greenwood (Eds) Communities and organizations
Research in the Sociology of Organizations (Vol 33 pp 336) Bingley UK Emerald
Group
Pietila M (2014) The many faces of research profiling Academic leadersrsquo conceptions of
research steering Higher Education 67 303316
Raff R A (2000) Evo-devo the evolution of a new discipline Nature Reviews Genetics 1
7479
Rafols I Leydesdorff L OrsquoHare A Nightingale P amp Stirling A (2012) How journal
rankings can suppress interdisciplinary research A comparison between innovation stu-
dies and business amp management Research Policy 41 1262-1282
Schiene C amp Schimank U (2007) Research evaluation as organisational development The
work of the Academic Advisory Council in lower Saxony (FRG) In R Whitley amp
J Glaser (Eds) The changing governance of the sciences The advent of research evalua-
tion systems (pp 171190) Dordrecht the Netherlands Springer
Seidel M-D L amp Stewart K J (2011) An initial description of the C-form In C Marquis
M Lounsbury amp R Greenwood (Eds) Communities and organizations Research in
the Sociology of Organizations (Vol 33 pp 3772) Bingley UK Emerald Group
Sommer R J (2009) The future of evo-devo Model systems and evolutionary theory Nature
Reviews Genetics 10 416422
Technopolis (2011) Mini country report The Netherlands Retrieved from httpeceuropa
euenterprisepoliciesinnovationfilescountryreportsnetherlands_enpdf Accessed on
February 12 2014
139Where Have All the Scientists Gone
Van den Besselaar P amp Horlings E (2010) Focus en massa in het wetenschappelijk onder-
zoek de Nederlandse onderzoeksportfolio in internationaal perspectief Den Haag the
Netherlands Rathenau Instituut
Van der Meulen B (2007) Interfering governance and emerging centres of control University
research evaluation in the Netherlands In R Whitley amp J Glaser (Eds) The changing
governance of the sciences The advent of research evaluation systems (pp 191202)
Dordrecht the Netherlands Springer
Von Stuckrad T amp Glaser J (2012) Es ist hochstens eine Kollegenschelte moglich aber die
bringt nichts - Kontingente und strukturelle Handlungsbeschrankungen der intrauniver-
sitaren Forschungsgovernance In U Wilkesmann amp C J Schmid (Eds) Hochschule
als organisation (pp 223243) Wiesbaden Germany VS-Verlag
Wenger E C amp Snyder W M (2000) Communities of practice The organizational frontier
Harvard Business Review 78 139145
Westerheijden D F De Boer H amp Enders J (2009) Netherlands An lsquoEchternachrsquo proces-
sion in different directions Oscillating steps towards reform In C Paradeise E Reale
I Bleiklie amp E Ferlie (Eds) University governance Western Europe comparative per-
spectives (pp 103125) Dordrecht the Netherlands Springer
Weyer E (2014) From loose to tight management Seeking evidence for archetype change in
Dutch and English higher education PhD thesis University of Twente CHEPS
Enschede the Netherlands
Whitley R (2008) Universities as strategic actors Limitations and variations In L Engwall amp
D Weaire (Eds) The university in the market (pp 2337) London Portland Press
Whitley R (2010) Reconfiguring the public sciences The impact of governance changes on
authority and innovation in public science systems In R Whitley J Glaser amp
L Engwall (Eds) Reconfiguring knowledge production Changing authority relationships
in the sciences and their consequences for intellectual innovation (pp 349) Oxford
Oxford University Press
140 GRIT LAUDEL AND ELKE WEYER
MERGER MANIA IN SCIENCE
ORGANIZATIONAL
RESTRUCTURING AND PATTERNS
OF COOPERATION IN AN
ACADEMIC RESEARCH CENTRE
Julien Barrier
ABSTRACT
While the formal structures of universities may predominantly reflect cere-monial rather than functional purposes attempts at changing them areusually a fertile ground for academic conflicts Taking this apparent contra-diction as a starting point the aim of this article is to explore the intriguingrole of formal structures in academic settings Drawing on a case study ofa merger and organizational restructuring process in an academic researchcentre it shows how symbolic responses to institutional pressures may haveactual consequences on research practices beyond myth and ceremony
Keywords Decoupling and recoupling processes inhabited institu-tions perspective mergers in higher education organization ofacademic research scientific cooperation
Organizational Transformation and Scientific Change The Impact of Institutional
Restructuring on Universities and Intellectual Innovation
Research in the Sociology of Organizations Volume 42 141172
Copyright r 2014 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN 0733-558Xdoi101108S0733-558X20140000042005
141
INTRODUCTION
There is an enduring intriguing puzzle in academic institutions On the onehand the design of their formal structures often seems to serve ceremonialrather than functional purposes predominantly reflecting conformity toinstitutionalized myths that is widespread taken-for-granted beliefs aboutrational models of organization (Meyer amp Rowan 1977) Universities areoften cited as a prime example of decoupling between structures and coreactivities For instance the rise of technology transfer offices in Germanuniversities has been described as a symbolic response to institutional pres-sures having little if any impact on actual university-industry relation-ships (Krucken 2003) On the other hand as pointed out by Musselin(2007) if university formal structures were all about myth and ceremonywhy do attempts at changing them usually prove so difficult and why arethey such a fertile ground for academic politics
Taking this apparent contradiction as a starting point the aim of thisarticle is to consider how and in what circumstances the formal organiza-tion of research units could affect patterns of collaboration and coordina-tion drawing on a longitudinal case study of the merger and restructuringof a French university research centre in electronic engineering Followingrecent developments in organizational institutionalism promoting a moreactor-centred approach (Hallett amp Ventresca 2006) this empirical setting istaken as a case for exploring the micro-level processes involved in thedesign adoption and transformation of structures in academic organiza-tions while not neglecting their relation to broader institutional pressures
Indeed concentration and mergers of research units are arguably notonly a significant phenomenon in the academic field in Europe but are alsoa key research site for analysing the interplay between institutional pres-sures and local work arrangements Like the lsquomerger maniarsquo in the hospitalsector in the United States (Kitchener 2002) the surge of policies advocat-ing the merger and concentration of research units exemplifies the rise ofnew powerful rational myths in the academic field For more than a decadenow policy makers in various European countries including France(Vinck amp Zarama 2007) have called for a concentration of scientificresources and personnel in lsquocritical sizersquo institutes or research centres inorder to boost their performance and outputs1 a trend Bonaccorsi andDaraio (2005) refer to as a lsquopolicy mythrsquo given the ambiguous nature ofthe relation between organizational size and scientific output
The rest of the article is structured as follows After outlining the analy-tical approach and the empirical materials of this article background data
142 JULIEN BARRIER
on the French higher education and research system are provided Thenresults from the case study are presented and discussed Although formalstructures are decoupled from core work patterns (eg cooperation pat-terns) to a certain extent our results show how reshaping structures maynonetheless have real consequences by triggering an unexpected chain ofreactions
BEYOND MYTH AND CEREMONY HOW
FORMAL STRUCTURES MAY MATTER IN
ACADEMIC SETTINGS
In order to present the theoretical background of the article this sectionoutlines three analytical issues that informed the formulation of ourresearch question and the analysis of empirical data
First this article considers the organizational formalization of academicsettings as a key yet under-researched issue Since the late 1970s most scho-lars in the field of Science and Technology and Society (STS) have beenmore concerned with bench practices or micro-processes of network forma-tion than with organizations and institutions (Frickel amp Moore 2006) atendency most evident in early laboratory studies (eg Latour amp Woolgar1979) This is not to say that science studies have ignored altogether organi-zational issues showing for instance how scientists mobilize organizationalresources to support their research agendas or the emergence of a disciplineBut STS scholars have showed little interest for the massive processes oforganizational rationalization that are so crucial to institutionalists whostress the dramatic expansion of formal structures in the field of higher edu-cation and research over the last decades (Drori Meyer Ramirez ampSchofer 2006 Krucken amp Meier 2006) As a result while the core activitiesof scientists have been increasingly dependent on their involvement intomultiple intersecting organizational forms (Barrier 2011) the consequencesof this trend at the micro-level are rarely addressed
Second although the increasing formal organization of academic activ-ities may often result in window dressing and symbolic compliance oneshould be cautious not to overestimate the extent of decoupling betweenstructures and practices The notion of decoupling originally put forward byMeyer and Rowan (1977) pointed to the gaps between the adoptionof structures reflecting conformity to institutional myths and everyday prac-tices in organizations in order to question the then-dominant assumption
143Merger Mania in Science
that formal structures were always effective means of coordination and con-trol2 Yet the original meaning of the notion has somehow suffered from itsvery success as decoupling is often assumed to be the default response oforganizations in general and academic institutions in particular to insti-tutional pressures Indeed Bromley and Powell (2012) state that much ofthe recent literature referring to decoupling has unfortunately equated adap-tation to institutional pressures with mere window dressing Resisting thisinterpretation they further note that lsquoeven ceremonial adoption can uninten-tionally trigger a chain of reactions that have real organizational effectsrsquo(Bromley amp Powell 2012 p 3) Thus instead of assuming that decouplingoccurs invariably it is arguably more fruitful to untangle the complex pro-cesses that mediate edit and shape the translation of institutionalized mythsinto practices a theme central to the lsquoinhabited institutionsrsquo perspective inorganizational institutionalism (Hallett amp Ventresca 2006)
Studies in this vein make at least two relevant contributions to ourresearch questions First they favour longitudinal approaches and focus oninteractions and the construction of meaning at the micro-level in order toidentify the complex indirect processes translating institutional pressuresinto core activities how they are lsquorecoupledrsquo (Hallett 2010 Sauder ampEspeland 2009) Second they draw attention to the internal diversity ofresponses to institutional pressures Binder (2007) shows that departmentswithin the same organization are structured along different institutionallogics which in turn shape the way they creatively adapt to externalpressures
Third a key issue is to capture how the design of formal structures mayshape and be shaped by intra-organizational dynamics Although aca-demic research units may exhibit significant variation in their organizationdepending on institutional context size and discipline they can arguablybe described as loosely coupled systems (Weick 1976) This is especiallyrelevant in the French context where they usually come as a confederationof research groups of different sizes more or less connected to each other(Louvel 2010) Having said that the question remains to understand whatnonetheless binds researchers together A first answer is to look at patternsof interdependence among researchers In the experimental sciencesresearch groups are usually based on a division of labour with functionalinterdependence stemming from the specialization of their members onvarious epistemic tasks (Shinn 1988) Yet this form of interdependenceis mostly significant at the research group level which is usually thebasic organized unit of knowledge production it is far less significant atthe institute or research centre level In addition beyond epistemic
144 JULIEN BARRIER
interdependence one may identify resource interdependence resulting fromthe need to manage of critical resources such as posts funding or researchfacilities meaning that research groups and centres alike tend to be animportant locus for the formation of local rules governing the allocation andmanagement of resources However looking at functional or task interde-pendence might not be sufficient
In a paper on the role of formal structures in universities departmentsbasic research units institutes schools and so on Musselin (1990) arguesthat they strikingly differ from similar structures in hierarchical mechanis-tic bureaucratic organizations In universities she argues the contours ofstructures are not as much based on patterns of interdependence asthey reflect the formation of defensive territories Basically they enablemembers to draw boundaries to assert their autonomy vis-a-vis othergroups (eg members of the chemistry department would not have a sayabout what is going on in the physics department) which favours theemergence of in-group identity and facilitates the development of localrules of cooperation The existence of a structure does not mean thatcooperation will ensue but it supports the maintenance of relationshipsthat would be fragile otherwise Consequently to understand how institu-tional pressures transform formal structures and affect work floorpractices one must not only examine how they relate to functional interde-pendence but also how they may destabilize or reconstruct existing defen-sive territories
DATA AND METHODS
This article is based on a longitudinal case study of a large research centrethat resulted from the merger of two research units This organization willbe referred to as the Electronic Engineering Institute (EEI a pseudonymaiming to protect the anonymity of interviewees) The study was part of alarger project on the changing conditions for academic research in Franceover the 19802007 period (Barrier 2011) It focused on the field ofacademic research in electronic engineering and micronanotechnologywhich was selected as an appropriate case for analysing the reconfigurationof relations between government academic science and industry The pri-mary aim of this project was to understand how changes in fundingregimes especially the rise of project funding affected the organizationof research and the professional autonomy of academic researchers
145Merger Mania in Science
Empirically the design of the project was mainly based on the longitudi-nal study of two large research centres including the EEI These two orga-nizations were conceived as focal research sites that allowed for thecollection of rich data on several dimensions including funding patternsbut also the internal management and organization of the research centresas well as their relations with their institutional environment Data werealso collected to position these cases in a broader national context
In addition to the realization of a total of 128 semi-directive interviews(37 of which were conducted for the EEI case study) with academics mem-bers of university governance boards research administrators policymakers and industrial researchers this project combined archival researchquantitative data on funding as well as bibliometric data This designallowed for a strong contextualization of local observations and the trian-gulation of data It also allowed for the exploration of emerging themesFor instance when fieldwork was carried on at EEI it was decided to col-lect additional data on the merger process since preliminary interviews pre-sented the merger as a central issue Subsequently interviews with researchadministrators at the national level included questions about mergersFinally in order to get a more precise view of intra-organizational coopera-tion within EEI publication data were collected from the Web of scienceWhile cooperation does not necessarily result in joint publication (Laudel2002) co-authorship patterns can yield further insights to complement andinterpret interview data with researchers MS Access was then used to sortout raw data and identify co-authorship networks Then the softwarePajek (De Nooy Mrvar amp Batagelj 2005) was used to create examine anddraw up these networks
BACKGROUND
Positioning the EEI in the French Public Research System
A lsquojoint research unitrsquo between a public research organization the CentreNational pour la Recherche Scientifique (National Centre for ScientificResearch CNRS) and three higher education institutions the EEI reflectsa distinctive feature of the French science system (Louvel 2010 ThevesLepori amp Laredo 2007) In this context the organization of academicscience as joint research units combining public research organizations withuniversities is the rule rather than the exception
146 JULIEN BARRIER
The EEI receives funding from both the CNRS and higher educationinstitutions and includes both CNRS and higher education personnel Witha minority of CNRS researchers the staffing of EEI reflects a general trendin fields such as computer science or electronic engineering where the cen-tre of gravity of research units leans towards university professors ratherthan CNRS researchers This relates to the demographics of these disci-plines in France in which the growth of research activities has been primar-ily driven by a rising number of students in the 1990s leading to a rapidincrease in the number of university positions not matched by posts inpublic research organizations (Laredo amp Mustar 2002) (Table 1)
Without going into too much detail it is important to note that theaffiliation of a research unit to the CNRS depends on an evaluation per-formed by the Comite National3 This permanent committee is in charge ofthe evaluation of CNRS research units including their creation and disso-lution and decisions about the affiliation of new units to CNRS While
Table 1 The Electronic Engineering Institute in 2006
Main Research Directions Microelectronics Electromagnetism Telecommunication
Devices and Systems Signal and Image Processing
Staff composition (rounded
numbers)
CNRS researchers 5
Professors 60
Technical and administrative staff 25
PhD students 100
Localization and
Institutional affiliations
MidCity a large city with a student population of about
40000
UniMedium medium size multidisciplinary university
TechSchool public engineering school
EngineeringSchool private engineering school (since 2004)
CNRS
Operating budget (without
staff salaries)
Core funding 16
Project funding from national European or international
agencies 45
Funding by local and regional government 17
Funding from industry 22
147Merger Mania in Science
criteria of scientific excellence and performance are central in the evalua-tions performed by the Comite National it also pays significant attention tothe organizational capacities displayed by research centres (FixariMoisdon amp Pallez 2009) A central script in the definition of research unitsas organizations is that they are expected to represent lsquomore than the meresum of their partsrsquo by promoting the sharing of facilities or resources aswell as encouraging scientific synergies among their members (eg thedevelopment of research projects or themes involving different researchgroups) Observations of debates at the Comite National show that mem-bers often encourage restructuring in research units to foster more cohesionor to reduce internal conflicts (Fixari et al 2009) In sum CNRS evalua-tion processes rely on an institutionalized organizational template basedon the notion that research units are (or should be) cohesive integratedorganizations4
The affiliation of research units with the CNRS gives access to materialand symbolic resources In addition to providing extra funding in the formof CNRS block grants it allows units to host CNRS researchers whounlike their university colleagues are full-time researchers without manda-tory teaching duties5 In more symbolic terms an affiliation to CNRS istraditionally considered as a strong distinctive quality signal in the Frenchsystem prompting university research units to compete for the lsquoCNRSlabelrsquo
lsquoMerger Maniarsquo in the Academic Field The Growing Concentration ofResearch Units in Electronic Engineering
Since the late 1990s it has become common place among French academicleaders and research administrators to criticize the lsquoorganizational disper-sionrsquo of research structures and the overall lack of lsquolegibilityrsquo of the Frenchresearch system Thus the concentration of research units into larger orga-nizations has been strongly encouraged The main arguments for this policywere not only to develop economies of scale in the use of research facilitiesand administrative resources but also to develop lsquocritical sizersquo research col-lectivities and boost the national and international lsquovisibilityrsquo of researchcentres This evolution concerns all disciplines and the field of electronicengineering is no exception Data on CNRS units in electronic engineeringconfirms a growing concentration of researchers into larger and larger insti-tutes a trend that accelerated during the 20042012 period (Table 2)6
148 JULIEN BARRIER
The analysis of original data and secondary sources suggests that thislsquomerger maniarsquo has not only been fuelled by policy ideas circulating at thenational level but also by the interactions between three set of actorsCNRS universities and local government First although the position ofCNRS regarding the concentration of research centres has varied overtime the global trend has leaned towards the creation of large researchinstitutes While interviews with the CNRS leadership and members of theComite National indicated some reluctance with the merger movementadvocated by policy makers at the ministry level they were nonethelessseen as a lever to promote rationalization in the use of experimentalfacilities and to promote restructuring initiatives in order to enable theemergence of new research themes or groups
Second in the early 1990s university leaders have fostered the creationof new organizational layers with the merger of research units or thecreation of clusters Given the limited institutional autonomy of Frenchuniversities the creation of new structures indirectly allowed the leadershipto define priority areas in the allocation of resources (Barrier amp Mignot-Gerard 2013) Predisposed to jump onto the lsquomerger bandwagonrsquo univer-sity leaders thus contributed to its advent
Third during the same period local governments especially regionalauthorities have become more and more involved in the funding of highereducation and research taking up a major role in financing the construc-tion of buildings and research infrastructures (Aust amp Crespy 2009) Inaddition they have sought to foster coordination among institutionslocated in the same metropolitan area Such initiatives have come to beviewed in increasingly positive terms as the notion of lsquoinnovation clusterrsquohas become more and more popular among policy makers As a conse-quence local governments have favoured mergers between research units
Table 2 Distribution of Research Units Affiliated to CNRS by Size(Number of Staff Members) in the Field of Electronic and Electrical
Engineering 19842012
lt25 2550 51100 101150 151300 gt300
1984 378 378 178 22 44 0
1994 206 294 206 147 88 59
2004 70 233 326 209 70 93
2012 67 100 200 100 300 233
149Merger Mania in Science
UNFOLDING THE CREATION OF THE EEI
The aim of this section is to describe how two research units TelecomLaband MicroLab merged to form the EEI in 2002 Because TelecomLab wasthe main driver of the merger process and since its former researchers com-pose the bulk of the institute created in 2002 the focus on the narrativewill be mainly centred on TelecomLab
How the Idea of a Merger Came About Aligning Local Arrangements withRising Institutional Pressures
The Early Days of TelecomLab An Organizational Umbrella LooselyAffected by Institutional Pressures (19831994)From its creation in the 1980s up to the 1990s the organization ofTelecomLab can be best described as loosely coupled first it was weaklyintegrated internally second institutional pressures to conform withCNRS recommendations were largely disconnected from work practices
TelecomLab was created in 1983 as a joint research unit between CNRSUniMedium and TechSchool in a context where UniMedium encouragedfaculty members to create formal research units and seek an affiliation toCNRS Within this context the creation of TelecomLab resulted from arather opportunistic strategy to fit into university priorities and acquire theCNRS label It was little more than an organizational umbrella for researchgroups that were independent from each other Although all groups wereinvolved in telecommunications-related research each group was essentiallyconceived as an independent entity For most of the 1980s and the 1990srelations between research groups were often tense whether verging oninterpersonal competition or mutual ignorance
Each research group had been built around a single man All of them were strong char-
acters And the relationships between those guys were not easy (hellip) We didnrsquot know
well people from other groups and we often discovered what the others were doing
only when we met them at conferences Sometimes we were also in competition when
applying for funding That was really uncomfortable (Professor)
The examination of TelecomLab board minutes confirms that thissituation of minimal cooperation that is limited to the maintenance ofan administrative umbrella persisted well into the 1990s For example aprofessor showed irritation with the suggestion that the board should meetmore often he argued that the number of meetings should be proportional
150 JULIEN BARRIER
to the money that was allocated by the lab thereby it was not necessaryto meet frequently7 Likewise while the 1996 activity report (coveringthe 19921995 period) rhetorically stresses the scientific cohesion acrossresearch groups the examination of publication outputs reveals thatthere was only one paper authored by members of different researchgroups and it was not a scientific article but a paper published in a pro-fessional outlet to present research themes at TelecomLab
Regarding external pressures during most of the 1980s up to the mid-1990s affiliation with the CNRS was viewed as a matter of prestige ratherthan a strategic concern CNRS was perceived as a distant parent organiza-tion which contributed little to the funding of research activities especiallywhen compared to contracts with industry and specialized telecom anddefense-related government labs The level of third party funding was higheven according to engineering sciences standards In addition TelecomLabdid not have any CNRS researchers and PhD candidates from TelecomLabrepeatedly failed to get recruited at CNRS thus the growth and renewal ofstaff depended on recruitments at UniMedium and TechSchool which inturn depended mainly on teaching needs In other words direct resource-dependence on CNRS was weak
That is not to say that affiliation to CNRS was deemed unimportantsince it provided symbolic status and some material resources Whilethe scientific performance of TelecomLab and the overall relevance of itsresearch activities were praised in evaluation reports window dressing wasdeveloped to exhibit compliance with the managerial and proceduralcriteria used in Comite National evaluations For instance activity reportswere carefully written to stress complementarities and cohesion acrossresearch groups although they were thin or non-existent Likewise aresearcher explained that he used not to disclose the full amount of hisindustrial contracts fearing that the lab would be considered too focusedon applied research according to CNRS standards
Aligning TelecomLab with Environmental Pressures The Crystallization of aMerger Project (19941998)At the beginning of the 1990s TelecomLab was involved in a project invol-ving the creation of a lsquoMicrowave and Radarrsquo Cluster at MidCity consist-ing in the relocation of TelecomLab and other research units in a newbuilding This project was essentially driven by opportunistic motives put-ting an appealing organizational umbrella on a set of heterogeneous activ-ities to secure resources But it also had side effects as it promptedTelecomLab members to develop new interpretations about their
151Merger Mania in Science
environment with the idea that further coordination might be developed atthe local level including other research units on campus
By 19941995 these considerations started to echo concerns about theposition of TelecomLab in a changing environment Paul a senior profes-sor and the head of a research group based at TechSchool became thechampion of a new project for TelecomLab First he argued that lsquosome-thingrsquo had to be done for TelecomLab to become lsquostrategicrsquo for CNRSwhich had started to encourage the creation of large research units Therehad been concerns over the level of CNRS support at TelecomLab sincethe early 1990s In comparison to other research units also affiliated toCNRS TelecomLab members estimated that they did not fare well Risingstudent enrolment had fuelled the creation of new faculty positions atUniMedium and TechSchool leading to the growth of TelecomLabHowever TelecomLab was still not getting any researcher positions fromCNRS This issue was considered all the more problematic as CNRS waspromoting the allocation of research positions to many research centresoutside of the Parisian area which could have benefited TelecomLabSecond Paulrsquos discourse echoed changes in the funding environment asfaculty members thought that the lsquovisibilityrsquo of TelecomLab had to beimproved A large share of project funding at TelecomLab came from thedefence and telecom sectors In 19951996 two sets of reforms in thesesectors were perceived as a threat to TelecomLab activities DGA the mainpublic agency supporting defence-related RampD made sharp spending cutsRampD in the telecom sector was also in trouble with the privatization ofFrance Telecom and the subsequent reform of CNET the government labfor telecom research These changes fuelled the idea that competition forfunding would soon become harsher and that it was necessary to defendthe collective reputation of TelecomLab From this point of view it wasalso believed crucial to maximize chances to maintain the CNRS label
The CNRS label provides an official recognition to industrial partners Mentioning
your CNRS affiliation when you apply for a grant for example DGA grants also adds
weight to your application They wonrsquot give you more money but they will pay more
attention to your proposal (Professor)
Third Paul built on these concerns to advocate a merger withMicroLab a small CNRSUniMedium unit specialized in microelectronicsSince 19931994 UniMedium had begun to encourage closer connectionsbetween the two labs in order to promote the pooling of experimentalequipment Paul pushed the idea further by proposing a merger to createan institute that would be considered as a major regional hub in electronics
152 JULIEN BARRIER
and telecommunications Paulrsquos views were shared by colleagues atMicroLab who inferred from the observation of lsquosuccessfulrsquo research unitsthat lsquoundersizedrsquo laboratories would be more and more disadvantaged inthe competition for academic positions and funding and may eventuallylose the CNRS label8 They had in mind the creation of another researchcentre the Microelectronics and Physics Institute9 an example of regionalconcentration of research capacities they wanted to follow
We saw what was going on at the national level There was the Microelectronics and
Physics Institute They had gone through a merger they were getting positions money
from the Ministry of research industry local governments and so on And we were
just bystanders waiting on the side of the road By 19951996 we started to realize that
we had to jump in if we wanted to still exist within five years (Senior professor)
This line of argument dramatizing the consequences of inaction wascentral in Paulrsquos discourse In a 1998 presentation he argued that it waslsquothe last chance for MidCity to become a recognized place in electronicsable to compare with other established clusters in Francersquo In sum the mer-ger would provide for a new organizational umbrella that could boost thevisibility of TelecomLab activities and foster synergies between researchers
Some researchers disapproved of Paul as a person casting him as tooambitious But his commitment to develop a new project for TelecomLab inthe absence of competing alternatives put him in a forceful position to defendthe idea of a merger In 1999 members of TelecomLab governing boardvoted unanimously for the merger and for the choice of Paul as the new direc-tor of the research unit that would result from the merger the Electronicsand Engineering Institute The official creation of EEI was planned for theautumn of 2000 However over a few months rising conflicts about thedirection of the merger compromised the completion of the project
Designing the New Institute Solving Old Problems ReactivatingDormant Conflicts (19992002)
Seizing the Creation of a New Organizational Umbrella to RestructureCooperation among ResearchersThe merger proposal had been mainly motivated by the prospect of anincrease in external legitimacy But the very rhetoric of local synergies thatwas put forward to draw external support started to turn into an organiza-tional reality When discussions about the actual implementation of themerger started at fall 1999 a shift occurred as debates moved towards the
153Merger Mania in Science
definition of the governance and organization of EEI By the same tokenthe merger became increasingly understood as an opportunity for restruc-turing and developing a new culture of cooperation thus potentiallybringing the future institute in closer alignment with CNRS organizationalscripts
First Paul pushed forward a reconfiguration of the boundaries ofresearch groups to foster more interaction between researchers working onsimilar research topics While MicroLab was merely to be included as oneresearch group former TelecomLab members were to be combined into twolarge groups that cut across the lines of previous groups Second proposalswere made to pool resources the use of experimental equipment acrossresearch groups would be rationalized and a share of external fundingwould be redistributed across research groups under certain conditions(eg if a research group did not spend all the money from a contract therest would be allocated to a pool that could be used to support otherresearch groups) Third it was argued that research group leaders shouldlimit or avoid internal competition for external resources in particularwhen applying for project funding This issue was explicitly related to craft-ing and maintaining a positive perception of the research centre
Recently answering a call for proposals different members of EEI who are now
regrouped into the Systems research group have submitted competing proposals with
different industrial partners It is important that funding applications should be first
disclosed to the heads of research groups in order to avoid this lsquocacophonyrsquo that is
harmful to the lsquobrand imagersquo we all want for our research unit (EEI governing board
September 1999)
This last point illustrates how the projection of a new image started toentail more than mere window dressing However the more the mergermaterialized into new internal organizational arrangements the more itcame to be seen as a threat to the prevailing order that governed interac-tions within TelecomLab It triggered various conflicts among senior mem-bers who were unwilling to commit themselves to the full adoption of theproposed changes
When Internal Conflicts Outweigh External Pressures Breaking up the NewInstituteConflicts about the merger grew so intense that by spring 2000 the projectwas about to be abandoned Growing concerns about the internal reorgani-zation had been heightened by Paulrsquos proposals about the choice of a dep-uty director and the overall direction of the merger he argued that EEI
154 JULIEN BARRIER
should grow bigger by absorbing more research groups Because Paulsought the support of TechSchool leadership to back his propositions hisactions were soon interpreted in terms of a lsquohostile takeoverrsquo byTechSchool over the future governance of EEI
These suggestions reactivated a series of dormant conflicts among seniorstaff Although these conflicts primarily revolved around Paulrsquos leadershipand the rivalry between TechSchool and UniMedium they also involvedcomplex long time interpersonal quarrels within TelecomLab As a resultit was proposed to break up EEI into two research units Mirroring the for-mation of interpersonal coalitions the boundaries of these two units didnot follow exactly the MicroLabTelecomLab divide nor the boundaries ofTechSchool and UniMedium but cut across research themes and institu-tional affiliation of faculty members
The Institute Resurrected Realigning Internal and External ExpectationsThe merger process could have stopped here However it was soon revivedIn the autumn of 2000 the response of Comite National was clear the exis-tence of two research units was declared unacceptable It threatened to sus-pend the CNRS label if the two units did not eventually merge supportingthe lsquocreation of a single research unit gathering all key actors in Electronicsin MidCityrsquo10
This decision was perceived by several researchers as an electroshock torevive the project Although CNRS did not initiate the merger proposal itacted as an lsquoinstitutional ratchetrsquo preventing the abortion of the process Inaddition to the dire prospect of losing the CNRS label it was feared thatthe process had gone too far to be aborted it would send a very negativesignal about the capacity of the local academic electronics community tocarry out collective projects in the future In other words given the growinglegitimacy of mergers and the expectations raised by the EEI merger project both internally and externally the process had triggered a dynamic ofcommitment among actors (Becker 1960) making the prospect of an abor-tion of the project increasingly costly This interpretation is further sup-ported by the fact that the initiative to revive the merger came from a loosecoalition of junior members from UniMedium and TechSchool who fearedthat their future would be compromised by the inability of senior membersto reach an agreement11
There was a long history of conflicts and we came to see it as a deadlock We were the
young generation and we thought that our future was being hijacked that we needed a
large structure with an international visibility It was time for us to express ourselves to
155Merger Mania in Science
say what we wanted to do even if it meant to drive out those who were in power
positions (Professor)
For them reviving the project was the only viable alternative While aca-demics can decide to leave a research unit to join another laboratory therewas no other large research unit in electronic engineering at MidCity atbest only a few could have lsquoabandoned the shiprsquo The initiative of juniorswas described as a lsquocouprsquo against seniors as the main protagonists pre-viously involved in the merger including Paul were deliberately putaside of the process
Discussions among juniors rapidly converged on the need to find a newdirector among faculty members who had kept some distance from internalrivalries Joseph a senior professor who had been long involved in theadministration of the electronics and physics teaching department atUniMedium was found to be a consensual person who could pacify thesituation He officially became the director of EEI in 2002 The same yearthe Institute was re-affiliated to CNRS with a lsquoprobation clausersquo that is itdepended on the capacity of the new institute to solve internal conflicts anddemonstrate its scientific relevance before being fully affiliated to CNRSin 2004
EXTERNAL AND INTRA-ORGANIZATIONAL
IMPLICATIONS OF THE MERGER
This section is devoted to the consequences of the merger It focuses ontwo issues how the EEI came to be seen as a nodal organization at thelocal level and how the merger affected intra-organizational dynamicswithin EEI
From Rhetoric to Reality and Reciprocally The Construction of aNew Collective Identity
Crafting an External Image Turning the Institute as a Nodal Organizationat the Local LevelThe new director of EEI Joseph was instrumental in turning the creationof EEI into a powerful rhetorical vehicle to draw support from localstakeholders Presenting EEI as a potential nodal organization at the locallevel enabled him to mobilize support which contributed to turning the
156 JULIEN BARRIER
institute into a key local actor In other words while the merger partlyresulted from external pressures to comply with an organizational templatethe legitimacy drawn from the merger was strategically reinvested to leverfurther support
During the first years of his mandate Joseph spent most of his time pay-ing visits to local stakeholders involved in the ICT sector (eg businessassociations territorial development agencies local government) Whileindividual connections existed before Joseph was the first to engage in asystematic approach to the creation and maintenance of a network of rela-tions with local stakeholders Lab council minutes over the 20022006period frequently mention the participation of EEI to local and interna-tional tradeshows in the field of electronics and telecommunications
The mere size of EEI and the absence of internal conflict allowed Josephto fashion himself as the main spokesperson for the local electronic engineer-ing academic community The lsquosuccess storyrsquo of the creation of EEIresonated with narratives circulating about the need to promote more coor-dination at the local and regional level among actors in the ICT sectorindeed local officials were becoming concerned with the decline of the sector(two corporate RampD centres closed down at MidCity in 2002 and 2003)Articles in local newspapers presented the creation of EEI as a positive movetowards overcoming industrial challenges in the ICT sector12 and the insti-tute came to be perceived as a key organization by local stakeholders
Before the creation of EEI there was no large institute in the field of electronics here
simply because people were pulling dirty tricks on each other hellip For us it is very
important to develop synergies to build projects across universities engineering
schools public research organizations hellip We found the restructuring of EEI to be very
positive it brings unity and strength (Local government administrative officer in
charge of higher education innovation and research)
EEI started to be often cited as a prime example to illustrate the aca-demic research potential of MidCity becoming a showcase for visits bybusiness or government representatives
However the increasing connections between EEI and local stakeholdersdid not have only symbolic aspects First external stakeholders saw thedirector of EEI as a contact point with the academic community in the ICTfield Local government administrative officers for instance explained thatit was crucial for them to have a limited number of trusted interlocutorswithin local academic institution in order to delegate to them the coordina-tion of proposals for research and higher education projects For exampleMidCity local government solicited EEI to support the development of
157Merger Mania in Science
new specialized teaching programmes in line with local industrial develop-ment projects they supported In turn the growing renown of EEI and itsinvolvement with local stakeholders meant that the research centre hadbecome a flagship operation for UniMedium and TechSchool Thegrowing importance of EEI also prompted a large research group fromEngineerSchool another institution at MidCity to seek affiliation withEEI in 2004 which contributed to the growth of the Institute
Thus EEI was able to draw support from a wide array of actors Thisproved critical as illustrated by the creation of a lsquocompetitiveness clusterrsquoat MidCity in the field of ICT in 20042005 in the context of a newnational policy initiative13 Since competitiveness clusters would serve aschannels for RampD funding their creation was a matter of fierce competi-tion among firms academic institutions as well as across industrial sectorsand scientific disciplines competing for the inclusion of their goals andpriorities into the definition of clusters at the local level (Younes 2012)
Initially the perimeter and the orientations of the MidCity ICT clusterhad been defined by a small group of representatives from industryRepresentatives from higher education and research institutions were onlyconsulted in a second phase When this phase started EEI was not consid-ered a fully relevant partner since the RampD agenda developed by industrialrepresentatives emphasized research themes in image processing andcomputer science Over a few months with the help of local governmentrepresentatives and EEIrsquos parent academic institutions (UniMediumTechSchool and EngineerSchool) which were statutorily associated tothe creation of the cluster Joseph defended the idea that EEI should be arelevant partner for the cluster
ComputerLab [a large and well established research institute in computer science at
MidCity] had been obviously included in the cluster but my job was to make sure that
EEI would not be forgotten I had to go to a lot of meetings (hellip) It was necessary to
show that we could fit into the agenda (hellip) We had to identify every bit of research
[within EEI] that could be relevant for them as it was dispersed in several research
groups
These efforts proved successful since EEI eventually became a memberof the ICT cluster and was involved in several RampD projects supported bythe cluster Although it is difficult to find hard evidence that fully supportsthis interpretation it is very likely that without both the legitimacy and theinternal coordination capacity lsquospeaking with one voicersquo that resultedfrom the creation of EEI the inclusion of the research unit in the ICT clus-ter would have been compromised
158 JULIEN BARRIER
Building a New Organizational SagaThe creation of EEI set up the foundations for a new organizational saga(Clark 1972) The process that led to the creation of the institute wasreframed in positive terms stressing the ability of researchers to overcomedivergences for the lsquogreater goodrsquo Moreover because junior professors hadplayed a major role in the resurrection of the merger project marginalizingolder faculty members interviewees estimated that the merger had pro-moted a more collegial approach to the organization of research groupsand decision making In addition it was believed that the mergerbrought individual benefits to lab members Some interviewees expresseddoubts about any actual change being brought by the creation of a newstructure it was just an administrative umbrella Yet most of them especially junior faculty members were convinced that the merger pro-vided both an increased reputation and an organizational interface thatfacilitated the funnelling of resources to EEI research groups
We have more visibility Thatrsquos important to attract prospective students hellip Large cor-
porations are hearing stuff about EEI They think telecommunications Thatrsquos
EEI hellip CNRS thinks telecommunications Well thatrsquos MidCity (Professor)
Organizational Restructuring and Changes in Patterns of CooperationThe Mixed Effects of the Merger
The merger also had an impact on the research conducted by EEI particu-larly the internal pattern of cooperation Initially one of the goals of themerger was also to promote an internal reorganization of research groupsin order to foster more cooperation between researchers and more coher-ence across research themes A first reorganization in 1999 defined threelarge research groups Given the internal conflicts of 2000 this structuringwas rather virtual But the same structure remained in place when EEI wasofficially established in 2002 Later in 2004 research group boundarieswere rearranged as one of them was split in two to form a new group (theImage research group) in the same year a team of researchers from anotherinstitution at MidCity EngineerSchool joined EEI to form a fifthresearch group the Communications research group
Appeasing Tensions Developing a New Culture of CooperationIn addition to being involved in the mobilization of external supportJoseph managed to reduce tensions and to promote more cohesion withinEEI Efforts were made to limit internal competition to develop more
159Merger Mania in Science
coordination and to pool resources For example instead of being directlyredistributed to research groups 40 of core funding was reserved bydefault for actions in the lsquocollective interestrsquo of EEI members (eg support-ing external communication) New rules were also emerging to define prio-rities in the allocation of PhD bursaries and to decide on investments inshared equipment While interviewees made it clear that these forms ofinternal coordination were emerging and not fully stabilized they viewedthem as a notable achievement given the conflicts which had plaguedTelecomLab
Although these changes mitigated conflicts and the intensity of directcompetition across research groups the impact of the merger on patternsof internal scientific cooperation was mixed On the one hand publicationoutputs do not indicate significant change in the proportion of scientificpapers co-authored by members of different research groups suggestingthat inter-group cooperation remained extremely limited14 On the otherhand while the co-authorship of papers is a good indicator of fruitful tightscientific cooperation interview data suggests that the merger facilitatedlsquolooserrsquo forms of cooperation Given the epistemic distance betweenresearch groups operating in cognate yet distinct fields cooperationmainly took the form of advice and exchanges around ideas devices analy-tical models or equipment
We could not really function like [laboratory X] where they reorganize teams on a pro-
ject by project basis Itrsquos simpler for them they are all computer scientists so you can
have many combinations Here itrsquos different we are quite heterogeneous Irsquom a remote
sensing specialist I know as much as you do about microelectronics hellip Of course if I
have a propagation problem Irsquom not going to read all the literature Irsquoll go to see the
specialists from the other group (Professor)
Another form of cooperation was the joint supervision of master-levelstudents and research interns allowing sharing expertise and testing therelevance of emergent ideas In addition researchers also reportedinstances of cooperation in large collective research projects especiallywith industry The basic rationale for the promotion of internal coopera-tion on projects was to minimize the coordination costs involved in largeprojects with multiple partners This was deemed especially relevant forindustrial projects since they usually require more integration andfinalization
People from the Systems group collaborate with our group [Telecom devices] We can
develop an antenna and all the system around it Then they can tell their prospective
industrial partners look we can do a lot here No need to go looking somewhere
160 JULIEN BARRIER
else hellip You can develop different basic blocks in the same place thatrsquos easier for an
industrial partner to assemble and coordinate all the blocks afterwards (hellip) We have
been able to do this for project X for instance (Professor)
Again while they did not necessarily resulted in tight scientific coopera-tion (ie researchers involved in different work packages in a large projectmay adjust and learn from each other without developing joint resultsworthy of publication) the development of such projects represented astrong departure from the former situation In sum while it is absolutelyclear that the restructuring could not force cooperation upon researchers especially given the diverse scientific profile of EEI it is important tonote that it fostered attempts at cooperation and thus enabled potentialcooperation to occur The emergence of a collective external identity inaddition to the regulation of internal competition had increased the incen-tives for cooperation
Cooperation within Research GroupsThe restructuring created some changes in cooperation patterns withinresearch groups albeit with different outcomes from one group to anotherThree different configurations emerged from interviews and bibliometricdata on co-authorship links between group members as shown in Fig 1The two periods presented in Fig 1 (19992003 and 20042008) corre-spond to two successive formal reorganizations of the institute first in 1999with the creation of three research groups and second in 2004 with a limitedrestructuring of research groups To draw co-authorship maps the list ofresearch group members for the two periods was retrieved and then matchedwith ISI data These graphs could be compared to long-exposure photo-graphs capturing the sedimentation of co-authorship links in two differenttime periods Since networks also evolve within each period the differencesbetween periods should not be interpreted in terms of sharp change In addi-tion Fig 2 put group co-authorship networks into the larger context of EEIduring the same period
In a first configuration illustrated by the lsquoImage research grouprsquo theconstruction of a group was mainly driven by symbolic purposes and hadalmost no impact on intra-group cooperation The group consisted of themerger of two small research teams one working on remote sensing andthe other on image processing which had different orientations andresearch themes The creation of this lsquoumbrella structurersquo intended to bal-ance the size of research groups in order to present image processing as akey topic for EEI to external audiences
161Merger Mania in Science
2004-2008
8 2
1
3
10
5
4
3
712
11
13 14
7
15
6
810
12
210
9
21
4
398
5 1015
1213
14 18
76
11
10 12
13
17
15163
4 2
8
1167
84
5
6
11
12
7
9
6
Image Research Group
Microelectronics Research Group
Telecom Devices Research Group
1999-2003
Ellipses Former boundariesof TelecomLab groups
2004-2008
1999-2003 2004-2008
Fig 1 The Evolution of Co-Authorship Networks in Three Research Groups
Note Nodes represent researchers who authored scientific papers only permanent
faculty members are taken into account excluding PhD students post-docs and lab
technicians Links represent co-authorship relations their thickness represent the
relative intensity of links (ie number of co-authored papers) To facilitate
comparison within a group across the two periods an identification number has
been attributed to each researcher In addition two colours have been used for
nodes to distinguish between researchers who are active in both periods (black
nodes) and researchers who are active in only one period (white nodes) For
instance white nodes appearing only in the 19992003 period represent researchers
who have either left the group (eg retirement promotion) before the following
period or who did not publish any paper in the following period
People working on image processing were really under the critical size [to be a
research group] and CNRS made remarks about it Well from a certain angle it
might be said that my team works on image So we left the Systems group to create
the Image Research group Thatrsquos also because the Systems group was too big and
we needed to balance the size of research groups But we donrsquot have the same defini-
tion of image as the other part of the Image research group (hellip) Wersquore just two
teams that get along with each other We actually exchanged some algorithms but
Fig 2 The Evolution of Co-Authorship Patterns in the Electronic Engineering
Institute
163Merger Mania in Science
that was mainly on a case by case basis We do not have any cooperation on a joint
project (Professor)
In this case the formal boundaries of the group do not correspond to aresearch collective Research is actually performed by small informal teamsof 23 group members who collaborate on a regular basis But these smal-ler teams within the group are independent since their cognitive activitiesare not based on a complex division of labour (mainly design modellingand data processing with few experimental tasks)
In a second configuration illustrated by the lsquoMicroelectronics researchgrouprsquo the creation of a group merely turned an integrated research col-lective into a formal structure No need was felt to restructure MicroLabafter its inclusion in EEI Indeed MicroLab used to function on a highdivision of labour between its members This situation resulted from aheavy emphasis on complex experimental work going hand in hand with aspecialization of members on different aspects of the research process Inaddition to a vertical division of labour where senior researchers were incharge of looking for funding for the whole group in order to allowjuniors to focus on experimental work there was a significant horizontalspecialization among junior members as illustrated by the following inter-view quote
Professors X and Y really focus on bringing the money in going to meetings they have
a managerial role hellip As junior professors we are also involved in administrative work
but we are more orientated towards experimental work hellip And we also work for other
group members so to speak Irsquom in charge of running the materials processing equip-
ment and Z is preparing thin layers for the whole group (Junior professor)
The tight interdependence among researchers in this group is furtherindicated by co-authorship patterns which suggest a strong integration atthe group level with a relatively stable configuration over time
In a third configuration illustrated by the lsquoTelecom devices researchGrouprsquo the restructuring had a stronger impact on cooperation As thisgroup included researchers coming from three different groups atTelecomLab it initially had a high potential for conflict But under the lea-dership of younger academics the restructuring was precisely seized as anopportunity to foster the sharing of resources and to limit intra-groupcompetition
We want to avoid the old rubbish situation hellip We have defined 5 main research
topics hellip People can work on one or two research themes But they canrsquot work on all 5
We try to avoid that hellip And therersquos one person in charge of each of them who has to
channel all the relevant information to the others hellip Just a week ago I had three
164 JULIEN BARRIER
contacts with people from industry I transmitted one of them to X [a colleague in
charge of another research theme] And my colleague sent me a proposal for an indus-
trial collaboration that did not fit into his research theme (Professor)
Areas of specialization still reflected divisions among former groups to acertain extent but the restructuring also allowed for the emergence ofshared themes Although change was facilitated by the marginalization ofthe former heads of research groups one of them retired and the twoothers progressively disinvested themselves from research at EEI intervie-wees explained that they still put much effort lsquoto keep the situation into con-trolrsquo The limitation of internal competition was a necessary albeit notsufficient condition for cooperation to develop between group membersThe merger and the subsequent restructuring was used as a lever to exploitpotential cognitive complementarities between researchers as suggestedboth by interview data and the visualization of co-authorship patterns
I supervise half of my PhD students with X There is also a PhD starting soon in coop-
eration with Y hellip For a long time I kept the research line Z for myself This is new
(Professor)
The evolution of co-authorship patterns confirms that researchers thatpreviously had few or no contacts begun to develop closer collaborationBut one may note that they significantly differ from those of the highlyintegrated Microlectronics group This relates to epistemic and materialfactors Knowledge production in this group does not depend as much onheavy tightly interdependent experimental task It combines diverseactivities (eg fabrication of relatively simple devices experimental charac-terization modelling and simulation) The elementary structure of thesetasks and the nature of the functional interdependencies they create was unaffected by the restructuring But it enabled the exploration of newcombinations of topics and expertise
DISCUSSION AND CONCLUSION
In analysing this case of merger and restructuring in an organized researchunit this article shows how the formal organization of research activities ischanging in relation to the increasing diffusion and significance of newinstitutional scripts in the academic field In this perspective in line withrecent efforts to revisit the concept of decoupling in institutional theory(Hallett amp Ventresca 2006) it argues against a tendency in the literature to
165Merger Mania in Science
overemphasize the extent of symbolic compliance in organizations15
Instead our intent was to examine how institutional myths are translatedinto local arrangements and practices Three main contributions can beoutlined from our case study
First of all this article sheds further light on an important trend in theorganization of public research in Europe that is the concentrations andmergers of research units to form lsquocritical sizersquo organizations (Bonaccorsi ampDaraio 2005) Departing from quantitative studies of the relations betweenorganizational size and scientific performance (eg Horta amp Lacy 2011) itused a case study to understand how this institutional myth is edited andtranslated at the local level This case confirms the growing legitimacy ofan emerging organizational template in the French research system thatis the large regional-based research institute but it also reveals two keyfeatures of the enactment and translation of this myth First our case illus-trates the performative dimension of the lsquocritical sizersquo myth as the mergerunfolded like a self-fulfilling prophecy The merger was perceived as a solu-tion to assert the position of electronic engineering research at MidCitywhich in turn fuelled actions intending to capitalize on the merger to estab-lish EEI as a key organization at the local level Although the merger wasinitially a defensive response to changing environmental pressures it waslater instrumentally mobilized as a rhetorical vehicle to boost the renownof the research unit in order to consolidate external networks A secondfeature concerns how the issue of organizational size was framed by actorsWhile economies of scale are usually salient in discourses promotingmergers in academia in our case actors reframed the merger as an organi-zational expansion turning it into a way to extend and diversify sources ofexternal support
In addition because notions of lsquocritical sizersquo put forward by policymakers are ambiguous our data suggests that the increasing legitimacy ofEEI did not so much derive from its conformity to a static well definedorganizational template than it stemmed from the process of aggregationand expansion In other words lsquobeing bigrsquo may matter less than lsquogettingbiggerrsquo This interpretation supported by the recent collection of follow-upinformation about EEI which has continued to grow since fieldwork wascarried out by integrating research teams from other academic institutions is consistent with the idea that institutional scripts of rationalization haveincreasingly caused reform to become endemic in organizations (Bromley ampPowell 2012)
Second our empirical findings may contribute to further exploration ofthe interplay between institutional myths organizational structures and
166 JULIEN BARRIER
practices in the scientific field issues that are not frequently addressedtogether Concerning the link between institutional pressures and organiza-tional structures a first interesting aspect of our case is that change withinEEI did not take the form of a perfect compliance to external pressuresnor was it mere window dressing Environmental pressures operatedthrough indirect processes catalysing rather than triggering change bycontributing to destabilize existing local arrangements in a chain reactionInterestingly while mergers are generally understood as a way to reinforcerelations between merged entities the merger had less impact on the linksbetween MicroLab and TelecomLab than within TelecomLab The mergerproposal threatened established defensive territories (eg the fragile equili-brium in relations between research groups at TelecomLab) a situationexacerbated by divergences about the direction of the merger which in turnprovoked a conflict among seniors almost suspending the merger processThis disequilibrium enabled junior academics to foment a lsquocouprsquo againstseniors leading to their subsequent marginalization and eventuallybrought the organization into closer alignment with the institutionalscripts diffused by CNRS (ie research centres as cohesive integratedorganizations)
Regarding the impact of organizational restructuring on research activ-ities two main results can be outlined On the one hand it appears ratherlimited in terms of changes in scientific output This certainly relates to thefact that the restructuring was unconnected to major epistemic change atthe field level in contrast with cases of thorough departmental overhaulsaccompanying the emergence of a new discipline or specialty where localreorganizations are interwoven with deep multi-level changes in the struc-ture and epistemology of a field (eg Wilson amp Lancelot 2007) In ourcase this dimension was absent the restructuring being primarily driven bya combination of non-discipline-specific institutional pressures and con-cerns over the coordination of research at the local level On the otherhand however especially in comparison with the Vinck and Zarama(2007) study of a merger between two research units in the field of micro-electronics in France our empirical evidence indicates that the merger andrestructuring did have some effects on research practices albeit mainly interms of coordination and regulation of internal competition16 Then ourresults suggest that while organizational restructuring may not affect thebasic patterns of scientific work (eg nature of core tasks types of func-tional interdependencies among researchers) structures are only partiallydecoupled from core activities Their impact is indirect because they areused by academics to define lsquodefensive territoriesrsquo (Musselin 1990) they
167Merger Mania in Science
may inhibit or enable potentially relevant cooperation They do not asmuch bring coordination by supervision as they provide spaces forexchange learning and mutual adjustment to occur This last point sug-gests that it would be worthwhile to investigate the effects of organizationalrestructuring over longer periods of time to understand to what extent(and in what conditions) mere membership to the same structure mightfavour the exploitation of epistemic complementarities
Third our results complement and extend Musselinrsquos propositions aboutthe properties of formal structures in universities (Musselin 1990 2007)Although Musselin mainly discusses the relation between structures andfunctional interdependence our case suggests that formal structures mayactually produce another form of interdependence among their constitu-ents namely lsquointerdependence of fatersquo (Lewin 1948) In contrast with taskor functional interdependence it refers to the idea that the fate of a groupmember is inseparable from the fate of the group as a whole just like thepassengers of a raft after a shipwreck In our case the creation of a newinstitute initially intended to provide researchers with a new organizationalfacade to draw further support and legitimacy But the virtual existence ofEEI as a project paradoxically began to display potentially real (and nega-tive) consequences for researchers when it was nearly aborted Because theproject had raised external expectations about the evolution of electronicengineering research at MidCity as a whole its abortion would have hadcollective reputational costs for this group thus making local academicsinterdependent in the face of adversity17
In other words in this particular context the formalization of a newstructure had created a latent community of fate While interdependence offate may be particularly salient when a group faces collective threats ornegative sanctions it is also significant when actors benefit from the collec-tive reputation of the structure they belong to This is well illustrated bythe fact that the increasingly positive external perception of EEI as a keyactor in ICT at the regional level was seen by members as a factor enhan-cing their individual reputations further consolidating their sense of mem-bership to EEI This property of formal structures is particularly visible inthe case of research centres in the French system which are a salientdeeply institutionalized level of aggregation in the evaluation of scientificperformance Indeed in addition to establishing a strong sense of member-ship one can hypothesize that a key precondition for a structure to pro-duce a community of fate is the fact that they are subjected to collectivesanctions positive or negative Then this property of formal structures
168 JULIEN BARRIER
may also be significant in other academic systems granted that the struc-ture in question be it a department an institute and so on is taken asan object of external scrutiny evaluation or steering
Coming back to the initial puzzle of this article the ambiguous roleof formal structures in academic settings this case study suggests thatthe impact of formal reorganization on research practices depends on thealignment of external and intra-organizational conditions While it is clearthat impact is likely to be all the more significant as change in structuresresonate with wider environmental pressures it ultimately depends onexternal legitimacy being tightly linked to the mobilization of materialresources In other words when the symbolic value of structures is nota key asset or at least not believed or anticipated to be a key asset byactors in the competition for material resources changes in structuresare likely to go unheeded In addition although it is also clear that thealignment between formal structures and epistemic patterns of coopera-tion is a crucial factor a result consistent with Vinck and Zarama(2007) the impact of reorganization on practices does not depend asmuch on an lsquoobjectiversquo fit between these two elements as it depends onactors using or not structures as a way to strategically pursue theirepistemic interests
NOTES
1 For example the French General accounting office described the fact that80 of academic research units counted less than 25 members as a lsquohandicap forinternational competitionrsquo (Cour des Comptes 2005 p 126)
2 Since the legitimacy and ultimate survival of an organization depends on theadoption of institutional scripts they might decouple their practices from structuresin order to protect the efficiency of their core activities
3 Its full name is Comite National de la Recherche Scientifique but to avoidconfusion with CNRS it is generally simply referred to as Comite NationalMembership of the Comite National consists of both CNRS and university person-nel who are appointed every 4 years (23 of the members are elected and 13 areappointed by the CNRS) It is divided into about 40 sections this number hasvaried over time and successive reorganizations corresponding to scientificspecialties For instance the EEI is affiliated to a section covering electronic andelectrical engineering photonics as well as micro and nanotechnology
4 Our examination of evaluation reports in different research units and probesinto CNRS central archives (especially 040074 STIC boxes 16 and 17 060107 INGbox 4 and 060111 ING box 16) indicate that the application of these organizational
169Merger Mania in Science
criteria is consistent both over time and across research units in the field of electro-nic engineering
5 Legally they are employees of the CNRS at the national level affected tolaboratories at the local level Upon acceptance by CNRS researchers can decide totransfer from one research centre to another
6 Data were compiled from different CNRS surveys and directories To iden-tify research centres operating in the field of electronic and electrical engineeringwe relied on their affiliation to the relevant section of Comite National
7 In another instance board members complained that their colleagues did notshare information about their discussions with prospective industrial partners
8 MicroLab had been seeking to merge with another unit Actually a part ofMicroLab members were not located at MidCIty but in another smaller city By1994 this part of the lab decided to leave MicroLab to join another research unitHence MicroLab was considered undersized by CNRS A former proposal tomerge with a laboratory in physics had been rejected by CNRS for being not con-vincing in terms of scientific complementarities
9 Created in 1992 this CNRS research unit was the first to result from anambitious merger involving three different laboratories in the field of electronicengineering10 Report CNRS archives 040085 STIC box 811 Because there were no other existing research units at MidCity in electronic
engineering12 With titles like lsquoA major actor in the field of telecommunications TechSchool
and UniMedium unites to create a new research institutersquo (local newspaper article2002) and lsquoElectronics seeking a new lease of life 180 scientists prepare the futureof industrial innovationrsquo (local newspaper article 2003)13 In brief competitiveness clusters are territorial-based structures gathering
firms and academic institutions in order to foster the development of specializedlocal innovation networks in targeted areas (eg aeronautics in the Toulouse area)giving access to special funding for RampD projects Their creation depended on abottom-up competitive process a national call for proposals invited local actors tolsquodefine their goals their relevant territory and the actions they wanted to taketogether in order to reach those goalsrsquo (Younes 2012 p 341)14 From 1999 to 2003 6 out of 247 publications (24) were co-authored by
members of two different research groups From 2004 to 2008 the figure was 7 outof 349 publications (18) See also Fig 215 Our argument should not be mistaken as the trivial notion that institutional
pressures can actually have an impact on practices But given the narrow under-standing of the notion of decoupling that has become prevalent in the literature itis relevant to insist on that issue to explore further the significance of the diffusionof scripts for organizational practices16 In the case they investigated Vinck and Zarama (2007) concluded that the
merger did not have any significant impact on scientific activities and barelyaffected preexisting collaborative patterns17 This holds true because there was no alternative for most academics as they
could not easily lsquoabandon the shiprsquo For instance transferring to another researchcentre was not an easy option because MicroLab and TelecomLab were the twomain centres at MidCity
170 JULIEN BARRIER
ACKNOWLEDGEMENTS
This article draws data from a research project funded by the EuropeanCommission (PRIME Network of excellence) I also acknowledge additionalsupport from ENS Lyon IFE and CNRS (Chair on Higher Education ampResearch Policy) and I thank Pierre Barrier and Barbara Bovy for their parti-cipation to data collection Participants to the lsquoOrganizing Sciencersquo Subthemeat 2012 EGOS conference provided helpful comments on a preliminary ver-sion of this article Many thanks to Jochen Glaser Grit Laudel UweSchimank Richard Whitley and an anonymous reviewer for their construc-tive comments on earlier versions
REFERENCES
Aust J amp Crespy C (2009) Les collectivites locales face a lrsquoenseignement superieur et a la
recherche Pouvoirs locaux 10 6472
Barrier J (2011) Following scientists following the money Project funding professional
autonomy and changing work patterns in academic research Sociologie du travail
53(4) 515536
Barrier J amp Mignot-Gerard S (2013) Leadership et changement dans une organisation
pluraliste In I Vandangeon-Derumez amp D Autissier (Eds) Le changement organisa-
tionnel Etudes de cas commentees (pp 121139) Paris France EyrollesEditions
drsquoOrganisation
Becker H (1960) Notes on the concept of commitment American Journal of Sociology 66(1)
3240
Binder A (2007) For love and money Organizationsrsquo creative responses to multiple environ-
mental logics Theory and Society 36(6) 547571
Bonaccorsi A amp Daraio C (2005) Exploring size and agglomeration effects on public
research productivity Scientometrics 63(1) 87120
Bromley P amp Powell W (2012) From smoke and mirrors to walking the talk Decoupling in
the contemporary world Academy of Management Annals 6(1) 483530
Clark B R (1972) The organizational saga in higher education Administrative Science
Quarterly 17(2) 178184
Cour des Comptes (2005) La gestion de la recherche dans les universites Report to the
President of the Republic Paris France
De Nooy W Mrvar A amp Batagelj V (2005) Exploratory social network analysis with
Pajek Cambridge Cambridge University Press
Drori G Meyer J Ramirez F amp Schofer E (2006) Science in the modern world polity
Institutionalization and globalization Stanford CA Stanford University Press
Fixari D Moisdon J-C amp Pallez F (2009) Lrsquoevaluation des chercheurs en questions
(19922009) Paris France Presses des Mines
Frickel S amp Moore K (2006) Prospects and challenges for a new political sociology
of science In S Frickel amp K Moore (Eds) The new political sociology of science
(pp 333) Madison WI The University of Wisconsin Press
171Merger Mania in Science
Hallett T (2010) The myth incarnate Recoupling processes turmoil and inhabited institu-
tions in an urban elementary school American Sociological Review 75(1) 5274
Hallett T amp Ventresca M (2006) Inhabited institutions Social interactions and organiza-
tional forms in Gouldnerrsquos patterns of industrial bureaucracy Theory and Society
35(2) 213236
Horta H amp Lacy A (2011) How does size matter for science Science and Public Policy
38(6) 449460
Kitchener M (2002) Mobilizing the logic of managerialism in professional fields The case of
academic health centre mergers Organization Studies 23(3) 391420
Krucken G (2003) Learning the lsquonew new thingrsquo On the role of path dependency in univer-
sity structures Higher Education 46(3) 315339
Krucken G amp Meier F (2006) Turning the university into an organizational actor In
G Drori J Meyer amp H Hwang (Eds) Globalization and organization World society
and organizational change (pp 241257) Oxford Oxford University Press
Laredo P amp Mustar P (2002) Innovation and research policy in France (19802000) or the
disappearance of the Colbertist State Research Policy 31(1) 5572
Latour B amp Woolgar S (1979) Laboratory life The social construction of scientific facts
Beverly Hills CA Sage Publications
Laudel G (2002) What do we measure by co-authorships Research Evaluation 11(1) 315
Lewin K (1948) Resolving social conflicts Selected papers on group dynamics New York
NY Harper
Louvel S (2010) Changing authority relations within French academic research units since
the 1960s From patronage to partnership In R Whitley J Glaser amp L Engwall
(Eds) Reconfiguring knowledge production Changing authority relationships in the
sciences and their consequences for intellectual innovation (pp 184210) Oxford
Oxford University Press
Meyer J amp Rowan B (1977) Institutionalized organizations Formal structure as myth and
ceremony American Journal of Sociology 83(2) 340363
Musselin C (1990) Structures formelles et capacites drsquointegration des universites francaises
et allemandes Revue Francaise de Sociologie 31(3) 439461
Musselin C (2007) Are universities specific organizations In G Krucken A Kosmutzky amp
M Torka (Eds) Towards a multiversity Universities between global trends and national
traditions (pp 6384) Bielefeld Germany Transcript Verlag
Sauder M amp Espeland W N (2009) The discipline of rankings Tight coupling and organi-
zational change American Sociological Review 74(1) 6382
Shinn T (1988) Hierarchies des chercheurs et formes des recherches Actes de la Recherche en
Sciences Sociales 74 222
Theves J Lepori B amp Laredo P (2007) Changing patterns of public research funding in
France Science and Public Policy 34(6) 389399
Vinck D amp Zarama G (2007) La fusion de laboratoires Processus de gestion et constitu-
tion drsquoune entite pertinente de lrsquoactivite scientifique Revue drsquoanthropologie des connais-
sances 1(2) 276296
Weick K E (1976) Educational organizations as loosely coupled systems Administrative
Science Quarterly 21(1) 119
Wilson D amp Lancelot G (2007) Making way for molecular biology Studies in History and
Philosophy of Science Part C 39 93108
Younes D (2012) Choosing the industry of an industrial cluster in a globalizing city Journal
of Change Management 12(3) 339353
172 JULIEN BARRIER
CONTRADICTORY
CONSEQUENCES OF
INSTITUTIONAL CHANGES ON
INTELLECTUAL INNOVATION IN
THE PUBLIC SCIENCES
INSTITUTIONAL CONDITIONS
AND CHANGING RESEARCH
PRACTICES IN SWITZERLAND
Martin Benninghoff Raphael Ramuz
Adriana Gorga and Dietmar Braun
ABSTRACT
This article analyses in what way Swiss academic institutions have hada favourable or unfavourable influence on changing research practicesby following developments in four scientific areas Bose-EinsteinCondensates Evolutionary Developmental Biology Large-Scale Assess-ments in education research and Computerised Corpus Linguistics Basedon empirical evidence we argue that overall a number of institutionalconditions have had a positive influence on the decisions of scientists todare a switch to a new scientific field One finds however also differ-ences in the working of these institutional conditions leading to quickeror slower developments of the four selected scientific areas
Keywords Public science systems Switzerland research practicesscientific innovations universities research councils
Organizational Transformation and Scientific Change The Impact of Institutional
Restructuring on Universities and Intellectual Innovation
Research in the Sociology of Organizations Volume 42 175202
Copyright r 2014 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN 0733-558Xdoi101108S0733-558X20140000042006
175
INTRODUCTION
This article contributes to the continuing discussion about the institutionalconditions of knowledge development (Bonaccorsi 2007 BonaccorsiDaraio amp Geuna 2010 Braun 2014 Whitley 2000 2003 WhitleyGlaser amp Engwall 2010) by analysing one particular research systemSwitzerland and assessing how Swiss academic institutions have had afavourable or unfavourable influence on intellectual innovations byfollowing the growth of research in four scientific areas Bose-EinsteinCondensates (BEC) Evolutionary Developmental Biology (ED) Large-Scale Assessments (LSA) in education research and Computerised CorpusLinguistics (CCL) In comparison to other contributions in this editedvolume where the institutional context varies while the dependent variablethe change in research practices is constant that is only one scientific fieldis analysed we look into the relationship of only one institutional contextwith a variety of scientific fields in which a change in research practices hastaken place
While the former research design enables the identification of the institu-tional conditions that are favourable or unfavourable for one scientificfield our approach is aimed at clarifying whether one and the same institu-tional context matters in the same way for changes in research practices indifferent scientific fields A strong variation in outcomes that is in thecapacity for new scientific fields to grow would suggest that the influenceof the institutional context differs between kinds of scientific field It mightbe positive for some scientific fields but not for others which helps us tounderstand the particular needs of institutional support for different scien-tific fields An overall strong capacity to spread in contrast would meanthat the institutional context has a similar positive influence throughoutscientific fields An overall weak capacity to change research practices indi-cates an institutional context with deficient support structures
Although the capacity to generalise on the base of this design remainslimited because only four case studies are considered they do represent themajor branches of the sciences physical sciences (BEC) life sciences(ED) social sciences (LSA) and humanities (CCL) Such contrasted casescan give at least an insight into the differences between change in researchpractices across the sciences and the impact of favourable or unfavourableinstitutional conditions Further research is needed to corroborate ourfindings
We proceed in the following way The next section presents conceptualconsiderations that have guided our empirical research The nature of
176 MARTIN BENNINGHOFF ET AL
scientific growth and diffusion will be defined and operationalised andwhat is needed to change lsquoresearch practicesrsquo discussed The different needsof scientific fields and the institutional conditions that can influence theirgrowth will be presented We then provide a short overview of the generalinstitutional conditions in Switzerland before we discuss the four case stu-dies The discussion describes the growth of the four cases in Switzerlandand explores the extent to which Swiss institutions have had a generallysupportive or constraining influence on all four of the developing scientificfields or rather had different effects across them In the conclusions wecome back to our research question To what extent do different kinds ofscientific institutions affect the expansion of various scientific fields
CONCEPTUAL CONSIDERATIONS
Scientific growth and diffusion are for us part of the process of lsquoscientificinnovationrsquo that is the production and diffusion of new knowledgecomponents This signifies the decision of scientists or researchers tochange their research practice1 or in other words to lsquoswitchrsquo and start towork in a yet relatively unexplored cognitive domain Original findingsare usually the starting point of the process of scientific innovation Suchfindings can only become scientific innovation if they are acknowledgedand become accepted by other scientists This happens if these otherscientists start to explore further the implications of these findings A newscientific field is created if a significant number of scientists are drawninto this exploration process and change their research practices in orderto be able to do so Institutionalisation processes in form of the creationof regular research networks associations journals and chairs in universi-ties follow
We are interested in the decision of scientists to change their researchpractice as the decisive building block for the establishment of a new scien-tific field If lsquopioneersrsquo the scientists that have made the discovery do notfind lsquofollowersrsquo the original findings will become forgotten and part of thehistory of science When therefore do scientists decide to change researchpractices In order to answer this question we use an economic interpreta-tion of scientific behaviour (Braun 2012a Kitcher 1995 Mirowski amp Sent2002 Shi 2001) Such a view presupposes that scientists take decisions in this case to enter or not to enter a new scientific field by using infor-mation about the costs and benefits involved A switch to a new field must
177Institutional Conditions and Changing Research
be lsquoattractiversquo meaning it should clearly result in more benefits than thescientist has at the moment and costs to switch should be low
Decision-Making Variables of Scientists
There are a number of variables that enter into the decision processes ofscientists considering entering new areas Only some of them are subjectto direct manipulation by institutional means in universities and fundingagencies Both the search for scientific reputation and for more lsquoeffective-nessrsquo that is better results in terms of knowledge advancement are rela-tively independent of external factors They are either decided by processestaking place in the scientific community or are a result of the cognitiveadvances that can be made once a switch has taken place Scientists canexpect some degree of better effectiveness but this is based on cognitiveevaluations of the possibilities of the new scientific field
Other variables can be manipulated or influenced
(1) Material resources have a high status in decisions of scientists Theresearcher will for example assess whether she has reasonable chancesof getting or maintaining an attractive employment when switching toa new area and she will ask whether there are sufficient chances to findresearch money research personnel and research infrastructure in thenew area If for example funding agencies do not fund the new areathe costs to find other research money will rise and the attractiveness ofthe new field will decline
(2) The time at disposition for research is another decision variableExploring new areas often need more time than continuing to work inestablished areas If research organisations or funding agencies do notprovide lsquoprotected timersquo for such longer-term investments a switchmight become unattractive Another time aspect is the amount of timeavailable for doing research compared to other activities of scientistslike teaching and administration The change of research practicesneeds a considerable amount of time that must be invested in researchInstitutional conditions define to some extent the distribution of timeof research available for scientists They can grant more or less time forresearch in different areas
(3) Academic support refers to the possibility of institutionalising thenew field within the scientific community New fields can engenderunfriendly behaviour of fellow scientists working in established fields
178 MARTIN BENNINGHOFF ET AL
either by subverting publications by not granting access to scientificassociations or by resisting to the institutionalisation of the new fieldwithin the university Switchers who meet such resistance have highercosts to carry in order to overcome such resistance and many may bedeterred to switch at all when they realise that the new field may raisesuch academic resistance New interdisciplinary fields are particularlysubject to such resistance and will in general raise higher switchingcosts Academic support can however also be benevolent and favour-able for scientist The new area may be seen as a fruitful complementto existing research topics in an institute or department The lsquoswitcherrsquomay then find backing and support to build up the new research areawhich lowers switching costs
Material resources time and academic support are therefore we sug-gest the major conditions that enter into the decision-making of scientistswhen thinking about switching research practices
Types of Scientists
Switching costs are not always the same for all scientists Apart from thefact that scientific fields may raise different cost levels for examplebecause of high infrastructural needs for research equipment as in the caseof ED and BEC one can distinguish between two types of scientists inthis respect
One is the distinction between those scientists who are heavily involvedin the lsquobuildingrsquo of the new research area and those who are prepared towork in the new area and simply use what the lsquobuildersrsquo have developedThis distinction becomes most appropriate in the case of CCL and LSAwhere the creation of the lsquoresearch toolrsquo that is databanks is the maininitial contribution to be made Obviously this needs considerable timeinvestments and material resources The same is true for BEC in the case ofwhich scientists search for methods to lsquofreezersquo the movement of atomsUsers in all three cases profit from the work of builders and develop scienti-fic knowledge by using the research tools prepared Investments to be madeare accordingly much lower It needs of course both types of scientists tospread the field throughout the scientific community
The second distinction of types of scientists with regard to switchingconcerns the strategies scientists can employ in order to change researchpractices Time investment is the important variable here Scientists may
179Institutional Conditions and Changing Research
either abandon completely their former research practices (and invest alltheir energy and time in the new field) or they may use only part of theirtime while maintaining also part of their research activities in existingresearch areas One could call the first type of scientists lsquofull switchersrsquo andthe second type lsquopartial switchersrsquo Full switchers take considerably morerisks when switching because if they fail to succeed in the new area theywill suffer from loss of reputation while a partial switcher can to someextent fall back on the reputation she has built up in existing research areasand which has been further developed even when exploring the new fieldFull switchers need more lsquoprotected spacersquo that is institutional support interms of money and time resources than partial switchers Institutionalconditions support of public research organisations like universities andfunding agencies can influence the decision of scientists in this respectGenerous and unconditional support by these institutions makes it easierfor scientists to take the risk of full switching
Institutional Conditions
Institutional conditions vary in time and space and lead to different switch-ing costs and opportunities for scientists to change research practicesWhat kind of institutional conditions present attractive conditions forswitching Which ones are detrimental to changing research practices Weendeavour below to formulate some expectations about the relationship ofthe three decision-making variables we have formulated above materialconditions time resources and academic support and institutionalconditions
Institutional Conditions and Material ResourceslsquoRationalrsquo scientists planning to switch prefer one can assume havingabundant resources easy access to financial resources and a flexible use ofthese resources The most fundamental institutional condition in thisrespect is to what extent funding money is available in the national researchsystem if there are conditions of lsquoaffluencersquo or of lsquoscarcityrsquo (see Heinze ampMunch 2010 Ziman 2002) In periods of affluence funding money cannourish a large number of different research areas which raises the chancesof scientists to find financial support even in those areas of research thatare generally considered as being risky Having abundant funding allowsscientists to distribute risks among a large number of projects Chances to
180 MARTIN BENNINGHOFF ET AL
become funded in this case is high for scientists going into new and unex-plored knowledge fields lsquoScarcityrsquo by contrast reduces the willingness offunders to take such risks Because fewer projects can be funded funderstend to support areas where results are certain which means most of thetime mainstream research with incremental knowledge progress There ismoreover a tendency to become more lsquoselectiversquo and to cut down averageresources per project Scientists have therefore higher lsquosearch costsrsquo in orderto find funding they must do with less resources and have fewer chancesto launch high-risk research projects
Research systems can however also be lsquoselectiversquo even in periods ofaffluence In this case it is a matter of belief about the best way to supportdiscoveries and innovations One narrative is that choosing the best scien-tists or projects and to supply them with abundant resources will result inhigher productivity than a distribution of funding money over a large num-ber of scientists and projects This narrative can lead to investments into aselected number of high risk and new research fields Switchers who suc-ceed in obtaining such funds can probably count on excellent material con-ditions but the chances to obtain these funds diminish The knowledge areathe scientist wants to work in must fit with those areas selected by fundersIf funders accept that the selected areas are chosen by scientists themselvesthe chance to obtain such funds rise but need considerable investments inorder to compete with other projects
A funding system with a plurality of funding sources is another institu-tional condition that might have a positive effect on the change of researchpractices It increases the chance to find the funding for research projectsbecause such a system allows for more variety in research Shrinkingfunding resources in one funding agency may be compensated by risingresources in another
Getting funds with abundant research money is one aspect To disposeof these funds in a flexible way is another The flexibility of fundingdepends on the policies of public research organisations in particular uni-versities to grant their scientists long-term and unconditional researchmoney Funding agencies usually give money for a short period of timeand demand a relatively precise outline and budget Recently there havebeen changes however with the launching of longer-term grants for exam-ple for centres and also occasionally to individual scientists (the advancedgrants of the ERC are a good example here) Such policies would contri-bute to a higher flexibility in the use of funding money which is importantfor switchers who work in areas that are new and must still be explored
181Institutional Conditions and Changing Research
Institutional Conditions and Time ResourcesThe fundamental decision to be made in research systems is to what extentfunding resources should be available only for shorter periods of time which reduces risks for funders as the invested amounts of money for oneproject remain low or for longer time periods which raises investmentcosts and needs substantial trust in the capabilities of scientists to beproductive Often it is institutional funding which is needed to supportlong-term areas of research or it may be project funding with a long-termperspective Another important condition is the availability of stableemployment positions that can be granted to scientists working in newareas One can think here of tenure-track or tenured positions The capabil-ity of funders to grant a long-term research perspective depends not onlyon large amounts of own resources they must also have the competencesto decide about such investments The change in some continentalEuropean universities to a new public management governance regime hasfor example given them greater operational autonomy which also includesthe use of financial resources This gives them the power to grant suchlong-term support Funding agencies have also recently started to changetheir funding toolbox by developing long-term grants In research systemsthat are built on scarce resources and a lsquoselective modersquo of choosingresearch projects long-term support will have difficulties to be financedSwitchers must in this case live with more uncertainty about the continuingfinancing of their research work and accept higher transaction costs insearching new funding
Another aspect of time resources is the possible substitution of researchtime by other tasks in universities like teaching or administration To havesufficient time for doing research in new areas is therefore also a matter ofhow public research organisations and funding agencies are liberatingresearchers from too many other tasks Funding agencies can develop fund-ing instruments that take this aspect into account and offer a lsquoresearchleaversquo for example Universities must have the flexibility and this isincreasingly the case to redistribute research time among its academicpopulation The stronger role of management and the strategic orientationof universities allow today to give some academic scientists more researchtime while burdening others with more teaching Such flexibility can allowlowering switching costs for scientists
Institutional Conditions and Academic IntegrationThe integration of new areas into established sciences depends to a largedegree on the scientific community When it comes however to the
182 MARTIN BENNINGHOFF ET AL
institutionalisation of new areas in the form of recognised positions in uni-versities (eg professorships) institutional conditions in universities domatter The creation of new professorships research institutes or researchcentres in new cognitive domains can be influenced by the openness of theuniversity towards the introduction of new areas by the influence of stake-holder interests in universities by the existence of a policy of priority set-ting or by the authority of university management to impose solutions inuniversities These conditions can contribute to more flexibility in institu-tionalising new cognitive domains within universities
THE SWISS RESEARCH SYSTEM
We now outline the major features of the organisation of research andresearch funding in Switzerland that help to understand the general institu-tional environment switchers have to work in
According to the OECD (2006 p 8) Switzerland performs very welllsquoin terms of nearly all available indicators of science technology andinnovation often holding with a leading international positionrsquo It is nowonder that the country is generally regarded as a highly attractive systemfor doing research This attractiveness is demonstrated by the fact thatSwitzerland has a high influx of foreign researchers About 50 of profes-sors in 2012 came from a foreign country2 This is not only is a good indi-cator of attractiveness but let us also assume that Switzerland should bevery capable of lsquoimportingrsquo new and flourishing scientific areas relativelyrapidly into the country It then depends on institutional conditionsfor example academic integration or financial support whether there is aspread of the new knowledge
Research overwhelmingly takes place in ten cantonal universities and intwo federal institutes of technology all relatively small in size (Braun ampLeresche 2007)3 Increasingly universities of applied sciences set up in1996 and charged with more applied oriented research are also becomingimportant research places in the system Extra-university research institutesare rare in Switzerland which means that academic integration usuallytakes place in the institutional context of higher education institutions
Switzerland is a federal country This means that both the federalgovernment and a large number of member states the cantons financecantonal universities while it is the federal government alone that pays forthe two federal institutes of technology (Ecole Polytechnique Federale de
183Institutional Conditions and Changing Research
Lausanne (EPFL) Eidgenossische Technische Hochschule Zurich(ETHZ)) During a long process that started in 1988 and lasted until 2009all Swiss universities have acquired operational autonomy and negotiatetheir portfolios of activities in the form of contracts with the politicalauthorities The precise conditions and the degree of operational autonomycan vary substantially between universities (Braun 2012b)
The last decade has been characterised by a period of lsquoaffluencersquoResearch funds have been continually expanded by the federal parliamentand cantonal parliaments These increased funds resulted in considerablegrowth rates for university budgets (Baschung Benninghoff Goastellec ampPerellon 2009) though most notably it was the very well-funded federalinstitutes of technology and the five larger cantonal universities (GenevaZurich Basel Lausanne Bern) that profited most among the 12 highereducation institutions The federal institutes and the larger universitiesare also the main place for scientific discoveries The relative generousfunding has made it possible that even today Swiss researchers can counton substantial institutional and therefore long-term support for doingresearch Institutional funding has not been decreased though there hasbeen a clear tendency and political will to strengthen competition forfunds and financing by research grants from funding agencies (Leporiet al 2007) The binding of about 30 of federal funding to cantonal uni-versities to a number of performance criteria4 has together with a growingcompetitiveness of universities for funding and reputation resulted in pres-sure on scientists to submit projects for research funding Research grantshave in consequence risen considerably during the last 10 years (Braun2012b)
The Swiss National Science Foundation (SNSF) is the only fundingagency for basic research in the country and has the primary role in sup-porting the development of knowledge in universities Relatively generousfunding has enabled the continuation of high success rates in Switzerlandand throughout most disciplines At the end of the 1990s the SNSF devel-oped a major centre programme the so-called lsquoNational CompetenceCentres of Researchrsquo (NCCR) which allows for 12 years of financing incertain areas of research (Braun amp Benninghoff 2003) Because of the largesums of money that were linked to this funding instrument it not onlygave researchers a longer-term perspective for doing research but func-tioned also as a selective device for research within universities Universityleaders had to decide which research grants (and knowledge domains) theywanted to support
184 MARTIN BENNINGHOFF ET AL
CHANGING RESEARCH PRACTICES IN FOUR
SCIENTIFIC DOMAINS IN SWITZERLAND
Research Methods
In order to follow the process of changing research practices in the fourscientific fields we conducted interviews with different type of actors in uni-versities (switchers and middle management) These interviews allowed hav-ing information related to the local context where researchers switch Theinterviews were based on the lsquolife-historyrsquo approach (Glaser amp Laudel2009) One difficulty was to identify individual researchers in the four fieldsThis was done by looking at institutional web sites and by referring to theISI Web of Science In total we conducted 70 interviews 22 for BEC 16for ED 18 for CCL 14 for LSA These interviews were recorded and fullytranscribed In order to preserve the anonymity of interviewees we do notmention any names of researchers
The Diffusion of Four Knowledge Domains in Switzerland
In this section we are interested whether Swiss researchers belonged to thelsquopioneersrsquo in the four cases or whether the knowledge area has been discov-ered elsewhere and Swiss researchers were among the followers to integratethe new field within the academic landscape of Switzerland If the latter isthe case how quick have Swiss scientists been to take up the new areas incomparison to other countries How rapid has then been the spread withinthe country that is the diffusion of the new area in different research insti-tutions and how many scientists have changed research practices to thenew field or in other words what has been the scope of the spread Havethere been differences between the four fields in this respect
Based on our empirical work it becomes clear that all four knowledgefields were diffused in Switzerland though there are differences in pacescope and paths of institutionalisation
ED In Line with International DevelopmentSwitzerland followed the pace of institutionalisation of ED elsewhereDuring the 1990s Swiss researchers participated in the discovery of thegene hox which led to the institutionalisation of experimental ED As at
185Institutional Conditions and Changing Research
the international level work on ED really started during the 2000s whenthe sequencing of different genomes had taken place The ED Swiss com-munity contributed to the further development by own original work (egthe identification of ED model animals) Therefore we can consider Swissscientists as pioneers in ED
An important factor for this role in ED was academic supportSwitzerland has a strong research tradition in molecular biology inSwitzerland since the 1930s and later also in genetics The fact thatthe different research domains that exist today in ED are implantedexactly in those universities which had developed this research tradition inmolecular biology and genetics is a strong indication of the relevance of aresearch tradition for the institutional integration of proximate fields ofknowledge
Looking at the ED switchers in Switzerland one finds three lsquofullrsquo switch-ers and eight lsquopartialrsquo switchers who develop new theories and methodsespecially by using new animal models and techniques coming from genetics(which is unusual in evolutionary biology and extremely costly if animalmodels are used) In comparison to other domains (ecology genetics popu-lation development genetics etc) ED until today represents a rather smallcommunity which is moreover not well connected internally due to theanchoring in different lsquomother disciplinesrsquo
Bose-Einstein Condensate Fast-Tracking the International Development byAttracting Foreign ResearchersSwitzerland has also an excellent tradition in physics which is above alllinked to nuclear physics and solid-state physics while atomic physics andtherefore cold atoms research has traditionally found little attention Thisexplains why BEC has been developed around 2000 through two scientistscoming from abroad (see Laudel et al this volume) While BEC of atomswas experimentally realised in Germany and Netherland and above all inthe USA during the 1990s Switzerland in this case was only a followercountry However a Swiss researcher has been a pioneer in realising BECwith polaritons which he discovered in 2006 A third method of condensa-tion BEC by magnetic systems has also started to be used in the 2000s inSwitzerland
Today researchers in three universities are working with the last twomethods They can be considered to be lsquopartial switchersrsquo that is they arenot investing all their research time to this issue while the two researchersworking on cold atoms in two different universities can be considered to be
186 MARTIN BENNINGHOFF ET AL
lsquofull switchersrsquo who were able to overcome the high costs involved inGermany They were appointed in Switzerland once they had switched
One finds a further research strand in BEC treating the Bose-EinsteinCondensate from the point of view of theoretical physics (which can con-cern both methods presented above) It requires less investment mainlythe cost of computers with considerable calculation powers (and this canbe costly) It also requires good interactions with experimentalists andtime to master the specific physics of this field It is developed in fouruniversities with four scientists participating They can also be consideredas partial switchers with relative low switching costs linked only to thelearning of the methods and the use of highly performance computers forsimulation
CCL On the Fringe of the International DevelopmentAt the international level CCL is not a recent knowledge area (see Engwallet al this volume) It has been built up in different languages over manyyears with a particularly important progress in English language studiesScientists working in the Swiss science system have not been among thecorpus lsquobuildersrsquo and may therefore be considered as being lsquofollowersrsquoRecently however three projects emerged in Switzerland among them alsquodigital dictionaryrsquo for the German languages which has already beenfinished
There are only two Swiss CCL-builders who work almost exclusively inthis domain and have an interest in its theoretical and methodologicaldevelopment Both researchers did however not switch in Switzerland butin Germany Most researchers in Switzerland are lsquousersrsquo who work withexisting corpora in order to answer their various research questions Threeresearchers in three universities were identified doing this In Switzerlandone cannot speak of a Corpus Linguistics community both because of thesevery different ways to deal with CCL but also because of the linguistic frag-mentation of the country which gives room to different epistemologicaltraditions and implies more native language studies (four) than in othercountries
LSA in Educational Research Working through the BacklogFirst steps in LSA were the organisation of national surveys linked to theinternational programmes that existed Switzerland was not among thepioneers in this respect but started to participate driven by scientific andpolitical interest on the national level since the 1990s first in the so-called
187Institutional Conditions and Changing Research
TIMMS (lsquoTrends in International Mathematics and Science Studyrsquo)5 andthen later in the PISA (lsquoProgramme for International Student Assessmentrsquo)studies organised by the OECD TREE (lsquoTransitions from Educationto Employmentrsquo) has been a nationally developed programme in themid-2000s Those who wanted to participate had to coordinate the build-up of own databanks with international databanks (one full switcher) Theother group of scientists wanted to use these databanks Among thesescientists two had to switch which meant largely to learn the different epis-temological tradition of LSA in comparison to epistemological traditionsused in disciplines researchers were anchored in Two others were familiarwith this tradition and therefore only applied the data
ComparisonOur first question focuses on the main dynamics at work in the Swiss scien-tific landscape that is whether scientists working in the Swiss science sys-tems have participated in the set-up of the four areas as lsquopioneersrsquo orwhether they have been lsquofollowersrsquo who adopted changing research prac-tices developed elsewhere Comparing the four fields we have identifiedfour different paths of changing research practices In ED Swiss scientistshave been among pioneers backed up by a long research tradition in therelevant disciplines nurturing ED Switzerland keeps pace with other coun-tries here in the development of the research area even if the scope is notso important compared to other life sciences communities The spread ofBEC became possible because Switzerland was able to attract pioneersfrom another country Pioneering work has however also been done bySwiss researchers in one of the methods applied in BEC Despite someresearch projects CCL remains at the fringe of international developmentin the area Finally LSA researchers were able to keep in line with interna-tional developments in their subject area ED and LSA are the twodomains in Switzerland which were introduced early seen from an interna-tional perspective But if ED has taken a gradual diffusion path in linewith international developments LSA has experienced ups and downs inits development due to scientific and political interests By contrast CCLand BEC were introduced in Switzerland much later than in the pioneeringcountries The role played by foreign researchers has been important forboth innovations However the diffusion paths are different CCL remainson the fringes of international developments while Swiss BEC researchersare becoming internationally competitive and contribute in innovative waysto its further cognitive development
188 MARTIN BENNINGHOFF ET AL
THE INFLUENCE OF INSTITUTIONAL CONDITIONS
ON CHANGING RESEARCH PRACTICES IN THE
SWISS RESEARCH SYSTEM
In the following we endeavour to take stock of the importance of the insti-tutional conditions in Switzerland for the decision of scientists to changeresearch practices
Institutional Conditions and Material and Time Resources
The scientists we interviewed have in general confirmed the high attractive-ness of the Swiss research system in terms of excellent research conditionsabove all in terms of funding grants high salaries and often generous start-up funds These conditions of lsquoaffluencersquo have led many switchers abroadto choose Switzerland as a host country These opinions are in line withexpectations we had about the Swiss research system The influx of foreignresearchers helps Switzerland to excel in areas that were initially not estab-lished BEC is an example though foreign researchers have also played acrucial role in ED and CCL
The attractiveness is linked to the stable and well-financed position ofprofessors in Switzerland both in the two federal and in the five cantonaluniversities examined Such positions allow building up new researchdomains especially if the necessary infrastructure has been included in thestart-up financing Switzerland has also not followed the internationaltrend to submit academics to constant evaluation processes and thereforeto a continuing pressure to publish This gives also more time to invest innew knowledge fields that still need further discovery The obvious attrac-tiveness of tenured positions in Switzerland is given the statements of thesubstantial number of foreign researchers we interviewed the majoradvantage of Switzerland as a research system Both BEC and CCL butalso ED have profited from these attractive conditions by drawing in for-eign researchers Professors profit from excellent working conditions andmoreover from continuing substantial lsquofreedomrsquo in the choice of theirresearch topics which makes lsquovarietyrsquo in the research system possible
Certainly one finds in Switzerland a stronger willingness of the univer-sity management compared to the past to think strategically both in termsof national and international competition and to orient the researchprogrammes within universities partly to pre-defined areas Despite such
189Institutional Conditions and Changing Research
tendencies professors are often hired without insisting on their fit to one ofthese priority areas but purely on the base of lsquoexcellencersquo A university policyopen to lsquovarietyrsquo and this is the result of appointments on the base ofexcellence leads to a patchwork of research fields and disciplines withinuniversities This does not mean that nominations of professors on the baseof strategic considerations are impossible but they seem still to be in theminority compared to nominations on the basis of excellence In the case ofBEC this principle of excellence has allowed the new research field tobecome established within the institutions in question CCL on the otherhand failed to find sufficient support in universities because humanities arerarely among the priority areas in Swiss universities The institutionalisationof ED was favoured by disciplinary traditions within universities that werecognitively near to the ED and which were not greatly altered by changes inthe programmatic orientation of the universities Like CCL LSA did notmeet favourable circumstances in terms of attention by the university man-agement but it was able to find external support that helped to installresearch units within universities
lsquoSeed moneyrsquo to set up new research projects can be important forswitchers to start up exploring new areas In this case we found examplesboth of university leaders demanding that seed money should be distribu-ted on the base of pre-selected research topics in the university and thatonly the proven excellence of the researcher should count
All this confirms that Swiss universities certainly have integrated a morestrategic-oriented policy in employment matters and in the support ofresearch but that a policy of lsquoexcellencersquo which gives scientists a strongfreedom in their choice of research topics still prevails thereby creatingopportunities for new fields to become integrated into universities
A policy to support a lsquovarietyrsquo of research topics instead of limiting vari-ety by selecting research topics on the base of pre-defined research pro-grammes is also supported by the research grant funding system Almostall interviewed scientists mentioned the attractiveness of the SNSF fundingwith its declining but still relatively high success rates that would increasethe chances to find support for research projects in many areas Grants addoften the necessary resources for professors to hire research personnel theycannot finance out of their institutional budgets These positive effects weretestified to above all by researchers in ED and BEC it was possible toobtain SNSF basic grants in new areas even if this has not been the pre-vious research field of the researcher Such generous funding has supporteda strategy of lsquopartial switchingrsquo of researchers who kept on working in theirexisting knowledge fields while being able to acquire supplementary
190 MARTIN BENNINGHOFF ET AL
resources for new research projects that could be invested to explore newareas
Institutional funding and research grants seem to work in a complemen-tary way in Switzerland Institutional funds of universities have played arole as seed money for the exploration of the new domains have helped tobuild research centres or have simply created an institutional space forhosting the research tools (as in the case of LSA and ED) Research grantshave helped to find necessary resources in terms of research personnel andrunning costs of research projects They were also at the base of a strongerconcentration of resources in universities by way of the NCCR grants Thiscomplementarity opened up considerable possibilities for researchers Morecompetitiveness by increasing pressure on researchers to find externalresearch money did not mean moreover that fewer projects were fundedResearchers must undergo the selection process but the chances of beingfunded are higher than in many funding agencies in other countries This isaccording to us very important for explaining the ability to change researchpractices in Switzerland More competitiveness under these conditions doesnot work as a disincentive in this respect
The funding structure in Switzerland has influenced switching strategiesof researchers lsquoPartial switchingrsquo we found has been the dominant switch-ing strategy in all our four cases of changing research practices (Table 1)
Many researchers were able to work in the new fields because they didnot become lsquofull switchersrsquo Staying anchored within the existing researchareas turned often out to be important for getting resources in risky newareas There were positive spill-overs when researchers applied for researchgrants as has been described and often potential switchers got their
Table 1 Numbers of Switchers in Four Cases
Scientific Fields Full
Switchers
Partial
Switchers
Total
Switchersa
Evolutionary developmental biology 3 8 11
Bose-Einstein condensate 2 18 20
Computerised corpus linguistics 0 16 16
Large-scale assessments in education
research
1 4 5
aTotal switchers means the total of researchers that we have interviewed and the not the total
of switchers in Switzerland Concerning the later we do not have data
191Institutional Conditions and Changing Research
employment positions in universities only in the traditional disciplinaryareas Partial switching allows moreover reducing the risks for the scien-tistrsquos reputation as it allows continuing profiting from the established repu-tation as long as the new area does not yet allow similar reputational gainsThe only disadvantage of partial switching though interviewed researchersdid not mention this is probably that time resources that can be invested inthe new areas are more limited in the case of partial switching compared tofull switching Such more limited time resources of switchers can howeverat least partially be compensated by the employment of researchpersonnel
There were only very few lsquofull switchersrsquo (5) in the four knowledge areaswe explored and among them some (2) had already switched abroadPartial switching (which can be either full professors assistant professorsor postdocs) seems therefore to be the main preferred switching strategyunder Swiss conditions of research reducing above all switching costsrelated to reputation resources and time
Especially with reference to BEC and ED one can say that the chancesto find adequate resources in the natural and life sciences both materialand in time were very high in Switzerland According to our investigationsthe situation for CCL and LSA was however somewhat different Weneed however to distinguish here between those who lsquousersquo the databasesthat are the main objects in these two fields and those who are activelybuilding these databases
Being a user in CCL or in LSA does not need full switching and impliesfew resources in both cases Users are contributing to the development ofthe field by working with the research tool and thereby expanding its recog-nition and significance But usually this does not need a particular positionat the university above all it does not need long-term positionsCognitive investments can be kept relatively low partly even very low assome users testified who were well acquainted with the methods that areused in LSA Funding can be acquired by the usual submission for fundingprojects There are therefore no major obstacles in this respect
Investing in building a corpus linguistic or a large assessment databasegenerates in general much higher switching costs in terms of material andtime resources
Large-scale corpus building for example is a longer-term project andneeds secured funding In Switzerland universities usually do not supportsuch an enterprise The SNSF has done it occasionally (a lsquodigital diction-aryrsquo for the German language was funded) but is increasingly reluctant tofund lsquoinfrastructurersquo or research tool projects which are the centre of both
192 MARTIN BENNINGHOFF ET AL
CCL and LSA The only other agency which in this case could compensatethe lack of funding is the Swiss Academy of Humanities and SocialSciences It has however limited resources and must distribute this moneyamong several competing projects As interviewed switchers statedthe money distributed by the Academy does not cover all expenses linkedto such a long-term project In addition to other problems like the var-ious language regions in Switzerland finding research grants is thereforea major problem for corpus-builders in Switzerland One solution forthis would be the financing by the NCCR programme of the SNSF whichis however extremely competitive and has been designed for large networksof researchers and institutions This puts the natural and life sciences inan advantage compared to the social sciences and humanities which aremore based on small group and individual research In the beginning inthe early 2000s one funding round of NCCR projects was reserved to thelatter disciplines but since the mid-2000s all disciplines compete witheach other for NCCR-funds Social science and humanities project are itturned out clearly in the minority In addition it turned out that also uni-versities have been reluctant to support this field Corpus building remainstherefore a precarious undertaking in Switzerland in terms of materialresources
Builders in LSA have on the contrary found more SNSF support Inthis case the SNSF helped to support LSA as programme funding of theSNSF financed the first LSA projects the TIMMS data But it neededalso the support of policy-makers if LSA wanted to be successful above allbecause the LSA data in particular linked to the OECD PISA-programmeneeded data on pupils in schools on a national scale To achieve suchnation-wide coverage was more complicated in Switzerland than in manyother countries because of its decentralised federal structure It took severalyears to coordinate efforts in order to obtain such a nation-wide coveragePolitical support was also important for the financing of the surveys It wasthe Federal Statistical Office which for a couple of years took charge ofthe management of the data before an independent expert committee wasentrusted with this task
Swiss research networks in the field of the LSA are very heterogeneousand include academics and researchers who are still working in or comeoriginally from cantonal or federal offices In addition the integration ofthis research field into the universities was quite different one universityopened an academic position in this field within the Faculty of Educationwhile another one has established a separate organisational unit (evaluationinstitute)
193Institutional Conditions and Changing Research
Funding has been relatively piecemeal and fragmented The variousdatabanks are under different authorities The whole research area has devel-oped in an incremental way and needed a combination of academic entrepre-neurship support of faculties research funding by SNSF programmes andpolitical support by cantonal governments Finally Switzerland has beenable though with some time lag to international developments in this areato set up the research tools in LSA and even to develop a new research toolwith the TREE programme Two universities have been helpful with thecreation of one specialised professorship and one competence centre Thenomination of a professor designed for developing LSA has been possiblebecause the university in question had re-designed its faculty structure sincesome time and had created a number of interdisciplinary faculties Thisallowed the creation of an interdisciplinary education department whichdeveloped an interest in such a professorship
In summary to build the databanks as a research tool in these two caseswas certainly a demanding task for potential switchers with clearly higherswitching costs with regard to material resources than the ones switchers inthe case of BEC and ED had to carry
Another point of discussion that is related to resource needs is thedemand for lsquolong-term supportrsquo for switchers All four cases needed longer-term support in order to be developed In principle such support can comefrom universities funding agencies or stakeholders We observed thatswitching both lsquofullrsquo and lsquopartialrsquo did not occur when the academic posi-tion was not tenured Only being a professor seems to allow researchers totake the risk of entering in still unknown fields The obvious advantage isthe long-term contract and the resources that may be linked to the post ofa professor
Long-term support by the SNSF was for a long time difficult to attainUsually funding projects are around three years Follow-up projects arehowever possible but according to our interviews mostly in the naturalscience division of the SNSF do we find a tradition of supporting researchprojects on a more continuing base This tradition is well known amongnatural scientists who count on this support and often receive it These con-tinuing funds have been advantageous for the development of BECresearch in Switzerland On the other hand it has been much more difficultfor ED to get money for large animal facilities The SNSF has not beenparticularly helpful in this respect Universities provided some resources toset up animal facilities though
The NCCR programme is nowadays the only funding instrument thatcan be used for long-term support for a very limited number of
194 MARTIN BENNINGHOFF ET AL
interdisciplinary projects mostly from the natural and the life sciencesThough none of our cases have obtained a NCCR funding we findresearchers most notably in BEC who participate in NCCRs and cancount therefore on recurrent funding for their projects on BEC up to 12years
We already discussed the difficulty for CCL to build up databases by find-ing adequate long-term funding LSA by contrast has found long-term sup-port for the continuation of the various surveys In the case of TREE theSNSF policy-makers and the host university together finance the surveyAnother university has created as already indicated a professorship forLSA studies and still another was prepared to build an evaluation centre andto coordinate the PISA study which is largely financed out of political funds
Costs Linked to Scientific Integration
We stated in the beginning that it might be difficult for new cognitive areasto find academic integration especially if it contradicts existing paradigmsof research Resistance of established disciplines and academics could beexpected to be the strongest in the case of ED because traditional evolu-tionary views in biology are contested One could expect little resistance inthe case of BEC as there was no contestation of existing paradigms Onthe contrary BEC gave answers to long-asked questions in atomic physicsInterdisciplinarity played no role in this case In the case of CCL and LSAacademic integration depended on epistemological orientations in the disci-plines switchers were originally anchored in like sociology psychology oreconomics If the epistemological tradition corresponded to the ones validin CCL or LSA there was obviously no reason except for struggles aboutthe distribution of finances to resist academic integration
With regard to ED we could not detect on the base of our interviewsovert academic resistance against research done in the ED field but therewere indications of more informal resistance above all from researchers inclassical evolutionary biology who were particularly challenged Theyattempted to prevent the employment of ED researchers The more visibleresistance was by scientific peers during the process of reviewing publica-tions in traditional journals ED switchers had a hard time in getting theirpapers accepted while own journals were still in the making and lackedvisibility
In the case of LSA epistemological confrontation did occur in one casewhere the employment of a new professor in this field caused considerable
195Institutional Conditions and Changing Research
resistance and could only be resolved after considerable time The confron-tation was decided in favour of LSA Two explanations can be stressedFirst there is a social demand for this kind of research that the facultycouldnrsquot ignore second there is also a research demand on the level ofscientific research community in education
We did not find a similar struggle in CCL Also BEC had no difficultiesto be accepted within the universities where it was practiced which is inline with our expectations
CONCLUSIONS
In the conclusions we endeavour to link the insights into the institutionalconditions of switching to the lsquospreadrsquo of the four new scientific fields inSwitzerland before we come back to our general question on the influenceof institutional conditions on changing research practices to what extentdo Swiss institutional conditions have a uniform or diverse influence onchanging research practices What are institutions of support which insti-tutions are detrimental to switching
Institutional Conditions and the Spread of the Four Scientific Fields inSwitzerland
In our case studies we noted that the four fields have followed differentlsquopathsrsquo in becoming part of the scientific community in Switzerland Whatare possible reasons
If we take ED first one can characterise the lsquospreadrsquo as gradual but rela-tively slow though in line with international developments Switzerlandbelonged here to the pioneers The obvious advantage for becoming a pio-neer has been the favourable academic environment in terms of existingresearch traditions that could nurture the interdisciplinary area of ED Thiscondition has facilitated scientists to explore the new field of ED As aninterdisciplinary area it met however also resistance in the academic com-munity when ED tried to find a space of its own within the scientific com-munity Together with the difficulties to obtain long-term support thismight explain somewhat the slow further development of the field inSwitzerland and the relatively small scope in terms of numbers of research-ers active in the area Long-term support exists in Switzerland up to a
196 MARTIN BENNINGHOFF ET AL
certain point by institutional means of universities ED is however an areawith high demands on material and time resources Funding agency sup-port is therefore a necessary condition For a long time however theSNSF has not had instruments to fund with a long-term perspective Thismay have contributed to the slow pace of development
BEC developed rather quickly in Switzerland once foreign pioneersbecame professors in Swiss universities Clearly Switzerland did not haveresearchers of its own that went into this developing area It needed theimport of foreign researchers to set the spread of BEC in motion In thiscase the original discipline atomic physics was not strongly established inSwitzerland This explains among other things that no Swiss researcherswent into the field The attractiveness of the Swiss research place and a pol-icy of excellence the employment of new scientists on the base of theirreputation and not on the fit with existing research lines in the university allowed to draw in foreign researchers who belonged to the pioneers in thisfield Generous financial support helped to develop the area from then onLong-term support was given not only in the ETHZ in this case but also bythe SNSF where we find for the natural sciences a policy of continuous sup-port for once accepted projects though the overall amount of money maydecline in comparison to other branches of the sciences The spread of BECthroughout Switzerland was then rather quick and the country becamequite competitive in the field
In the field of CCL Switzerland has been following pioneering develop-ments in other countries It was rather late in taking up these developmentsand until today the country remains rather at the fringe of internationaldevelopments Possible reasons for the retardation are that Switzerland is acountry with different language regions which has made it more difficult tolink to international developments It needed the lsquoimportrsquo of some foreignresearchers in the German region to give the field an impetus Againattractive working conditions made the lsquoimportrsquo possible Though someprojects were launched the pace of building up a CCL community wasslow Linguistic fragmentation is a further reason for the limited scope ofdevelopments Another important variable in this respect has been thehigh transaction costs for researchers to find institutional supportInfrastructures like the databases in CCL have difficulties in finding fun-ders the SNSF only reluctantly pays such infrastructures Support in uni-versities has been non-existent and other funders did not dispose ofsufficient funding resources
Finally the situation for LSA was not so different from CCL but never-theless LSA found ways to become supported on the long term Swiss
197Institutional Conditions and Changing Research
researchers were not among the pioneers in the area but some were eager tojump on the starting train of LSA in education Temporary initial supportwas given by the SNSF Further support was however more difficultwhich explains the ups and downs in the development of the area The fed-eral structure made it difficult to develop the databases Financial supportwas fragmented and piecemeal political support was important but alsothe preparedness of some universities to host the LSA researchers by givinga position or accepting an institute Flexible arrangements between fundersand the universities made continuing support finally possible TodaySwitzerland can keep in line with international developments and research-ers have some security in remaining funded The difference with CCL hasbeen probably the political interest in the field while CCL depended on thesupport of universities alone and never reached a high status on the prioritylist of universities
Uniform and Diverse Influence of Institutional Conditions
The discussion on the link between institutional conditions and the spreadin the four fields discussed in this article has already given some indicationsconcerning favourable and unfavourable conditions In this section we tryto bring these insights together
It seems to us that a number of institutional conditions have in generalinfluenced positively the decision of scientists to dare a switch to a newscientific field To these conditions belong without a doubt the generalfavourable financial conditions that were present since the end of the 1990sin Switzerland These financial conditions allow offering attractive posi-tions in universities for scientists These positions are internationally com-petitive meaning that they are offering enough advantages to draw in alarge number of scientists The attractiveness as an important variable wasconfirmed throughout our interviews in the four cases When Switzerlanddoes not have own researchers to promote new areas it can without majorproblems lsquoimportrsquo pioneers from abroad and integrate them into the sys-tem This is not yet enough to spread the new field further but it is a begin-ning (eg BEC ED) These attractive positions are found in all the majoruniversities but given the financial advantages of the federal technical insti-tutes and their research profile technical natural and increasingly so lifescientists are particularly favoured in the Swiss context
The employment as a professor in Switzerland gives in general a goodbase for long-term research a condition for developing new fields
198 MARTIN BENNINGHOFF ET AL
especially if the employment is linked to generous infrastructural support forresearch (eg ED) Such support is characteristic of new employments in gen-eral but again it is often more generous in the two federal technical institutes
The shift of universities to a NPM regime has contributed to more flex-ibility in the use of material and time resources of universities and hasallowed to lsquoexperimentrsquo in the case of institutionalisation of new scientificfields into universities (see above all the case of LSA) The policy of excel-lence applied to the nomination of new scientists we found in Swiss universi-ties has given the opportunity to completely new fields not yet anchored inthe universities to become integrated (ie BEC) On the other hand thisnewly won room of manoeuvre to decide on strategies and priorities couldalso have negative effects for switchers as the case of CCL testifies here thedecision of university leaders to value humanities to a lower extent thanother branches of sciences has been a hindrance for the developmentof CCL
A last common trait though not in itself an institutional condition wehave found in the four cases is that lsquopartial switchingrsquo is a frequent strategyof scientists used in all four cases It is questionable if this strategy is theoutcome of institutional conditions or of practices that scientists usuallyapply when considering switching Partial switching is a lsquorationalrsquo strategywe stated above in the sense that scientists avoid taking too much riskswhen switching to a new field We cannot judge whether partial switchingcontributes to a lower pace in the development of new fields as timeresources of switchers to be invested into the new area obviously are morereduced than in the case of full switchers Partial switching could expressthough a deficiency of the research system to create conditions of supportwhich let the scientist trust that she has sufficient time to develop the newarea and built up new reputation In other words long-term support mightbe judged as insufficient by partial switchers6 This point should be clarifiedin future research
In sum the Swiss research system certainly has some institutional condi-tions that have a general positive effect on the change of research practicesThere are however also differences in the working of these institutionalconditions if we take a closer look at the cases One can mention the pointsgiven below
We found a different potential of universities in Switzerland to promoteswitching Not discussing the difference between small and large cantonaluniversities here which has not been part of our investigation one canstate that the federal technical universities have in general more attractivefinancial conditions than the large cantonal universities This helps above
199Institutional Conditions and Changing Research
all the natural and technical sciences which are the main focus of these uni-versities The high standing of these universities in international rankingsdemonstrates also in general the potential for the development of newscientific fields
Infrastructural support is relatively well developed for natural and forlife sciences in universities Continuous support by the SNSF is higher inthe case of the natural sciences The position of the social sciences andhumanities is more problematic in this respect They are subject to highersearch costs when looking for institutional support and must often dependon several funding sources Priority setting in universities had detrimentaleffects especially in the case of CCL
We have found that the academic environment that is existing researchtraditions in universities matter Their presence allows to more easily con-vince scientists to switch in fields that are based on these research traditionsED has been an example while BEC and also CCL with no tradition inbasic supporting disciplines needed to import foreign researchers in orderto develop the new areas even if some ED switchers are also coming fromabroad
This overview demonstrates that some institutional conditions in a coun-try can have an overarching influence that have a positive effect on thechange of research practices There are however also institutional effectsthat lead to different chances for the spread of new scientific fields InSwitzerland a number of institutional conditions are conducive for switch-ing throughout the different branches of the sciences but it has becomeclear that the natural and life sciences are favoured by a number of condi-tions in comparison to our areas in the social sciences and humanities
Having said this we are aware that our study has its limits when gener-alising our findings We have treated only four cases out of a large varietyof scientific disciplines and research domains Nevertheless these caseshave been chosen as examples of the main branches of the lsquosciencesrsquo nat-ural science life science social sciences and humanities The differences wefound in the support of these areas in Switzerland indicate at least that onemight find such differences also for other fields in these branches Futureresearch is needed to confirm this Our study can in general guide suchfuture research by having pointed to the kind of institutional support andinstitutional deficiencies of relevance for changing research practices Theseinsights can be taken up in future research and brought to a test either byreiterating case studies or by developing a more quantitative large-scaledesign that could use our findings as hypotheses of investigation
200 MARTIN BENNINGHOFF ET AL
NOTES
1 Research practices are built on specific sets of concepts and theories and theuse of specific methods and research infrastructure in a cognitively delimitatedknowledge field2 Calculation based on data from the web-site of the Bundesamt fur Statistik
wwwbfsadminch3 The biggest university Zurich has 25000 students4 The main criteria are the number of students and the total amount of SNSF
research projects5 TIMSS was developed by the International Association for the Evaluation of
Educational Achievement (IEA) to allow participating nations to compare studentsrsquoeducational achievement across borders (see httpenwikipediaorgwikiTrends_in_International_Mathematics_and_Science_Study)6 The comparative analysis of ED in this edited volume suggests such a hypoth-
esis (see Laudel et al this volume)
REFERENCES
Baschung L Benninghoff M Goastellec G amp Perellon J (2009) Switzerland Between
cooperation and competition In C Paradeise E Reale I Bleiklie amp E Ferlie (Eds)
University governance Western European comparative perspectives (pp 153175)
Dordrecht the Netherlands Springer
Bonaccorsi A (2007) Explaining poor performance of European science Institutions versus
policies Science and Public Policy 34(5) 303316
Bonaccorsi A Daraio C amp Geuna A (2010) Universities in the new knowledge landscape
Tensions challenges change An introduction Minerva 48(1) 14
Braun D (2012a) Why do scientists migrate A diffusion model Minerva 50(4)
471491
Braun D (2012b) Die Forderung wissenschaftlicher Innovation an Schweizer Universitaten
In T Heinze amp G Krucken (Eds) Die institutionelle Erneuerungsfahigkeit der
Forschung (pp 65100) Berlin Germany Springer
Braun D (2014) University governance and scientific innovation In C Musselin amp
P Teixeira (Eds) Reforming higher education Public policy design and implementation
(pp 145173) Dordrecht the Netherlands Springer
Braun D amp Benninghoff M (2003) Policy learning in Swiss research policy The case of the
national centres of competence in research Research Policy 32(10) 18491863
Braun D amp Leresche J-P (2007) Research and technology policy in Switzerland In
U Kloti P Knoepfel H Kriesi U Linder amp Y Papadopoulos (Eds) Handbook of
Swiss politics (pp 735762) Zurich Switzerland Verlag Neue Zurcher Zeitung
Glaser J amp Laudel G (2009) Identifying individual research trails In B Larsen amp J Leta
(Eds) Proceedings of the 12th International Society for Scientometrics and Informetrics
(ISSIrsquo09) (pp 841845) Rio de Janeiro Brazil
201Institutional Conditions and Changing Research
Heinze T amp Munch R (2010 April 2223) Mechanismen der Erneuerungsfahigkeit
Working Paper Conference Institutionelle Erneuerungsfahigkeit der Forschung
Bamberg Germany University of Bamberg
Kitcher P (1995) The advancement of science Science without legend objectivity without illu-
sions Oxford Oxford University Press
Lepori B van den Besselaar P Dinges M Potı B Reale E Slipersaeligter S Theves J amp
van der Meulen B (2007) Comparing the evolution of national research policies
What patterns of change Science and Public Policy 34(6) 372388
Mirowski P amp Sent E-M (Eds) (2002) Science bought and sold Essays in the economics of
science Chicago IL University of Chicago Press
OECD (2006) OECD reviews of innovation policy Switzerland Paris France OECD
Shi Y (2001) The economics of scientific knowledge Cheltenham UK Edward Elgar
Whitley R (2000) The intellectual and social organization of the sciences Oxford Oxford
University Press
Whitley R (2003) Competition and pluralism in the public sciences The impact of institu-
tional frameworks on the organisation of academic science Research Policy 32(6)
10151029
Whitley R Glaser J amp Engwall L (Eds) (2010) Reconfiguring knowledge production
Changing authority relationships in the sciences and their consequences for intellectual
innovation Oxford University Press Oxford
Ziman J (2002) The microeconomics of academic science In P Mirowski amp E-M Sent
(Eds) Science bought and sold Essays in the economics of science (pp 318340)
Chicago IL University of Chicago Press
202 MARTIN BENNINGHOFF ET AL
COLD ATOMS HOT RESEARCH
HIGH RISKS HIGH REWARDS IN
FIVE DIFFERENT AUTHORITY
STRUCTURES
Grit Laudel Eric Lettkemann Raphael Ramuz
Linda Wedlin and Richard Woolley
ABSTRACT
Bose-Einstein condensation is a scientific innovation in experimental phy-sics whose realisation required considerable time and resources Its diffu-sion varied considerably between and within five countries that werecomparatively studied Differences between countries can be explainedby the variation in the national communitiesrsquo absorptive capacities whilewithin-country differences are due to the impact of authority relationson researchersrsquo opportunities to build protected space for their changeof research practices Beginning experimental research on Bose-Einstein condensation required simultaneous access to the universityinfrastructure for research and to grants The former is largely limited to
Organizational Transformation and Scientific Change The Impact of Institutional
Restructuring on Universities and Intellectual Innovation
Research in the Sociology of Organizations Volume 42 203234
Copyright r 2014 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN 0733-558Xdoi101108S0733-558X20140000042007
203
professors while the latter made researchers vulnerable to the majorityopinion and decision practices of their national scientific community
Keywords Scientific innovation emergence of fields authority rela-tions absorptive capacity experimental physics academic careers
INTRODUCTION
Anyone who wants to introduce a major scientific innovation in experimen-tal physics faces a challenge because designing and constructing a newexperimental system requires considerable time and resources In 1995physicists around the world were facing such a challenge if they wanted torespond to a major scientific breakthrough namely the first production ofa so-called Bose-Einstein condensate of cold atoms A Bose-Einstein con-densate (BEC) is a specific state of matter that occurs when gases of atomsor subatomic particles are cooled to near absolute zero (lt100 Nanokelvin)that is a state of very low energy A large fraction of the atoms collapseinto the lowest quantum state at which point quantum effects occur on amacroscopic scale The occurrence of the phenomenon was theoreticallypredicted by Bose and Einstein in 1924 The first BEC of atom gases wereproduced in 1995 by two US atomic and molecular optics (AMO) groupswhich innovatively combined several recently developed cooling technolo-gies (Cornell amp Wieman 2002 Griffin 2004 Ketterle 2002)
Meanwhile BECs have proven useful in the exploration of a wide rangeof problems in fundamental physics (particularly quantum theory) whichhas led to an explosive growth of experimental and theoretical activitiesPhysicists are using BEC as methods or try to manipulate this new state ofmatter for a wide variety of applications in the more distant future whichinclude atom lasers and quantum computers
There are two reasons why the development of BEC research is ofsociological interest First BEC developed in the national experimentalphysics communities at very different velocities For example it was takenup immediately in 1995 in the Netherlands and Germany while Spain fol-lowed eleven years later The extent to which BEC has become a significantpart of national communityrsquos research activities also varies Researchabout and with BECs has become a substantial and still growing part ofphysics research in Germany while it has disappeared from the agenda inSweden
204 GRIT LAUDEL ET AL
Second experimental BEC research remained very expensive and riskyfor several years until the early 2000s Researchers who wanted to developthe innovation had to make an explicit decision had to invest their wholeresearch capacity in this enterprise and had to control above-averageresources for above-average time horizons This is why BEC research was(and some of its strands still are) very sensitive to variations in the nationaland organisational governance of research
These two aspects are likely to be linked although in ways that arepoorly understood The aim of our article is to answer the question howscientific innovations can be developed by individual researchers on themicro-level under conditions of changing community expectations and indifferent systems of governance and research management With thisanswer we want to bridge the divide between macro-level diffusion studiesof fields and micro-level studies of individual research practices
CONCEPTUAL FRAMEWORK
The development of scientific innovations and of the fields that sometimesemerge with them has interested scholars in the sociology of science for along time After Kuhn (1962) introduced the notion of a paradigm and acorresponding scientific community sociologists became interested in theemergence of new paradigms Classical studies include those by Mullins(1972) on the phage group and the genesis of molecular biology Law(1973) on X-ray Protein Crystallography Mullins (1973) on ethnometho-dology and Edge and Mulkay (1976) on the emergence of radio astronomy(see also Chubin 1976 for a critical review of this research) From a cur-rent perspective many factors one would consider crucial for the emer-gence of a specialty are curiously absent from these studies The impact ofnational science policies and organisational conditions for research on theopportunities for the proponents of new fields to change their researchpractices is not discussed and the social conditions that are discussed(eg information exchange mobility a shared identity access to graduatestudents) were not linked to organisational or policy decisions The sameholds by and large for many of the constructivist studies that include thediffusion of new research practices (eg Cambrosio amp Keating 1995 1998Collins 1998 Fujimura 1988 1992 Pickering 1980 1995 Pinch 1980)
There are several reasons why these studies have neglected many condi-tions for changes of research practices that we would today consider essential
205Cold Atoms Hot Research
Conditions for research have changed considerably In particular tempor-ary positions and competition for grants which make much of currentresearch precarious are relatively recent developments as are the highereducation reforms that increase the power of university management inmany countries (see Whitley this volume) Furthermore the laboratorystudiesrsquo focus on the micro-level made it difficult to observe the impact ofmacrostructures such as institutions (Kleinman 1998 pp 285291 Knorr-Cetina 1995 pp 160163 Mayntz amp Schimank 1998 p 751)
Although these reasons for the neglect of macro-level conditions are lesspersuasive then two or three decades ago more recent studies of the emer-gence of new research fields still struggle with the micro-macro link Theystill seem to address either the macro-level of the diffusion of new findings(eg Fagerberg amp Verspagen 2009 Heinze Heidler Heiberger amp Riebling2013 Raasch Lee Spaeth amp Herstatt 2013) or the micro-level of innova-tors (Mody 2004) Studies on exceptional research (lsquocreative achievementsrsquolsquobreakthroughsrsquo) include organisational factors and sometimes grant fund-ing but were so far unable to establish systematic relationships between spe-cific conditions created by governance and specific kinds of achievements(Heinze Shapira Rogers amp Senker 2009 Hollingsworth 2008)
Studying the micro-macro link requires empirically investigating bothmacro-conditions and micro-level changes establishing how the former aretranslated into conditions for the latter and demonstrating how micro-levelchanges are aggregated In order to establish causal links specific macro-level conditions must be compared with regard to their impact on micro-level changes Although we limit our empirical study to this latter task(leaving the study of aggregation processes to further work) we still need aconceptual framework that solves three problems The framework mustenable a strict comparison of macro-level conditions that is of nationalresearch systems and influences exercised by scientific communities thecomparative investigation of translations of these macro-level conditionsinto conditions for individual researchers who decide to change theirresearch practices and a comparison of conditions for research that explainthe differential success of researchers who want to change their practicesWe use the concept of authority relations for the first and the concept ofprotected space for the second and third tasks
For an integrated assessment of changes in public science systems wedraw on the authority relations perspective (Whitley 2010) This focuseson how different authoritative agencies (the state research organisationsorganisational elites external funding agencies and national as well asinternational scientific elites) exercise authority over specific matters of
206 GRIT LAUDEL ET AL
governance which we can specify for our purposes as research goals Itsbasic assumptions are (a) that the changes that public science systems areundergoing have implications for the relationships between actors andthe way that they are able to realise their interests and (b) that authorityrelations as regards the selection of research goals are the main channelthrough which changes in the knowledge production system are effectuated
Authority over research goals can only be exercised through a fewchannels namely the allocation of resources reputation and career oppor-tunities The relative authority of actors depends on their control of thesechannels which enables the construction of a framework for the integrationof national governance processes into comparable patterns By applying theauthority relations perspective authoritative agencies at the macro-levelbecome linked to meso-level and micro-level actors because authority rela-tions include all actors who exercise authority regardless of the level at whichthey are located
Our framework for comparing micro-level conditions and linking themto authority relations builds on the definition of scientific innovations likeBEC as research findings that affect the research practices of a large num-ber of researchers in one or more fields (ie their choices of problemsmethods or empirical objects) Changing research practices incurs costs andmay be risky in several respects because
they partly devalue the knowledge and equipment a researcher has accu-mulated working on previous topics
a researcherrsquos reputation may suffer if the change requires learning orexperimental redesigns and thereby delays opportunities to publishresults and
the new line of research may deviate from the mainstream of theresearcherrsquos community which again creates the risk of losing reputation
Variations in authority relations affect the creation or diffusion of inno-vations by providing different opportunities for researchers to bear therisks and meet the costs of changes in their research practices (GlaserLaudel amp Lettkemann 2014) We use the concept lsquoprotected spacersquo forcomparing these opportunities as they are provided in our investigatedcountries (Glaser et al 2014 Whitley this volume) We define protectedspace as the autonomous planning horizon for which a researcher can applyhis or her capabilities to a self-assigned task Two dimensions of this vari-able are important here The first dimension is the time horizon for whichthe capabilities are at the sole discretion of the researcher that is the per-iod of time in which the researcher is protected from external interventions
207Cold Atoms Hot Research
into his or her epistemic decisions and external decisions on the use of cap-abilities The resource dimension reflects the research capacity theresearcher controls in this time horizon (personnel over which theresearcher has authority time available for research equipment consum-ables etc)
Researchers create and extend protected space mainly by career deci-sions (the search for positions that provide protected space) and the acqui-sition of funding from various sources including their organisations Thebuilding of protected space links the decisions about research to authorityrelations Applying these concepts to our empirical analysis makes it possi-ble (a) to identify the authority relations the investigated researchers wereembedded in when building their protected space and (b) to assess thescope of protected space that is the numbers of researchers in differentcareer stages whose organisational position makes it possible to build theprotected space that is necessary for a change of research practices (Glaseret al 2014)
METHODS AND DATA
We use data from a larger comparative project that studies the impact ofchanging authority relations in four countries on conditions for intellectualinnovations (RHESI) to which we added a case study about experimentalBEC in Spain Our main focus was on research groups who attempted toproduce BEC of cold atom gases We identified these groups from publica-tions using the keywords lsquoBECrsquo or lsquoBose-Einstein Condensationrsquo frominternet searches of experimental physics groups at universities and fromlsquosnowballingrsquo that is by asking interviewees about their national commu-nities Our attempt to include groups that were prevented from conductingBEC research by authority relations despite their interest failed becausesuch cases are almost impossible to identify empirically
Table 1 provides an overview of the interviews For all countries exceptGermany whose BEC community is too large by now we interviewedresearchers from all atomic physics groups that conducted BEC research(usually the group leaders and in some cases also group members) OneDutch group leader and two German group leaders declined to beinterviewed However there is only one German group about which wehave little information because no former group members could be inter-viewed In order to get a better picture of the structure of relevant physics
208 GRIT LAUDEL ET AL
Table 1 Overview of Conducted Interviews
Netherlands Germany Switzerland Sweden Spain
AMO BEC groups 5 (7 interviews) 8 2 (7 interviews) 1 (2 interviews) 1 (2 interviews)
Other BEC groups 3 (6 Interviews) Other physics groups 2 experimental
AMO physics
5 (8 interviews with
BEC theoreticians
1 from another physics
field)
3 (1 BEC theoretician
1 experimental AMO
physicist 1 other field)
Other informants 2 officers of funding
agency
1 officer of
funding agency
1 former funding
advisor to
ministry
Total number of
interviews
11 9 21 5 3
209
Cold
Atoms
HotResea
rch
communities we conducted additional interviews with other physicists Wealso interviewed officers of funding agencies
The interviews were conducted as semi-structured face-to-face inter-views1 A shared interview guide was used for interviews with BEC research-ers in order to ensure comparability of data Interviews with researchersconsisted of two main parts In the first part the intervieweersquos research wasdiscussed We explored the development of the intervieweersquos research sincethe PhD project with an emphasis on thematic changes and the reasons forthem In this part of the interview developments in the intervieweersquosnational and international communities were also discussed The discussionof the research content was prepared by analysing documents includinginternet sites Nobel lectures (Cornell amp Wieman 2002 Ketterle 2002Phillips 1998) and publications at various levels of sophistication frompopular science to an intervieweersquos research publications The preparationalso included a bibliometric analysis of the intervieweersquos publications forthematic links publications A visualisation of this publication network wasused to prompt narratives about the content of the research at the beginningof the interview (Glaser amp Laudel 2009a see also Laudel et al thisvolume)
In a second part of the interview conditions of research and the factorsinfluencing them were discussed Topics included the knowledge personnelequipment required to produce BECs source of material support andopportunities as well as constraints provided by the intervieweersquos academicposts The interviews lasted 60120 minutes All but one were recordedand fully transcribed
The analysis of interviews focused on the variables of the theoreticalframework The comparison of cases from the five countries is based on anassessment of the necessary protected space for early BEC experimentswhich could be derived from the interviews We then reconstructed animportant macro-level condition for all researchers who attempted tochange their research practices namely the international diffusion of BECresearch in the contexts of opinions and preferences in the internationaland national scientific communities For each attempt to begin the experi-mental production of BECs the building of protected space the authorita-tive agencies involved and the consequences of the exercise of authoritywere determined and compared Reasons for delayed and failed attempts toproduce BECs were also traced back to authority relations and interests ofthe actors involved Our comparison of changes of research practices at theresearcher level follows the distinction between supported delayed andprevented cases because this distinction emphasises the impact of authority
210 GRIT LAUDEL ET AL
relations The empirical analysis is followed by a generalising discussion ofthe macro-micro link and conclusions
EPISTEMIC PROPERTIES OF EXPERIMENTAL BEC
RESEARCH AND ITS REQUIRED PROTECTED SPACE
From the first attempts until the early 2000s manufacturing BECs of atomswas an exceptionally complex risky and expensive undertaking even bystandards of the wider field in which it is located experimental low tem-perature physics Manufacturing a BEC was strategically uncertain becauseit was not clear that the effect could be experimentally produced and tech-nically uncertain because it was not clear how by what specific experimen-tal setting the effect could be produced (on these notions of strategic andtechnical uncertainty see Laudel and Glaser 2014) The strategic uncer-tainty concerned the question whether any gas of cold atoms would staygaseous at the low temperatures and relatively high densities required forBEC It was quite plausible that all atom gases except hydrogen wouldbecome liquids or solids if cooled so far This question stood anew for eachnew kind of atom for which BEC was attempted replications of suchexperiments were the only exception The technical uncertainty was highfor all researchers who tried to produce a BEC for the first time Setting upthe experimental system required a researcher to go through a longsequence of steps of adjusting and fine-tuning the equipment In manycases parts of the equipment were built to order by technical workshopsThis is why for a long time (at least until the early 2000s) building andfine-tuning a BEC experiment took several years It was always possiblethat the researcher could not solve the technical problems involved inwhich case the experiment failed Although the technology for BEC pro-duction has advanced during the last two decades setting up a BEC experi-ment for the first time remains a risky and demanding endeavour for aresearch group
These epistemic properties of BEC research correspond with a large pro-tected space in terms of resources and long and often unpredictable timehorizons Achieving BEC in atomic gases required the combination of themost advanced techniques for cooling atoms and trapping those with thelowest energy The research involves complex task-specific equipmentwhich is usually built from components by the researcher Depending onthe research prior to the move to BEC several of the more expensive
211Cold Atoms Hot Research
components might already exist in the laboratory The equipment for aBEC experiment could cost 100000500000 Euros depending on whatwas available in the laboratory Consumables (mostly very expensivecooling liquids) caused additional recurrent costs At least two full-timeresearchers (almost always PhD students) were needed to build and adjustthe experimental setup parallel work of more PhD students or postdocswould be an advantage to accommodate the technological uncertainty
Owing to the inherent uncertainties the time horizon of the experiment(from setting up the experimental system to publishing the results) mayextend beyond the usual three-year grant cycle The reputational riskinvolved is high because the experiments can fail entirely and because littlecan be published until the experiment is successful
INTERNATIONAL AND NATIONAL DYNAMICS
OF BEC RESEARCH
The development and diffusion of experimental BEC can roughly bedivided in four phases Attempts to produce BECs began in the 1980s aftera new cooling method (laser cooling) was developed This first phase endswith the first successes in summer 1995 For about three years afterwardsresponses to this success were mixed because the experiments were too diffi-cult to replicate and the potential of BECs for further research remainedunclear With the first successful replications and BECs of other atomsBEC research began to grow in the AMO community after 1998 Since theearly 2000s the manufacture and use of simple BECs became routinisedand BECs began to be used as a method in other areas of physics
Phase 1 First Attempts to Manufacture BECs (1980s to Summer 1995)
Since it had always been clear that the experimental realisation of BEC inatom gases depended on achieving extremely low temperatures experi-ments aimed at producing BECs seriously began only after a new coolingtechnique laser cooling had been developed Still the majority of thescientific community believed that producing BECs was impossible becausethe atom gases would turn into liquids or solids when cooled down to thetemperatures necessary for BEC Some believed hydrogen to be the onlypossible exception which is why major experimental efforts began in the
212 GRIT LAUDEL ET AL
1980s when a few condensed matter physics groups tried to produce BECin spin-polarised hydrogen gas by combining several cryogenic methods Atthe beginning of the 1990s they were considered the leading experts in BECresearch However most AMO physicists doubted that a breakthroughcould be achieved in the near future A different route towards BEC waspursued by a small minority of researchers in the US who began to coolatom gases of alkalis This idea was met with even stronger scepticism thanthe hydrogen route
Bose-Einstein condensation was a vision of the past [early 1990s] and many people did
not believe in it They said this is not possible This will not work for all sorts of rea-
sons Just before the first Bose-Einstein condensate was produced in 1995 there were
predictions that it is not possible (German BEC researcher)
Of course all people who worked with cold atoms wrote in their applications about the
lsquoHoly Grailrsquo that this will at some time lead to BEC But this was so far away for
everybody that we thought that we will never achieve it Just imagine you always work
with buckets of water on a great plain and there are puddles everywhere and suddenly
somebody sets out to fill an ocean It is clear that it becomes more when I have more
buckets but it is unimaginable that it will work in the foreseeable future (German BEC
researcher)
In the early 1990s only two of the five national physics communitiesincluded in our study featured strong AMO physics communities TheGerman and Dutch communities had made major contributions to laser-cooling techniques Both national communities shared the internationalmajority opinion namely that BEC cannot be experimentally achievedNone of the German AMO groups pursued BEC at that time Researcherswho did consider attempts to produce BEC felt disheartened by the vastadvantage of the US groups in both resources and experience Since thenew method of laser cooling had created many new research opportunitiesthere was no reason to pursue the lsquoholy grailrsquo of BEC whose realisationwas doubtful anyway
In the Netherlands the strong tradition of low temperature physics and astrong AMO physics provided a supportive background for BEC researchOne of the Dutch research groups worked on BEC in atomic hydrogen TheDutch group leader belonged to the pioneers that paved the way towardsthe first BEC His major scientific achievements and being backed by thehydrogen community probably helped to continue this research despite theAMO communityrsquos doubts Another Dutch researcher became interested inBEC in the early 1990s but could not begin for lack of funding
The situation in the three other countries was quite different Switzerlandand Spain had no tradition in AMO physics and were unaffected by the
213Cold Atoms Hot Research
experimental development Switzerland has a strong tradition in nuclearand condensed matter physics instead while Spainrsquos experimental physicshas been dominated by nuclear and particle physics which had little episte-mic connections to AMO physics cold atoms and BEC research Swedenhas a long tradition in the field of AMO physics with many strong experi-mental groups However most research concentrated on molecular spectro-scopy building on the tradition of Nobel Prize winners Manne Siegbahn(1924) and Kai Siegbahn (1981) This narrow focus was criticised by aninternational evaluation committee in 1992
lsquoSweden for many years has been strong in Atomic Molecular and Optical Physics
almost all the work is of high quality and some is outstandingrsquo but also recommends
lsquo there should be less emphasis on traditional molecular spectroscopy and more on
newer fields of atom trapping laser cooling laser optics etcrsquo (Bradshaw et al 1992
pp 1617)
The fields whose strengthening was recommended were those on whichexperimental BEC built
Phase 2 Responses to the First Experimental Success(Summer 19951997)
In the summer of 1995 first empirical evidence of BEC was presented at aninternational physics conference at Capri Until the end of the year threeUS research groups were successful in producing a BEC from alkali gasesThis was immediately regarded as an outstanding contribution by AMOphysicists and by the wider physics community However the internationalAMO community was undecided whether these achievements marked theend of the long quest for the lsquoHoly Grailrsquo of BEC or the beginning of anew research area Would it open up opportunities for interesting new phy-sics or was it just the experimental confirmation of a theoretical predictionIn the year following the Capri conference it was unclear what directionresearch would take
Most German researchers tended to share the general beliefs of the inter-national community it was not clear whether BEC would indeed open anew research direction
Q [After the 1995 Capri conference] was the entire German community of the same
opinion
A (Laughs out loudly) It was very mixed Some [people] just went quiet and reverent
and saw that this is the Holy Grail But I would say that what you can do with it that
214 GRIT LAUDEL ET AL
it opens up an entirely new parameter space and that you can make a completely new
physics was recognized only by few back then (German BEC researcher)
Secondly the problem of the competitive advantage of the US groupshad not vanished Thirdly other AMO fields promised better careerchances Against this general trend in the German AMO physics commu-nity three groups (two professors one junior group leader) began BECresearch after the first experimental success was announced at the Capriconference in mid-1995 Although the community approved of grants onlyreluctantly all three researchers could immediately begin with the first stepsof building the experiments which included simulations and invitations tothe innovators from the United States in order to learn the tricks of thetrade from them The first German BEC (and first BEC outside the UnitedStates) was achieved in 1997
In the Netherlands the researcher who had originally worked on BECin hydrogen switched to the alkali route because the technology hadbecome superior to the complicated cryogenic approach for BEC in hydro-gen Four more researchers became interested in pursuing BEC research inalkalis in this phase However only one professor and a researcher whosemove was tolerated by his professor could start in that period Tworesearchersrsquo change of research practices was delayed by authority rela-tions as we will explain in more detail in the following section
Phase 3 Growth of BEC Research within AMO Physics (Since 1998)
The year 1998 witnessed new BECs being produced in many countries Theresearch now moved beyond the replication of the original results becauseit became obvious that BECs provided many opportunities for interestingtheoretical and experimental research Since then more than a hundredresearch groups worldwide have produced BECs BECs of atoms of otherelements photons and molecules have been produced The technologiesfor producing BECs have been improved which made it possible to useBECs as tools in several other research areas The growth of BEC researchwas thus accompanied by an internal differentiation
Again the German AMO physics communityrsquos attitude towards BECparalleled that of the international community The great potential of BECexperiments became widely accepted More groups began BEC researchand the experiments of all German BEC groups moved beyond the replica-tion of original experiments in new directions Today about 15 GermanAMO physics groups work on BEC
215Cold Atoms Hot Research
The new attitude of the German AMO community was also reflected inthe decisions on project funding Grants for BEC research were approvedwithout problems and the delays in producing BECs were tacitly acceptedby approving grants for researchers who had already had a grant but werenot successful within the first three years
In the Netherlands the growth period of BEC in AMO physics wasbriefly reflected in a dedicated grant programme for the support of BECresearch Two further Dutch atomic physics groups began their BECresearch Different from Germany the communityrsquos attitude to BEC wasless favourable
But the situation in Holland was not very favourable [several years] after the first BEC
was observed Some people thought why are you going to do this now Other coun-
tries are ahead of us why should we do that There was one experiment in Amsterdam
where they achieved Bose Einstein condensation and some thought well perhaps this
is sufficient why do you need four groups (Dutch BEC researcher)
Different to their German counterparts the Dutch funding agencybecame impatient soon and downsized the funding programme when noBECs were produced after three years Further grant funding was concen-trated on two groups one of which already had produced a BEC in 1999The funding problems caused delays and forced one group to give up theirattempts altogether In the end three more groups achieved BEC between2004 and 2006 Currently four groups continue BEC research
From 1998 onwards BEC research began in the three other investigatedcountries too In 1998 a young Swedish researcher on a 4-year fellowshipposition began BEC research which he intended to conduct parallel toanother line of research that was also based on laser cooling In spite of theresearcherrsquos appointment as professor at a Swedish university his group didnot succeed due to funding difficulties In 2009 the group leader left thecountry without having achieved BEC and the Swedish group dissolved
The Swiss physics community caught up with BEC research by recruitinglsquoready-madersquo BEC researchers This was not completely premeditated Theuniversities recruited whom they considered as the best candidates in quan-tum physics who happened to be BEC researchers However the decisionwas likely to be influenced by the high potential of the BEC field In 2000the first AMO physics groups began to work on BEC a second group fol-lowed recently Both groups expanded rapidly with several parallel BECexperiments with at least one group belonging to the international elite
The situation in Spain was unusual in that the first BEC experiment inSpain was initiated by a researcher from a different community namely the
216 GRIT LAUDEL ET AL
theoretical physics community of non-linear optics He had conductedtheoretical BEC research since the late 1990s and had made important con-tributions to the field But it was only after he became a full professor thathe considered the move towards experimental BEC and he started in2006 Taking into account his limited access to resources he wanted to pro-duce what could be considered a lsquostandard BECrsquo at that time (a BEC ofRubidium atoms) However this intention was not enthusiastically met bythe national community which doubted the value of a Spanish lsquosimplersquoBEC given how far other groups worldwide had moved ahead The com-munityrsquos prevailing attitude was that it would be better lsquovalue for moneyrsquoif he continued to produce theory papers Being fully autonomous as auniversity professor the researcher tried nevertheless but failed because hedid not receive grant funding
Phase 4 Utilisation of BEC in Other Areas of Physics(Since the Early 2000s)
With BEC increasingly proving its usefulness as an instrument for funda-mental research questions in physics and beginning to show remote applica-tion opportunities the production of BECs became interesting for otherareas of physics such as condensed matter physics In Germany severalgroups worked on BEC topics related to condensed matter physics
This diffusion of BEC in other fields of physics is most pronounced inSwitzerland where the strong condensed matter community turned to BECresearch Around 2001 a condensed matter physicist began BEC researchbecause he wanted to add a BEC research line to his research portfolioDue to the high technical uncertainty it took around seven years to achievesuccess which was immediately recognised as a major achievement by theinternational community Two more condensed matter physicists workedon BEC with completely different approaches In one case BEC was aserendipitous discovery while the other researcher is only marginally inter-ested in BEC and has not yet produced a BEC after eight years
Fig 1 summarises the diffusion dynamics of BEC research in cold atomsin the five countries we studied The Netherlands and Germany had startedearly on with BEC research Switzerland and Sweden followed relativelylate and Spain only recently started Germany has the strongest BECcommunity now with several groups belonging to the international elite TheNetherlandrsquos BEC community consists of four groups that are internationallyrecognised Switzerland has one group belonging to the international elite
217Cold Atoms Hot Research
and another strong group2 Several groups could not immediately start BECresearch (the delays are marked grey)
HOW VARIOUS SETS OF AUTHORITIES ALLOW
HINDER PREVENT BEC RESEARCH
Having established interests and influences of international and nationalscientific communities we now turn to a second set of macro-level condi-tions and their link to changes of research practices We begin at the micro-level of individual changes by asking how changes of research practices wereaccomplished A first comparison across all countries reveals that a largenumber of these changes were supported by the set of authority relations inwhich they were made while other attempts to produce BECs were delayedor even prevented by the exercise of authority as summarised in Table 2
The dynamics of the changes in research practices vary considerablywithin countries The shortest time between the start of BEC research andsuccessful publication is one year (for the researcher in Switzerland whohad successfully produced BECs before) the longest time is ten years (for aDutch researcher whose move to BEC was delayed by authority relations)
Fig 1 Diffusion Pattern of BEC Research in Five Countries
218 GRIT LAUDEL ET AL
The delays and failed attempts certainly were also influenced by the cap-abilities of researchers which we could not reliably compare across allcases in five countries (see Glaser amp Laudel 2009b on that problem)However we can trace the reasons for delays and failures back to particu-lar patterns of authority relations which systematically differed from thoseof the supported cases
Since the variance in dynamics is bigger within countries than betweencountries we begin our analysis by comparing supported delayed andprevented cases Findings on countries will be synthesised in a subsequent step
Authority Relations Supporting BEC Research
The ideal-typical situation of a researcher whose move to BEC researchwas supported by authority relations was somebody who controlled anadaptable infrastructure and had access to external grants in a way that
Table 2 Impact of Authority Relations on Changes of ResearchPractices
Impact of Authority Relations on Changes of Research
Practices
Supported Delayed Prevented
Occurrences Germany
Netherlands
Switzerland
Germany
Netherlands
Netherlands
Sweden
Spain
Time from first intention to begin
of work (years)
Immediate start Germany 27
Netherlands 36
Immediate start
Time from begin of work to
achieving BEC (years)
Germany 27
Netherlands 3
Switzerland less
than 1
Germany 2
Netherlands 49
Time from begin of work to
publication of results
Germany 38
Netherlands 4
Switzerland 1
Germany 38
Netherlands 610
219Cold Atoms Hot Research
supported the unpredictable time horizons Researchers in Germany theNetherlands and Switzerland found themselves in that situation They heldprofessorial positions which granted them discretion over the necessaryinfrastructure (laser equipment) and some personnel Building or changingthe infrastructure for BEC research was supported by resources providedby the university which were granted as start-up packages on appointmentas professor or as loyalty packages if a professor received an invitation towork elsewhere but stayed at the university Many professors already hadsubstantial parts of the necessary equipment (eg the lasers) and thus oftenneeded only relatively small amounts of additional funding of ca h100000Technical support provided by the research organisations was crucial forbuilding the experimental setups In terms of the required knowledge theAMO physics groups were either themselves familiar with laser optics andcooling technologies or they hired postdocs who had learned it in the lead-ing laboratories abroad
With the exception of one German professor whose start-up packagewas so generous that he could work without external grants for severalyears the money for additional equipment and personnel came from exter-nal grants For that (personnel) and for specific equipment additionalresources were needed This made professors dependent on their scientificcommunities whose attitudes towards BEC and grants for it changed overtime As we saw in the previous section the majority opinion in scientificcommunities was not in favour of BEC in the first two phases The Germancommunity would nevertheless provide grants albeit reluctantly so
The application was in June rsquo95 In May I believe was the [Capri] meeting In June I
sent the application out for review This review process was stopped by the DFG []
The approval of my application is well I would have to look it up but I think that it
lasted almost two years It was approved when I achieved the BEC (laughs) Maybe it
was just one or one and a half years or something like that So it was extremely tough
They posed further questions they did not answer for months and it was terrible
(German BEC researcher)
The Dutch community did not approve of any grants for BEC exceptthose for the early innovator until the late 1990s and cut funding soonagain in the early 2000s (see Glaser et al 2014 for a comparison ofGerman and Dutch decision practices on BEC grants) This attitude con-tributed to several delayed cases (see the following section)
For professors who held non-BEC grants their communitiesrsquo reluctancedid not matter due to another practice namely the complete transfer ofauthority over the use of the money to researchers once the grant wasapproved lsquoBootleggingrsquo the use of the grants for different purposes was
220 GRIT LAUDEL ET AL
tacitly approved by both the German and the Dutch funding agencies Thisled to the paradoxical situation that a community did not explicitly approveof grants for BEC but let researchers use grants on BEC that were approvedfor other purposes
Another tacit practice helped adjust the grant funding to the long andoften unpredictable time horizons of BEC research Grants were usuallygiven for three (Germany) or four years (Netherlands) while producing aBEC could take much longer The scientific community responded to thisdiscrepancy by awarding new grants regardless of the experimental successof previous ones
Well I must say that we have always been supported by the Deutsche
Forschungsgemeinschaft especially with these high-risk projects So in the case of BEC
which as I said took seven years you could have said many times lsquothatrsquos itrsquo and lsquothere
will never be resultsrsquo Nevertheless we have always been successful in writing applica-
tions (German BEC researcher)
The same pattern occurred in Switzerland in the 2000s where a con-densed matter group received consecutive grants for its non-atomic BECand succeeded after seven years
The ideal-typical pattern thus consists of a professor who controls alaboratory infrastructure and some personnel has access to technical work-shops and can utilise grants for a change of research practices There weretwo deviations from this pattern which nevertheless still included authorityrelations that made the change of research practices possible
Several researchers started BEC on non-professorial positions In thesecases the necessary access to infrastructure and personnel was not givenwhich made the researcher dependent on others German and Dutchresearchers below the professorial level could start their BEC work becausetheir professors (or directors of institutes) approved and granted themaccess to their infrastructure Researchers whose professors did not grantthat access were delayed in their move (see the following section) In onecase this access was granted by a Dutch faculty which wanted to compen-sate for the lack of grants for BEC
The other deviation from the ideal-typical pattern is that of the twoSwiss researchers who produced BECs of atoms Both researchers suc-ceeded in a very short time due to prior experience and above-averagefunding They both had produced their first BEC abroad and had to re-create experimental setups they were already familiar with when theymoved to Switzerland Both researchers received considerable start-upfunding from their university departments for equipment and several PhD
221Cold Atoms Hot Research
and postdoctoral positions In addition one of them was permitted by hisformer lab leader to take the equipment for his previous BEC experimentwith him Both researchers extended their protected space quickly by exter-nal grants from the Swiss National Science Foundation (SNF)
The ideal-typical pattern and the deviations from it make the authorityrelations that supported the move to BEC research clearly identifiableResearchers had to have authority over infrastructure and personnel whichwas granted by their universities to professors They also needed grantswhich the scientific community was initially reluctant to give directly Thisexercise of authority concerning dedicated BEC grants was compensatedfor by the transfer of authority over the use of grants to researchers Thusthe transfer of authority over resource by both universities and scientificcommunities enabled the building of the large protected spaces
Authority Relations Delaying BEC Research
In several Dutch and German cases authority relations delayed changes ofresearch practices because researchers could not build the necessary pro-tected space Either of the two processes identified above the transfer ofauthority over infrastructure from universities to professors and the transferof authority over grants from scientific communities (through fundingagencies) to researchers could be absent or halted If this was the casesome researchers waited for authority relations to improve which causedthe delays
The most frequent reason for delays was the lack of access to infrastruc-ture Two Dutch researchers on tenured but non-professorial positions hadto postpone their BEC research in the late 1990s because their professorswere not interested in BEC research and therefore would not lsquolendrsquo theirinfrastructure for this topic This lack of opportunities to build protectedspace contributed to their fears of not being able to compete with thegroups in the US Thus they only started around five years later with BECresearch Two German researchers were also delayed by missing access toinfrastructure One of them had to wait for a professorial position whilethe other had become appointed but did not receive the start-up packagefor several years He had successfully negotiated when he was recruited butdid not receive the money due to financial difficulties of his university
Missing project funding was a reason for delays in the NetherlandsThe main funding agency for physics research Stichting voor Fundamenteel
222 GRIT LAUDEL ET AL
Onderzoek der Materie (FOM) was initially reluctant to fund BECresearch In the first phase prior to the first experimental success onlyone researcher received grants for research on BEC in hydrogen Anotherresearcher who wanted to take up the idea of BEC in alkalis before thesuccess in the United States did not receive grants and had to postpone thestart of BEC research
At the end of the 1990s Dutch researchers joined forces and wrote abottom-up funding proposal dedicated to BEC research This proposal wasreviewed and had to be revised which delayed the funding programmeuntil 2000 This funding programme gave researchers who wanted to pro-duce BECs access to dedicated grants The programme was evaluated afterthree years and was stopped because no further BECs were achieved afterthe first success in 1999
A And then there was the [hellip] programme on cold atoms starting somewhere in 2000
and that lasted only three years and then it was stopped by FOM because they thought
there was not enough progress
Q In form of publications
A In the form of Bose Einstein condensates Because you have to realize it started in
2000 and then we of course promised Bose Einstein condensate here in Holland and
two years later there were still no Bose-Einstein condensates And then the funding
agency said we stop the programme (Dutch BEC researcher)
The funding agency FOM established a second funding programme forBEC but gave all funding of this programme to the group of the researcherwho already had produced a BEC and to a junior research group leader inhis department These groups were the only two whose research was nothindered by insufficient funding The other groups faced shortages Twogroups continued their BEC research but could not set up parallel experi-ments which was common in BEC research As a result both groupsrsquo suc-cesses were considerably delayed A third group had to give up BECresearch entirely (see below)
The cases of delayed success in producing BECs confirm the necessaryauthority relations derived from the cases that were not delayed Research-ers that were successful in the end were delayed because they had to waiteither for a position that gave them authority over the infrastructure andpersonnel provided by the university or for grants that provided the neces-sary complementary funding The following cases of unsuccessful experi-mental BEC demonstrate what happens if the two conditions are neversimultaneously fulfilled
223Cold Atoms Hot Research
Authority Relations Preventing BEC Research
A Dutch a Swedish and a Spanish group failed in their attempts toproduce a BEC Having tried for seven to eight years these groups had notproduced BECs and two abandoned experimental BEC research Whileepistemic reasons may have contributed to these failures the authority rela-tions in which these researchers worked also deviated considerably fromthose characterising successful cases The three cases have in common thatat some point the continuation of BEC research of all three groupsdepended on experimental success that is on an externally enforced timehorizon of their protected space that was too short
A Dutch group was confronted by the termination of the BEC fundingprogramme by FOM after three years (see the preceding section) The uni-versity did not compensate for the exclusion of the group from grant fund-ing and the group could not obtain enough grants to continue BECresearch by bootlegging money
A Swedish researcher on a 4-year fellowship had obtained knowledgeabout laser-cooling experiments in one of the leading US laboratories andmanaged to secure a grant from a major foundation for setting up a laser-cooling laboratory at his home university The start-up process was slowbecause neither the fellowship nor the infrastructure grants allowed for anyfunding of additional research positions The faculty did not allocate PhDpositions very likely because he was not an established researcher on apermanent position yet BEC research also was new to the physics depart-mentrsquos research agenda The researcher could later secure two PhD posi-tions for his laboratory which both were jointly funded by his departmentand by external grants Three years later in 2001 the researcher wasappointed as tenured associate professor at another Swedish universityThe recruitment came with no start-up funding and further funding forPhD students or equipment remained problematic for the group
The group felt there was a lack of support from the university for theBEC research Support further declined when a new dean of the depart-ment was appointed There was a lack of understanding for the technicaluncertainties of BEC research and disappointment about insufficient publi-cations Eventually the group leader managed to secure funds from smalllocal Swedish funding agencies to fund doctoral students for his laboratoryStill the group had only minimum resources in an environment where theycould not collaboratively use the infrastructure of others While fundingfor equipment did not seem to be a major problem the grants left littleroom for the experimental failures that were unavoidable under conditions
224 GRIT LAUDEL ET AL
of high technical uncertainty Technical support by workshops which hadbeen crucial for the German and Dutch BEC groups was almost non-existent The group could not recruit experienced postdocs from otherlaboratories One PhD student travelled to one of the well-known laser-cooling labs in the United States to obtain additional knowledgeComparing the conditions in the US lab with what was available in hishome setting he said that his group had to work lsquowith duct tape andhome-made solutionsrsquo to make the experiment work BEC was neverachieved In 2009 the group leader moved abroad to take up a professor-ship and the only attempt of establishing experimental BEC research inSweden ended
In Spain a theoretician who was internationally recognised for his con-tributions to theoretical BEC research became interested in experimentalBEC which he started when he obtained a professorial position in 2006Although the Spanish physics community did not support his idea of pro-ducing a standard BEC he managed to obtain some resources for theexperimental work A PhD student on a 4-year scholarship provided by theministry was awarded through a general PhD programme and not specifi-cally dedicated to BEC research but could be oriented this way Theresearcher also received two small 2-year grants from the national govern-ment and regional government (together around h60000) The researcherfurther bootlegged funding from grants for his theoretical research to buyequipment This enabled the beginning of building the experimental setupalbeit on a shoestring budget He got some ad-hoc assistance from hiscolleagues-experimentalists in the department There was no technical sup-port from the university at all Through visits of leading European labora-tories he and his PhD student obtained the necessary knowledge about thecooling techniques of BEC experiments The visits were made possible byhis reputation for theoretical BEC work
The reputational risk of this attempt at experimental BEC was reducedsince the researcher continued his theoretical BEC research and was thusable to continually publish However the time horizon of protected spacewas limited to two years because the Ministry made funding conditional onproving experimental success
Yes we finished the MOT [magneto-optical trap] and just on time In the Ministry
they said okay if you donrsquot have the MOT by this date you will not have more money
So we were really desperate and I remember the day when I was with this experiment you know in these experiments you have lots of things to tune - and I was there and I
was very impassioned with a detector we had there quite slow 100 times more and we
saw a flash there and okay and I said to [the PhD student] have you seen that and she
225Cold Atoms Hot Research
said yes and we were very delicately moving and then that was when we got it And yes
it was a really happy day One week before the deadline (Spanish researcher)
The group succeeded only with the first experimental steps but not withproducing a BEC Subsequent evaluations by members of the national phy-sics community questioned the scientific significance of the standard BECFinally the national funding agency rejected further grant applications fortwo successive years
If we look at the authority relations in the three cases of failed attemptsthe contingencies of BEC research become apparent Researchers simulta-neously need access to infrastructure and grant money for a sufficiently longtime horizon which means that they depend on the university having theseresources and granting them and on the approval of their research by theirscientific community In the Netherlands the funding agency decided not toprovide grants for BEC research to particular groups anymore because thetime horizons were at odds with common expectations In Sweden thelimitations of grants were not overcome by support from the universitypartly because university funding was limited and partly because theresearcherrsquos colleagues also expected quick results In the Spanish case BECresearch appeared somewhat like a suicide mission a scientific communitywith strong misgivings funding agencies controlled by the government withlittle understanding for the protected space needed Only the flexible use offunding made it possible to start the experimental BEC work at all
MACRO-MICRO LINKS IN THE DEVELOPMENT OF
SCIENTIFIC INNOVATIONS
Having identified the processes by which international and national scienti-fic communities developed the innovation and the role of authority relationsin the building of protected space for changing research practices at theindividual level we can now return to our question about the link betweenmacro-level authority relations and individual changes of research practicesWe focus our discussion on the exercise of authority by the actors involved
Absorptive Capacity of Scientific Communities and Exercise of Authority
Scientific communities exercise authority through their scientific elite whichcontrol the grant funding that has become a necessary source of protected
226 GRIT LAUDEL ET AL
space in experimental research In the case of basic research the state and(in Sweden) private foundations supplied these resources but transferredtheir authority over them almost completely to the scientific elites Theirdecisions in funding agencies which are macro-level actors bypassed allother actors including universities and their sub-units and directly shapedresearchersrsquo opportunities to build protected space This is why the commu-nitiesrsquo attitudes towards BEC research and to the use of grants for it werecrucial for micro-level changes of research practices (Table 3)
Our reconstruction of the emergence and growth of BEC research in theinternational scientific community and in five national scientific commu-nities shows that the idea of prior research creating absorptive capacitywhich has been introduced by Cohen and Levinthal (1990) to organisa-tional sociology is also applicable to scientific communities and theirresearch traditions We found four distinct situations each of which cre-ated specific responses to the innovations Germany and the Netherlandshad strong research traditions in the field in which the innovation wascreated and featured research groups that contributed to the groundworkfor the experimental success This could be expected from the largeGerman science system which is likely to host most research traditions
Table 3 Absorptive Capacity and Attitudes towards the Use of Grantsfor BEC in Five Communities
Germany Netherlands Switzerland Sweden Spain
Absorptive
capacity
High High Low (high in
condensed
matter
physics)
Low Low
Acceptance of
proposals that
contradict majority
opinion
Yes with
some
reluctance
Only for one
member of
international elite
Not
observed
Not observed No
Acceptance of time
horizons exceeding
terms of grants
Yes Only for one
member of
international elite
Yes Not observed No
Transfer of
authority over use
of grants to
researchers
Yes Yes Not
observed
Limited Yes
227Cold Atoms Hot Research
in some form The absorptive capacity of the Dutch community was his-torically contingent The Netherlands had a strong research tradition inlow temperature physics that dates back to the 19th Century which pro-duced many important contributions to cooling techniques
The example of Sweden highlights the contingent nature of such tradi-tions Sweden featured a strong AMO physics tradition which howeverwas narrowly focused on the legacy of two Swedish Nobel laureates to theexclusion of those subfields of AMO physics that provided the absorptivecapacity for BEC research Switzerland and Spain did not even have AMOphysics communities
Consequently the German and Dutch research groups were best posi-tioned to develop the innovation and responded first by producing BECsThe Swiss response is yet another illustration of the importance of absorp-tive capacity Apart from the recruitment of two researchers who producedBECs of cold atoms several initiatives to produce BECs were developed inthe condensed matter physics community which is a strong Swiss researchtradition Having no absorptive capacity Swedish and Spanish physics hadto rely on individual activists who both can be said to have failed due tothe wider physics communityrsquos lack of understanding for the intricacies ofBEC research particularly the technical uncertainties and the resultingunpredictable time horizons
The degree to which the communityrsquos majority opinion actually mat-tered depended on the decision style of grant funding processes (Glaseret al 2014) Of the three communities for which this could be observedbecause proposals that were at odds with the majority opinion were sub-mitted the German community exhibited a pluralistic attitude and (albeitreluctantly) funded proposals the majority did not consider worth fundingThe Dutch and the Spanish community did not with the Dutch communitynot believing in the possibility of BEC in alkalis (first phase) and in thescientific merits of continuing with BEC after the first experimental success(second phase) The Spanish community did not believe in the merits of aleading theoretician venturing into experimental BECs
The communityrsquos response to the unpredictable and often long timehorizons of BEC research is a third aspect of the exercise of authority thatsignificantly affected researchersrsquo opportunities to build protected spacewith grant funding The German and Swiss communities accepted this pro-blem and responded by approving new grants despite the previous ones notbeing successful The response of the Dutch community is rather surprisingbecause AMO physicists would know the problem (and have it experiencedthemselves) However the decision was made at the level of funding
228 GRIT LAUDEL ET AL
programmes with the wider physics community being included in a deci-sion about fields to which the money should go In Sweden the attitudetowards time horizons was not observed for the grant funding process butwas clearly visible in the critical attitude of the BEC researcherrsquos faculty con-cerning delayed publications In the Spanish case the impatience manifesteditself in intermediate lsquomilestonesrsquo for setting up the experiment that seem tohave generated at the ministerial as well as the community side of funding
The opportunities to build protected space were finally affected by theextent to which authority over the use of grants was transferred toresearchers once the grants were approved German Dutch and Spanishresearchers were able to overcome other obstacles set by their communitiesor by administrative delays in the grant funding process by lsquobootleggingrsquomoney from grants they received for other purposes This opportunity gavea clear advantage to lsquowealthyrsquo researchers that is to those who were suc-cessful in acquiring many grants In the Swiss case the transfer of authoritywas not observed because bootlegging was not necessary The same appliesto the Swedish case although some of the grants the BEC researcherreceived carried strong limitations (they could be used only for equipment)
Thus although scientific communities used only one channel to exerciseauthority over BEC research this channel turns out to be quite complexThe approval of grants and expectations concerning their use and outcomescarry the use of both specific authority concerning the content of research(in our case BEC research and its time horizons) and unspecific expecta-tions that include the extent to which authority over grant use is transferredto grant holders
The Translation of Authority Relations in Micro-Level Conditions forChanging Research Practices
The second essential source of protected space for researchers was thefunding provided by their research organisations (mostly universities for acomparison of universities and public research institutes see Glaser et althis volume) The authority exercised through this channel also created acomplex pattern that varied between countries The state was the ultimatesource of all resources for BEC research (a Swedish private foundation beingthe only exception) but transferred its authority over all these resources toother actors
Authority over resources for research is transferred to research councilsand universities in all five investigated countries The degree to which
229Cold Atoms Hot Research
authority is transferred varies and has been greatly increased with recenthigher education reforms in the Netherlands Germany Sweden andSwitzerland While the use of resources within universities was previouslyprescribed in great detail universities now have more authority over theuse of their resources
In all five countries control over the expensive infrastructure for experi-mental research was concentrated on professors This situation occurredregardless of the progress of higher education reforms Even in the Dutchsystem whose career structure and university governance was reformed inorder to reduce authority of professors the latterrsquos control over infrastruc-ture and PhD students remains This is partly due to scarcity Universitiesdo not have enough resources to provide all their academics with infra-structure and if a selective approach is necessary anywhere concentrationon the most highly reputed academics suggests itself
In university systems that are further reformed professors share theirauthority with their parent faculty which has the authority to allocateadditional resources or positions for PhD students We observed this splitauthority in the Netherlands and Sweden In Germany the authority overinfrastructure is still transferred exclusively to professors while the author-ity of the faculty was not relevant to the Swiss and Spanish cases
This made professors the gatekeepers of the BEC innovation in all fourcountries While the grant allocation process was more diverse and also pro-vided researchers below the professorial level with resources for BECresearch the second essential source for protected space the universityrsquosinfrastructure could only be utilised by professors or those whose researchprojects professors approved of This is why fellowship positions rarely pro-vided enough protected space for a change of research practices towardsBEC Even generously funded fellowships depend on pre-existing infrastruc-ture which made fellows dependent on those who control it The very fewexceptions from this pattern we observed confirm rather than challenge itThey include one decision of a Dutch faculty to provide a non-professorialresearcher with resources for infrastructure and a German researcher whohad accumulated grants that might have sufficed for a change of researchpractices (which was not tested because he was appointed professor soonthereafter)
CONCLUSIONS
The experimental manufacture of BEC constituted an extreme case becauseit required protected space that is large both in the resource and the time
230 GRIT LAUDEL ET AL
dimension from the beginning of the 1990s at least until the beginning ofthe 2000s This made it a very good test case for the observation and analy-sis of authority relations Our analysis enables some general conclusionsabout the impact of authority relations on researchersrsquo opportunitiesto change their research practices for following or generating scientificinnovations
A first important conclusion concerns the role of epistemic traditions forcreating absorptive capacity in scientific communities For scientific innova-tions to be taken up and developed in a country this country not onlyneeds interested researchers but also a basis in the form of prior knowledgeand techniques While larger countries can be expected to have this basisfor most innovations its existence in smaller countries depends on histori-cal developments and path dependencies This is why for many countriesthe opportunity to immediately respond to a scientific innovation is contin-gent on its research traditions
A second conclusion concerns the impact of authority relations onresearchersrsquo opportunities to build protected space If we start from theassumption that any researcher in any position might want to change theirresearch practices in order to either create an innovation or to develop aninnovation that has been created elsewhere we can assess how authorityrelations reduce this space
If building the necessary protected space depends on access to the infra-structure provided by universities the opportunity to create or develop aninnovation is limited to those who control it that is to professors and thosewhose research is approved by professors This dependence is underminedwhere faculties have the right to allocate infrastructure However thismoves a researcherrsquos dependence from individual professors to local organi-sational elites
If building protected space additionally depends on external grants theopportunity to create or develop an innovation is thematically limited toinnovations that are considered worth pursuing by a scientific communityOnly if a scientific community transfers the authority over this particulardecision to the researchers by not making the majority opinion a bindingfoundation for the approval of grants researchers can work against themainstream It was interesting to notice that all investigated communitiesmitigate some of the consequences of their decisions by transferring far-reaching autonomy over the use of grants once they have been awarded
These dependencies prevented some researchers from changing theirresearch practices while others experienced considerable delays It is nowpossible to assess the impact of some of the larger longer-term trends ofchanging authority relations in the public sciences (see Whitley this
231Cold Atoms Hot Research
volume) The transition from predominantly recurrent funding to a splitsystem in which recurrent funding must be supplemented by competitivegrant funding has ambivalent effects It increases the number of researcherswho can build protected space by including staff below the professoriallevel At the same time it makes all researchers dependent on the authorityof their national and international scientific elites which can cause consid-erable delays and may prevent some innovations altogether Higher educa-tion reforms and the transfers of authority it involves do not affect thelimitation of innovation opportunities to those who control the infrastruc-ture (ie professors) but may create additional opportunities for non-professorial staff by transferring authority over resources to faculties Thefunding of fellowships and temporary research groups is not sufficient forinnovations requiring large protected space but may be beneficial for lessdemanding innovations
Finally the increasing state interest in research serving societal goals didnot play a direct role for the purely basic BEC research However asLaudel and Weyer (this volume) suggest the total protected space for basicresearch might shrink due to funding problems and the increasing incor-poration of state priorities in science policies
NOTES
1 Telephone interviews were conducted with three of the researchers in theSwedish case2 The research of Swiss condensed matter physicists who work on BEC cannot
be compared to the other cases in terms of authority relations and protected spacewhich is why we didnrsquot include them in Fig 1 and Table 2
ACKNOWLEDGEMENT
We are grateful to Jochen Glaser whose suggestions have greatly improveda previous draft
REFERENCES
Bradshaw A M Cabbibo N Dalgarn A Kienle P Ramsey N Wilkins J amp Eberson
L (1992) International evaluation of Swedish research in physics Report to the research
council Stockholm Sweden Naturvetenskapliga forskningsradet (NFR)
232 GRIT LAUDEL ET AL
Cambrosio A amp Keating P (1995) Exquisite specificity The monoclonal antibody revolution
New York NY Oxford University Press
Cambrosio A amp Keating P (1998) Monoclonal antibodies From local to extended net-
works In A Thakray (Ed) Private science Biotechnology and the rise of the molecular
science (pp 165181) Philadelphia PA University of Pennsylvania Press
Chubin D E (1976) The conceptualization of scientific specialties Sociological Quarterly 17
448476
Cohen W M amp Levinthal D A (1990) Absorptive capacity A new perspective on learning
and innovation Administrative Science Quarterly 35 128152
Collins H M (1998) The meaning of data Open and closed evidential cultures in the search
for gravitational waves American Journal of Sociology 104 293338
Cornell E A amp Wieman C E (2002) Nobel lecture Bose-Einstein condensation in a dilute
gas the first 70 years and some recent experiments Reviews of Modern Physics 74
875893
Edge D amp Mulkay M J (1976) Astronomy transformed The emergence of radio astronomy
in Britain New York NY Wiley
Fagerberg J amp Verspagen B (2009) Innovation studies The emerging structure of a new
scientific field Research Policy 38 218233
Fujimura J H (1988) The molecular biological bandwagon in cancer research Where social
worlds meet Social Problems 35 261283
Fujimura J H (1992) Crafting science Standardized packages boundary objects and ldquotrans-
lationrdquo In A Pickering (Ed) Science as practice and culture (pp 168211) Chicago
IL The University of Chicago Press
Glaser J amp Laudel G (2009a) Identifying individual research trails In B Larsen amp J Leta
(Eds) Proceedings of the 12th International Society for Scientometrics and Informetrics
(ISSIrsquo09) Rio de Janeiro Brazil (pp 841845)
Glaser J amp Laudel G (2009b) On interviewing ldquogoodrdquo and ldquobadrdquo experts In A Bogner
B Littig amp W Menz (Eds) Interviewing experts (pp 117138) London Palgrave
MacMillan
Glaser J Laudel G amp Lettkemann E (2014) Hidden in plain sight The impact of everyday
governance on the emergence of fields In M Merz amp P Sormani (Eds) The local con-
figuration of new research fields On regional and national diversity Sociology of the
Sciences Yearbook Dordrecht the Netherlands Springer
Griffin A (2004) The first BEC conference in Levico in 1993 Journal of Physics B Atomic
Molecular and Optical Physics 37 Retrieved from httpiopscienceioporg0953-4075
377E02 Accessed on February 15 2014
Heinze T Heidler R Heiberger R H amp Riebling J (2013) New patterns of scientific
growth How research expanded after the invention of scanning tunneling microscopy
and the discovery of buckminsterfullerenes Journal of the American Society for
Information Science and Technology 64 829843
Heinze T Shapira P Rogers J D amp Senker J M (2009) Organizational and institutional
influences on creativity in scientific research Research Policy 38 610623
Hollingsworth J R (2008) Scientific discoveries An institutionalist and path-dependent per-
spective In C Hannaway (Ed) Biomedicine in the twentieth century Practices policies
and politics (pp 317353) Bethesda MD National Institutes of Health
Ketterle W (2002) Nobel lecture When atoms behave as waves Bose-Einstein condensation
and the atom laser Reviews of Modern Physics 74 11311151
Kleinman D L (1998) Untangling context Understanding a university laboratory in the
commercial world Science Technology amp Human Values 23 285314
233Cold Atoms Hot Research
Knorr-Cetina K (1995) Laboratory studies The cultural approach to the study of science In
S Jasanoff G E Markle J C Petersen amp T Pinch (Eds) Handbook of science and
technology studies (pp 140166) London Sage
Kuhn T (1962) The structure of scientific revolutions Chicago IL The University of Chicago
Press
Laudel G amp Glaser J (in press) Beyond breakthrough research Epistemic properties of
research and their consequences for research funding Research Policy Retrieved from
httpdxdoiorg101016jrespol201402006
Law J (1973) The development of specialties in science The case of X-ray protein crystallo-
graphy Science Studies 3 275303
Mayntz R amp Schimank U (1998) Linking theory and practice Introduction Research
Policy 27 747755
Mody C M (2004) How probe microscopists became nanotechnologists In D Baird
A Nordmann amp J Schummer (Eds) Discovering the nanoscale (pp 119133)
Amsterdam the Netherlands IOS Press
Mullins N C (1972) The development of a scientific specialty The phage group and the ori-
gins of molecular biology Minerva 10 5182
Mullins N C (1973) The development of specialties in social science The case of ethno-
methodology Social Studies of Science 3 245273
Phillips W D (1998) Laser cooling and trapping of neutral atoms Reviews of Modern
Physics 70 721741
Pickering A (1980) The role of interests in high-energy physics The choice between charm
and colour In K D Knorr R Krohn amp R Whitley (Eds) The social process of scien-
tific investigation (pp 107138) Dordrecht the Netherlands Reidel
Pickering A (1995) The mangle of practice Time agency and science Chicago IL The
University of Chicago Press
Pinch T J (1980) Theoreticians and the production of experimental anomaly The case of
solar neutrinos In K D Knorr R Krohn amp R Whitley (Eds) The social process of
scientific investigation (pp 77106) Dordrecht the Netherlands Reidel
Raasch C Lee V Spaeth S amp Herstatt C (2013) The rise and fall of interdisciplinary
research The case of open source innovation Research Policy 42 11381151
Whitley R (2010) Reconfiguring the public sciences The impact of governance changes
on authority and innovation in public science systems In R Whitley J Glaser amp
L Engwall (Eds) Reconfiguring knowledge production Changing authority relationships
in the sciences and their consequences for intellectual innovation (pp 349) Oxford
Oxford University Press
234 GRIT LAUDEL ET AL
HIGHLY ADAPTABLE BUT NOT
INVULNERABLE NECESSARY AND
FACILITATING CONDITIONS FOR
RESEARCH IN EVOLUTIONARY
DEVELOPMENTAL BIOLOGY
Grit Laudel Martin Benninghoff
Eric Lettkemann and Elias Hakansson
ABSTRACT
Evolutionary developmental biology is a highly variable scientific innova-tion because researchers can adapt their involvement in the innovation tothe opportunities provided by their environment On the basis of com-parative case studies in four countries we link epistemic properties ofresearch tasks to three types of necessary protected space and identifythe necessary and facilitating conditions for building them We foundthat the variability of research tasks made contributing to evolutionarydevelopmental biology possible under most sets of authority relationsHowever even the least demanding research depends on its acceptanceas legitimate innovation by the scientific community and of purely basic
Organizational Transformation and Scientific Change The Impact of Institutional
Restructuring on Universities and Intellectual Innovation
Research in the Sociology of Organizations Volume 42 235265
Copyright r 2014 by Emerald Group Publishing Limited
All rights of reproduction in any form reserved
ISSN 0733-558Xdoi101108S0733-558X20140000042008
235
research by state policy and research organisations The latter conditionis shown to become precarious
Keywords Scientific innovation emergence of fields authority rela-tions life sciences evolutionary developmental biology
INTRODUCTION
In this article we look at the development of a scientific innovation that inmany respects epitomises the varied nature of scientific change Differentfrom innovations that emerged from specific experiments or methodologi-cal developments (see eg Laudel et al in this volume on Bose-Einsteincondensation) evolutionary developmental biology (evo-devo) has a muchlonger trajectory with more diffuse roots It can be traced back to the endof the 1970s when it became more and more obvious that neo-Darwiniantheory was unable to account for all empirical findings of evolutionarybiology (Muller 2007) particularly the rapid changes in the forms oforganisms evident from the fossil record and the origins of traits that didnot constitute an adaptation to the environment
It became increasingly obvious that these explanatory deficits of neo-Darwinism were
due to its treatment of development as a lsquoblack boxrsquo and the consequent absence of
the generative rules that relate between genotype and phenotype (Muller 2007
pp 500501)
The discovery of genes regulating embryonic development in the 1980s(HOX genes) and advances in molecular and genomic techniques made itpossible to address specific questions by comparing the development of dif-ferent organisms which led to increased understanding of developmentalmechanisms on the molecular level (Gerson 2007 Raff 2000 Muller2008) This research recently received a further impulse by a breakthroughin the development of sequencing technologies which made the sequencingof whole genomes affordable for single research groups
These conceptual and methodological developments affect a variety offields Responses to the new ideas range from purely conceptual develop-ments to complicated experiments Research in evo-devo can be conductedin many ways and thus can adapt to specific organisational conditions andauthority relations However some features of different approaches to evo-devo make them sensitive to organisational conditions which is why theydo not thrive in all organisations and countries
236 GRIT LAUDEL ET AL
Evo-devo research can be taken up by researchers from either develop-mental or evolutionary biology fields by integrating the complementaryperspective and can be either theoretical or experimental Experimentalevo-devo research is always comparative but varies in the types of organ-isms that are used and in the ways in which the empirical evidence aboutthe organisms to be compared is acquired Most of the research comparestwo or more species but some scientists compare transgenic organismsbelonging to the same species The organisms studied can be classicalmodel organisms (such as mouse Drosophila or Arabidopsis) or they canbe selected because they are best suited to answer specific evo-devo ques-tions (eg hedgehogs or snakes) The comparisons required by evo-devoresearch can also take a variety of forms Many researchers experimentwith organisms from one species and acquire information about the organ-ism(s) they use for comparisons either from the literature or by collaborat-ing with colleagues who investigate the other organisms Others experimentwith more than one organism Most of these possibilities occur in threemajor fields in which evo-devo perspectives have taken hold namely zool-ogy plant biology and palaeobiology
The resulting combinatorial complexity of intellectual transition pathsto evo-devo and their varied intellectual and resource costs for researchersmake evo-devo both a very interesting and a very challenging case forsociologists for studying the impact of authority relations on the develop-ment of innovations Evo-devo is a very interesting object of study becauseresearchers can adapt their involvement in the innovation to the opportu-nities provided by their environment They can temporarily or permanentlychoose degrees of involvement whose costs match the niches provided forthem by the various interacting organisations At the same time this flex-ibility makes it much more difficult to attribute variations in the develop-ment of evo-devo research to specific authority relations it is not only thatthey enable or prevent evo-devo research but they also enable or preventcertain kinds of evo-devo research
In this article we take up this challenge by identifying the conditionsunder which specific research problems of evo-devo biology can be formu-lated and solved This amounts to formulating a lsquopopulation ecology ofresearch tasksrsquo that in many ways resembles the population ecologyapproach in organisational sociology (Hannan amp Freeman 1977 1989) Atthe same time we follow Hodgson (2013) who resolved the argument aboutselection versus strategic adaptation that has arisen in the context of popu-lation ecology in exactly the same way as it is done by our empirical objectie by applying an evolutionary developmental perspective We combine an
237Highly Adaptable but Not Invulnerable
evolutionary approach that asks which niches enable support or preventspecific kinds of research tasks with a developmental approach that askshow researchers together with other authoritative agencies co-created thesevery niches and adapted their research tasks to them
Analysing the ecology of evo-devo research tasks involves a causal argu-ment consisting of three steps which we present after introducing the theo-retical background and the methodological approach of our research Firstwe analyse the epistemic characteristics of different kinds of evo-devoresearch and identify the kinds of protected space that scientists need forengaging in the various lines of this research Second we identify the neces-sary and facilitating conditions for building these kinds of protected spaceand the authoritative agencies controlling them This analysis enables athird step in which we ascertain how researchers could build protectedspace in different countries
THEORETICAL BACKGROUND
The emergence of fields (or scientific specialties as they were called at thattime) enjoyed much attention in the sociology of science during the late1960s and 1970s1 Inspired by Kuhn analysts attempted to identify rela-tionships between cognitive dynamics and social patterns in the early stagesof development Focusing on the link between the dynamics of ideas andpatterns of interaction the studies of emerging scientific specialties did notpay much attention to the conditions under which the research was con-ducted and the opportunities for changing research practices provided bythese conditions In 1976 Edge and Mulkay produced a synopsis of studiesof emerging specialties that identified three common features of specialtyemergence the growth of the specialties from innovations at the margins ofestablished disciplines the mobility of researchers and ready access of theproponents of the new specialties to graduate students (Edge amp Mulkay1976)
Evo-devo does not though fit these patterns because research in thisarea grows on top of rather than beside existing fields The differentiationpattern observed in the cases studied in the 1960s and 1970s was one oflsquobranchingrsquo according to which new research areas emerged around newsets of problems research technologies or empirical objects (Mulkay1975) By contrast evo-devo adds a new layer of empirical research and anew frame of reference in which experiments can be designed and data
238 GRIT LAUDEL ET AL
compared This means that moves of researchers from old to emergentfields can be partial and may be reversed
The early studies of emerging fields also had a rather narrow focus inthat they did not link the observed social conditions for emergence to orga-nisational or policy decisions This might have been due to the differentsituations of scientists at the time the fields emerged particularly regardingaccess to research funds and tenured posts Most of the researchers in thesestudies were already on permanent contracts had relatively easy access toresources and were able to move between universities as opportunitiesarose Once established in tenured posts they were much less dependent onauthorities (organisational managers disciplinary elites and external agen-cies) than are their colleagues today
More recent science policy studies that look at the shaping of emergentfields (such as nanotechnology) by governance structures typically take thereverse perspective They look almost exclusively at policy measures chan-nelling resources to emerging fields and trying to influence the directions oftheir development (see Glaser et al 2014 for a critique) Scientific innova-tions that do not attract political attention eg due to lack of potential forapplications and the processes through which governance structures andprocesses affect the development of fields do not tend to be the focus ofthese analyses
Investigating the evolution of evo-devo research links the problemchoices of researchers to the niches created by contrasting governance struc-tures in different countries We consider evo-devo as a scientific innovationwhich we define as a research finding that affects research practices (choicesof problems methods or empirical objects) of a large number of researchersin one or more fields2 Changing research practices incurs costs and mayberisky eg by devaluing knowledge equipment and reputation accumulatedwith previous research Our comparative framework focuses on differencesin authority relations (Whitley 2010) as the key mediating factor connect-ing governance structures to changes of research practices and uses theconcept of protected space (Glaser et al 2014 Whitley this volume) tocompare the opportunities for researchers to change their practices underthe specific conditions created by different sets of authority relations
The state research organisations external funding agencies and scientificelites exercise authority over research choices through three main channelscontrol of resources the allocation of reputation and the provision ofcareer opportunities For these channels the relative authority of each setof actors can be assessed and a framework for the comparative analysis ofeveryday governance of research activities formed accordingly This
239Highly Adaptable but Not Invulnerable
framework can be linked to the micro-level of changing research practicesby identifying the authority relations that conditioned how researchersattempted to move to evo-devo research
The impact of particular sets of authority relations on the opportunitiesfor changing research practices towards evo-devo can be ascertained bycomparing the level of protected space that is required for different kindsof changes with the level that can be built by researchers in different sets ofauthority relations We define protected space as the autonomous planninghorizon for which a researcher can apply his or her capabilities to a self-assigned task The two critical dimensions of this variable are the time hori-zon for which the capabilities are at the sole discretion of the researcherand the resources (including personnel over which the researcher hasauthority and the actual time available for research) Researchers createand extend protected space mainly by career decisions (the search forpositions that provide protected space) negotiations with managers of theirresearch organisation and the acquisition of funding Important dimen-sions of this last variable are the likelihood of success in receiving fundingand the speed with which such funding can be obtained
METHODS AND DATA
We use data from the larger comparative project that studied the impact ofchanging authority relations in four countries on conditions for intellectualinnovations that is summarised in the editorial introduction Throughsearches in publication databases (Web of Science Google Scholar) andInternet webpages we detected research groups who had included evo-devointo their research portfolio We also asked our interviewees which otherevo-devo researchers in their national community they were aware of InSweden Switzerland and the Netherlands most evo-devo researchers couldbe interviewed In Germany with its large evo-devo community onlyselected cases could be investigated Researchers who presented evo-devoas a research focus on their websites were selected taking into accountthe variation of disciplines (zoology and plant biology) and of types ofresearch organisations (university and public research institute) Additionalinformation was obtained by interviewing experts in the field and heads offaculties and institutes The distribution of our interviewees is summarisedin Table 1 The table also lists the situations we investigated as casesnamely situations in which researchers successfully or unsuccessfullyattempted to build protected space Not all of our interviewees undertook
240 GRIT LAUDEL ET AL
these attempts some of them were postdocs whose research was determinedby the decisions of their group leaders Furthermore we consulted self-descriptions of the evo-devo community on the development of the field(eg Carroll 2006 Hall 2012 Laubichler amp Maienschein 2007 Minelli2008 and the authors quoted above)
The interviews with researchers consisted of two main parts In the firstpart the intervieweersquos research that contained the innovation was discussedin the context of the intervieweersquos research projects exploring the continu-ity and all thematic changes and reasons for them Parallel developmentsin the intervieweersquos national and international communities were alsodiscussed The discussion of the content of scientistsrsquo research required thedevelopment of the interviewersrsquo knowledge to an lsquoadvanced laypersonrsquosrsquolevel and the negotiation of a level of communication at the beginning ofthe interview (Laudel amp Glaser 2007 see also Collins amp Evans 2002 onthe level of expertise necessary for competent interaction) Therefore it wasprepared with Internet searches and publications at various levels of diffi-culty (from popular science up to an intervieweersquos publications) were usedInterview preparation also included a bibliometric analysis of the intervie-weersquos publications that enables the identification of thematically linkedpublications A visualisation of this publication network (see Fig 1) wasused to lsquostimulate the recallrsquo (Dempsey 2010) and to prompt narrativesabout the content of research (Glaser amp Laudel 2009)
In a second part of the interview conditions of research and the factorsinfluencing them were discussed Topics included the knowledge personnel
Table 1 Overview of Interviews and Cases
Germany Netherlands Sweden Switzerland
Researchers 7 group leaders
1 PhD student
11 group leaders
2 PhD students
3 group leaders
2 postdocs
7 group leaders
5 postdocs
Other informants 3a 4b
Total number of
interviews
8 16 5 16
Number of transition
situations
12 13 7 9
aOne researcher who gave background information on the evo-devo field one director of an
institute one officer of a funding agencybOne evo-devo background three heads of institutes or deans of faculties
241Highly Adaptable but Not Invulnerable
and equipment required to conduct evo-devo research sources of materialsupport and opportunities as well as constraints provided by the intervie-weersquos organisational positions The separation of the discussions of theseconditions of research from the content of research is important because itlimits the extent to which interviewees present their own subjective theoriesabout how current funding conditions made them conduct their currentresearch The interviews lasted 60120 minutes They were recorded andfully transcribed
For the comparison we developed typologies for evo-devo practicestransition situations and protected space In order to deal analytically withthe task of identifying influences of authority relations on intellectual tran-sition patterns we reduced the complexity of these patterns in two subse-quent steps First we used the combinations of variables with the strongestinfluence on transition costs to identify ten common transition situations tostart evo-devo In a second step we allocated these situations to differentcategories on the basis of the level of protected space they required
lsquoTHE SNAKE TOOK US ABOUT THREE YEARSrsquo TRANSITIONS TO EVO-DEVO
In order to identify the impact of authority relations on transitions toevo-devo we must first establish what kind of transitions could take place
Fig 1 Example for a Research Trail of an Evo-Devo Researcher (the circles are
publications the size of the circles indicates the number of citations the lines show
thematic connections between publications)
242 GRIT LAUDEL ET AL
and what protected space these transitions required The difficulty costand risks of a move to evo-devo depended on the epistemic properties ofthe move which included properties of the research task chosen by aresearcher and the researcherrsquos disciplinary background We derive the pro-tected spaces necessary for a transition in two steps First we identify theproperties of transitions to evo-devo that affected the necessary protectedspace Second we distinguish between large medium and small levels ofprotected space that the transitions required
Properties of Transitions
Four properties of transitions to evo-devo research that affected the neces-sary protected space included (a) the empirical strategy (b) researchersrsquo ori-ginal disciplinary background (c) the types of organisms used forexperiments and (d) the approach to comparisons The combined variationof these properties produces the enormous variance in evo-devo researchtasks and associated necessary protected spaces
Empirical StrategyOne of the most important distinctions for the transition to evo-devoresearch was that between experimental evo-devo research tasks and otherforms of evo-devo research (Table 2) The easiest way to engage in evo-devo research and one that was the entry ticket to evo-devo for many ofour interviewees can be described as conceptual extension Conceptualextension occurred when researchers continued their evolutionary biologyor developmental biology research line including the presentation of their
Table 2 Empirical Strategies Affecting the Protected Space Required forTransitions to Evo-Devo
Empirical Strategies Impact on Dimensions of Protected Space
Resources Time horizon
Conceptual extension None Low
Theoretical research None Low
Bioinformatics research None Low
Experimental research Depends on other properties Depends on other properties
243Highly Adaptable but Not Invulnerable