Mapping the enviro security field

RESEARCH

mapping the ‘enviro-security’field: rivalry and cooperation inthe construction of knowledgesarah saublet* and vincent larivièreUniversity of Montreal, 3744 Jean Brillant, 525-7 Montreal, QC, Canada H3T 1P1E-mail: [email protected]

*Corresponding author.

advance online publication, 24 June 2016; doi:10.1057/eps.2016.26

AbstractThis paper maps out the network of researchers on environmental securityfrom the end of the 1980s to 2014, providing a systematic analysis of howthe research is organized in this interdisciplinary field. Drawing on theWeb ofScience database, we generated a co-citation analysis that exhibits thecognitive structure of the field. Twenty interviews were conducted in orderto uncover relationships of cooperation, rivalry and conflict so as to under-stand the structure of academic debates inside the field over time. Theresearch findings highlight that central authors have had a long-lastinginfluence on the field despite the evolution of their productivity. We alsofound that the field is composed of six fairly structured groups aswell as a fewperipheral authors, which can in turn be distinguished by their epistemologi-cal and methodological choices, as well as geographical centre of gravity.

Keywords scientific field; sociology of knowledge; environmental security;co-citation analysis

One of the most striking develop-ments in recent decades has beenthe growth in the popularity of

environmental security as an area ofresearch investigation. Indeed, climatechange is now regarded as a securitythreat by the United Nations SecurityCouncil:

We must make no mistake. The factsare clear. Climate change is real, and itis accelerating in a dangerous manner.

It not only exacerbates threats to inter-national peace and security, it is (itself)a threat to international peace andsecurity.1

The linkage between these issues hasmost recently been noted by policy-makers and researchers who haveadvanced the argument that the droughtin Syria (because of climate change) wasa causal factor in the outbreak of the con-flict (Fountain, 2015). This is a concern

european political science: 2016

(1–20) & 2016 European Consortium for Political Research. 1680-4333/16 www.palgrave-journals.com/eps

1

http://www.palgrave-journals.com/eps

that has not been reflected in the findingsof policy reports, of which the Intergovern-mental Panel on Climate Change (IPCC)in its Fifth Assessment Report devoted achapter to the links between climatechange and security (IPCC, 2014). In asimilar vein, in June 2015 the G7 summitattached particular focus to the threatposed by climate-fragility risks and madereference to tackling the issue in prepara-tion of the Conference of the Parties to theUnited Nations Framework Convention onClimate Change in Paris that took place6months later in December 2015 (COP21).In the scientific community, however,

the consensus about the relationshipbetween the environment and violent con-flict is weaker.2 Scholars have been study-ing the impact of environmental factors oninternational peace and security for dec-ades (Homer-Dixon, 1994; Lee, 2009).After the end of the Cold War, a debateopened up as to whether the environmentshould integrate security studies as a newvariable (Brown, 1977; Ullman, 1983).This in turn led academics to considertwo kinds of questions: Is climate changea new cause of war? How will climatechange affect security issues? As scien-tists wrote more about this topic duringthe 1990s, the environment became animportant concern in International Rela-tions as well as in other disciplines in thehumanities and social sciences.

Every scientific choice – the choice ofthe area of research, the choice ofmethods, the choice of the place ofpublication, the choice between rapidpublication of partly checked resultsand later publication of fully checkedresults – is in one respect – the leastavowed, and naturally the least avow-able – a political investment strategy,directed, objectively at least, towardsmaximisation of strictly scientific profit,i.e., of potential recognition by theagent’s competitor-peers. (Bourdieu,1975: 22)

In line with this view, Collins (1998)shows that an intellectual network is aconstrained market where symbolic strug-gles take place in order to gain eminenceand recognition. Bourdieu (1975) andCollins (1998) describe the competitivenature of the scientific field. In scientificfields, we expect to observe ‘coalitions ofthe mind’, that is, competing groups.3 Byforming arguments and lineages, groupsstruggle for eminence within the field.From Collins’ view, groups can be meta-phorically represented as rival camps. Theproduction of knowledge does not onlystem from a pure interest in science. It isa socially situated activity through whichits actors look for recognition as legitimatespeakers. Thus, we expect to encounterrivalries, conflict and disagreements. Inthe “enviro-security” field, researchersdisagree on three main issues: the secur-itization of the environment and climatechange,4 the measure of climate change’simpacts on society and the interpretationof empirical evidences.5 Conflicts arise notonly from those disagreements but alsofrom intrinsic rivalries of the scientific field.We conceive rivalry as the social state ofthe scientific field in which actors areengaged in a competition to claim theirlegitimacy. They are ‘competitor-peers’:statistical studies published in highlyrecognized journals (Journal of PeaceResearch, Science or International Secur-ity) get a lot of attention compared withqualitative case studies that are often area

‘Environmental securitycan be described as ascientific field where

controversies, debatesand collaboration take

place among researcherswho are socially

situated’.

european political science: 2016 mapping the ‘enviro-security’ field2

oriented and published in specialized jour-nals. As researchers form lineages, coali-tions and cooperation are other definingaspects of the scientific field. Researcherscooperate when they offer explicit supportto each other, sustain relationships for-mally or informally6 or when they trainnew researchers to continue their work.It is, however, important to note that the

idea that the ES field is made up of well-structured groups is more often assumedthan empirically demonstrated. By lookingat the content of scientific contributions,authors have reached the conclusion thatthe field is fragmented (Deligiannis, 2012;Floyd and Matthew, 2013). Indeed, theyhave shown that this is not a homogeneousfield but a ‘polysemous category’ withdifferent approaches and methodologies(Floyd and Matthew, 2013). As Ronnfeldt(1997) has shown, research on the envir-onment and security is a three-generationfield. In the early 1980s, the first genera-tion focused on the integration of environ-ment factors into the concept of security(Brown, 1977; Ullman, 1983). The secondgeneration used case studies to identifyempirically the pathways from environ-mental stress to conflicts (Homer-Dixon,1994). Finally in the late 1990s, the thirdgeneration offered renewed methodologi-cal means to investigate the links betweenenvironmental degradation and interna-tional security (Hauge and Ellingsen,1998; Gleditsch, 1997). Dalby et al (2009)described three phases of environmentalsecurity, differentiating them according totheir institutional affiliations.In reflecting on these points, our paper

draws upon previous viewpoints of thefield regarding the idea that it is divided.But at the same time, we conceive thisheterogeneity as a marker of its competi-tive nature. We therefore adopt a socio-logical position on the evolving ES field. Tofind out how the field has been structured,we initially asked the following questions:Who are the leading researchers? Whatare their relevant publications? What are

their institutional affiliations? (Reid andChen, 2007). Is there evidence of substan-tial sub-networks according to geographyand research programs objectives? Froman initial analysis of co-citations relation-ships at the early stages of the empiricalresearch (1990–1999),7 we then focusedon the 2000–2007 period before proceed-ing to analyse the field between 2008–2014. As the field is a ‘constrained market’we expect that its organization and itsactors’ positions in it change over time.

The analysis highlights that authorsbelong to six different groups as well asperipheral authors. Although no singleauthor dominates, we point out to a corecouple of researchers who seem to remainessential through time (Homer-Dixon andGleditsch). The cognitive base of the field iscomposed of scientific publications from theearly 1990s. We refer to it as the originalgroup. It stills plays a significant part in2014. Overall, we show that the researchis mostly led by North American and NorthEuropean groups (Canada, the UnitedStates and Scandinavian countries). As co-citation analysis offers an incomplete mapof the field as the social content of citationsrelationships is ignored, in order to under-stand how the field came about we need tolook for the meaning of those relationships.To this end, semi-structured interviewswere undertaken with twenty researchers,with these findings in turn being used totriangulate the systematic evidence ontheir scientific publications by bringing thenarratives behind relations of intellectualproximity, uncovering patterns of coopera-tion, rivalry and sometimes conflict. Theevidence from these interviews corroborate

‘We conceive rivalry asthe social state of thescientific field in whichactors are engaged in a

competition to claim theirlegitimacy’.

sarah saublet and vincent larivière european political science: 2016 3

the existence of ‘rival camps’ or groups,while at the same time they help to nuancethe mainstream view of the field as beingonly contentious. The paper is organized asfollows. First, we present our methodologi-cal design. Second, we investigate theresults of our analysis. Finally, we discussthe social network of people working on theenvironment and security.

METHODOLOGY

Our analysis relies on a combination ofmethodological tools. We first use biblio-metrics to perform a co-citation analysis.This helps unveiling intellectual relation-ships between scientists. The data offers aformal look at those relationships. Wethen conducted twenty semi-structuredinterviews to balance the results from theco-citation analysis. Using interviews, weanalysed the structure of the field fromanother standpoint.

TOOLS AND DATA SOURCE

While in recent years bibliometrics hasbeen increasingly used for the evaluationof researchers and institutions, one of itshistorical uses has been to provide empiri-cal data on the structure of the scientificcommunity (Cole and Cole, 1973; Merton,1973). Among bibliometric methods thatallow for such measurements, author co-citation analysis, which is based on thefrequency at which authors appeartogether in the bibliography of scientificdocuments (Small and Griffith, 1974), isprobably the most well-known. Accordingto McCain (1990: 433), author co-citationanalysis ‘can be used to produce empiricalmaps of prominent authors in variousareas of scholarship’.We aim to measure and visualize the

size and the composition of the network ofresearchers working on environmentalsecurity. Compiling the data, we system-atically looked at the structure of

interpersonal relations and therefore ana-lyse linkages among pairs of researchers.To identify core researchers we borrowedtools from social network theory. First, wemeasure the centrality of actors in thenetwork, which can be defined as thesum of the links one entity has (Freeman,1979). This measure is used as an indica-tor of influence, popularity and prestige(Carrington and Scott, 2011) and as aresult can quantify actors’ prominence inthe network (Knoke and Yang, 2008).Centrality deals with actors’ involvementin many relations whereas prestige is ‘theextent to which a social actor in a network“receives” or “serves as the object” of rela-tions sent by others in the network’ (Knokeand Yang, 2008). It allows us to under-stand the positions of the actors within thestructure. Second, we capture the relation-ships between actors’ positions. Are thereties among actors? Do some actors havefewer ties than others? Are there sub-groups (clusters)? (Hanneman and Riddle,2011). Following these guidelines, we arenot only able to locate the main actors inthe network, but we can also reveal intel-lectual ties between them.

A co-citation occurs when two scientificarticles appear in the same bibliography(Small and Griffith, 1974). Two differentnames referenced together in several arti-cles suggest a connection or proximity(social or cognitive) between the two per-sons. We want to know who is linked towhom and how closely because that canreveal both social and intellectual struc-tures within a field of enquiry (White,2011). In order to conduct a co-citationanalysis, we compile bibliometric datausing a citation index. The database weuse in this paper is the Web of Science(WoS), which is the most used data-base for bibliometric analyses.8

KEYWORDS AND CORPUS

Bibliometrics allows us to create a corpusfrom which we can analyse authors’


relationships. We created a list of key-words and in order to be systematic listedthe words and expressions that are likelyto articulate the environment and secur-ity. The query on the WoS has been madefollowing those keywords listed in Table 1(see the online appendix). We appliedkeywords to titles only, given that theabstracts of the papers were not indexedin the database before 1996.9 We alsofocused on a list of scientific journals. Inexploratory trials, we indiscriminatelysearched disciplines and journals that ledto thousands of occurrences. However,the majority of the results were eitherfalse positives (articles that should nothave been retrieved) or false negatives(titles that should appear but do not). Thefact that social sciences are at a disadvan-tage in terms of indexation led us tonarrow our query further. To select the listof journals, we used the results of pre-vious queries and searched for the mostrecurrent journals, whose indexed articlestitles matched the keywords. We alsoused the compiled data derived fromBuzan and Hansen (2009). Their workinvestigated the history and evolution ofinternational security studies. They iden-tified major journals in the discipline. Weassumed that the link between the envir-onment and security has been well cov-ered by those journals. Finally, we alsolooked at secondary sources like existingliterature reviews to include the mostrelevant journals. Table 2 lists the journalsselected (see the online appendix). On thewhole, we isolated 492 titles that formedour corpus from which the co-citationanalysis was produced.

INTERVIEWS

We performed twenty semi-structuredinterviews to enrich the formal aspect ofthe mapping. To conduct our interviews,we selected individuals that showed up onthe map from the core to the periphery.Given that we assumed the central nodes

represented the most important people,we therefore chose to begin with them.We refined our list of interviewees basedon the knowledge gathered in exploratoryinterviews. A few researchers who did notappear on the map were nonethelessinterviewed, as they were considered tobe just as important by their colleagues.We carried out interviews up to 1.5 h, withthe aim of obtaining three main kinds ofinformation. First, what do they think isthe current organization of the field? Sec-ond, how do they think the field hasevolved? Finally, we triangulated theresults of the co-citation analysis withtheir perceptions of who was/is involvedin the field. To find out who was importantwe proceeded in two steps. First, intervie-wees were invited to give their own inter-pretation of the relationships laid out bybibliometrics. What did they see? Second,we asked them to define their position,status and affiliation. Who did they feelclosest to? On the one hand, we were ableto validate the existence of groups or‘camps’. On the other hand, we couldnuance this vision of conflictuality by tak-ing into account actors’ interpretation oftheir positions.10

EMPIRICAL ANALYSIS

In this section we present the findings ofthe co-citation analysis, as well as ele-ments of corroboration from the inter-views. We analysed its organization andshape between 1990 and 2015. Generat-ing a systematic map of the ES field, wewere able to highlight six visible groups ofresearch that we present chronologically.The co-citation analysis shows the ‘Tor-onto group’ understood as the group ofpeople that have worked under the super-vision of Homer-Dixon in the early 1990s.After 1999, only Homer-Dixon appears inthe data.We prefer to focus on the originalgroup that published seminal theoreticaland empirical work on the links between


environmental degradation and violence,rather than maintaining a blurred ideaabout the ‘Toronto group’ (the ‘primemovers’). This original group was nothomogeneous as the first debatesoccurred within it (Homer-Dixon and Levy,1996). We also found that scholars fromScandinavian countries formed a specificgroup either called the ‘PRIO group’ or the‘Oslo group’. This group is quite homoge-neous and it revolves around Gleditsch. Itis also the most represented group.Indeed, it is institutionalized and is theonly one to ‘own’ two academic journals(Journal of Peace Research and SecurityDialogue). There was also a group of scho-lars working together through the RobertStrauss Centre at the University of Texas,Austin. They participated in the ClimateChange and African Political Stability Pro-gram (CCAPS). We called them schemati-cally the ‘CCAPS group’. We also found the‘Berkeley group’ as being a prominentgroup, especially after 2009. Yet in con-trast to others, it was a small group withonly three scholars being part of it. Thisgroup is more oriented towards econo-metrics compared with the others thatbelong to the social sciences. We alsomention two other European groups thatseem to be less important, of which thedata singled out two scholars (Bernauerand Koubi) from a Swiss-based group thatwe called the ‘ETH Zurich’ group. Bernaueris the research leader of the group ‘inter-national political economy’ based at theCentre for Comparative and InternationalStudies. This group focuses on the condi-tions under which international environ-mental and economic problems can besolved. Climate change is conceived asone of the many variables that can impactsociety and induce political violence (Koubiet al, 2014). The data also showed oneresearcher (Scheffran) from the Universityof Hamburg, who leads the ResearchGroup on Climate Change and Security,the ‘CLISEC group’. In undertaking thisresearch we expected people’s importance

in the network to vary with their produc-tivity: those who contributed to theresearch in the early 1990s do not neces-sarily work on the same issues today andconsequently we expected that actors’positions might change through time. Infact, the analysis showed the constantpresence of the original group thatopened the way for the empirical investi-gation of the linkages between the envir-onment and security. We also highlighteda core couple of researchers that survivesacross time, but we did not single out oneresearcher as being dominant.

THE FIELD BETWEEN 1990AND 2015

1990–1999: THE PIONEERS

Figure 1 represents the cognitive base ofresearch on environmental security. It is asmall and loose network with a diffusecore.11 During that time, the research wasled by a few political scientists. With a fewexceptions, it originated mainly from NorthAmerica. Thomas Homer-Dixon was at thecore of the early empirical research as itwas the biggest node in the network. He ledthe first empirical and systematic projectson environmental stress and violence afterhe published a seminal article in Interna-tional Security in 1991 (On the Threshold:Environmental Changes as Causes of AcuteConflict) (Homer-Dixon, 1991).12 Scholarsfrom the field we interviewed confirmed hisstatus as a pioneer because he paved theway for a new way to think about environ-mental stress. Nevertheless, the figureshows that he is not the only dominantactor. In fact, the network is also composedof scholars who initiated a conversation onwhether it was relevant or not to broadenthe concept of security after the end of theCold War. That is the case with Brown(1977), Mathews (1989), Westing (1989)and Buzan (1991), with whom Homer-Dixon is co-cited. These authors are part of


the network as they all stress the impor-tance of including environmental factors assecurity concerns. Furthermore, the net-work is characterized by the closenessbetween Homer-Dixon and Myers, as wellas Kaplan who appear on the map. Myerswrote a seminal article in 1989 pointing tothe emerging risks of climate change(Myers, 1989). Kaplan published a famousarticle, ‘The Coming Anarchy’, in 1994. Thisarticle intended to link scarcity, overpopula-tion and violence (Kaplan, 1998). Kaplanexplicitly cites Homer-Dixon, Gleick (1991)and Renner (1996), all of whom are alsoearly contributors of the academic move-ment that has aimed to connect theenvironment and security.13 The co-cita-tion analysis does not give much impor-tance to Baechler despite the fact that hewas Homer-Dixon’s European counter-part in the early 1990s (he is not visibleon the map).14 They were both workingon the same issues at the same timeusing similar methodologies (case stu-dies). Baechler led the Environment andConflicts Project in Zurich, Switzerland,

which has investigated the causal rela-tionship between environmental damageand degradation and actual or possibleconflicts. Although scholars mentionBaechler’s influence (they know he haspublished work on the subject), he doesnot show up as an essential contributor.The reason why he does not appear in themap lies in the fact he was less cited inthe first place. Therein, the US-centricnature of the early ES research canexplain that.15

Figure 1 illustrates the first disagree-ments too. After Homer-Dixon’s first pub-lications from 1991 to 1998, scholars likeLevy (1995), Deudney (1990, 1991),Gleditsch (1998) and Dalby (1992) startedpublishing pieces questioning the sound-ness and risks of linking environmentalissues to security. In particular, Deudneyrefuted the idea that resource scarcitywould lead to war. From the mid-1990s,those scholars were engaged in debates.Sometimes these are referred to as a ‘dis-pute’ between the Toronto group and theOslo group led by Gleditsch.16 The Toronto

DALBY-S

HOMERDIXON-TF

SMIL-V

BROCK-L

RENNER-M

DEUDNEY-D

LICHBACH-MI

GURR-TR

KRASNER-SD

MATHEWS-JT

SWAIN-A

MISCHE-PM

BUZAN-B

OPSCHOOR-JB

PORTER-G

WESTING-A

MISCHE-PLEVY-MA

CHAZAN-N

DOKKEN-K

BROWN-N

BROWN-L

CHOUCRI-N

LIPSCHUTZ-RD

WESTING-AH

GOODLAND-R

*WORLDCOMMENVDE

MYERS-N

MEYERS-N

GALTUNG-J

*WORLDBANK

GOLDSTONE-J

BENEDICK-RE

LODGAARD-S

GLEICK-PH

MULLER-EN

OPHULS-W

WAEVER-O

FRENCH-HF

SUSSKIND-LE

GLEDITSCH-NP

*WORLDRESI

BARBIER-E

KAPLAN-RD

HAUGE-W

HOLST-JJ

*UNENVPROGR

CAMILLERI-J

ECKSTEIN-H

LIBISZEWSKI-S

LEVY-M

GREENHOUSE-S

CONCA-K

KAKONEN-J

ULLMAN-RH

HOWARD-PSMITH-D

BRUNDTLAND-GH

GRAEGER-N

ELLINGSEN-T

EHRLICH-PR

MATTHEW-RA

LOWI-MR

GOLDSTONE-JA

HOLSTI-KJ

YOUNG-O

DABELKO-GD

Figure 1 The early research 1990–1999.Note: Only relationships of three co-citations or more are shown.


group in essence refers primarily to Homer-Dixon and his co-authors that includedresearch assistants and students.17 Thisemerging debate concerns mainly Homer-Dixon and Gleditsch. Indeed, data showthat Homer-Dixon has been co-cited withHauge and Ellingsen (1998) who co-authored an article with empirical interestsand conclusions similar to Homer-Dixon’s.Hauge is a member of the Peace ResearchInstitute of Oslo, like Gleditsch. We will seelater that it is more complex than that but itis worth noting that from 1998 onward, asGleditsch publishes a critique on ‘Armedconflict and the Environment’, he repre-sents the other ‘camp’.

2000–2007: A DENSER FIELD

Figure 2 is a simplified picture of the field asit becomes more complex. More research-ers are involved.18 We noticed the intellec-tual base of the research located at theperiphery of the network (early authorscircle the network). Homer-Dixon is stillhighlighted as a central actor and he issituated at the crossroad between

original research and new research.Being a pioneer, he acts as a bridge andhe is a reference that keeps showing upin the network even though his last pub-lished empirical work on the subject goesback to 1998 (his work is published intwo major books in 1998 and 1999).Figure 2 shows how strong the intellec-tual base of ES is as we find the samescholars with the highest number of co-citation relationships (Gleick, Levy,Hauge, Myers, Mathews, Deudney etc.).During that time, the field is dividedbetween the original research led byHomer-Dixon’s case studies and deductiveapproaches, and PRIO research led byGleditsch. The debate between Homer-Dixon and Gleditsch is primarily a method-ological debate. How can one best solvethe empirical question of whether environ-mental stress induces conflict? The dis-agreements between Gleditsch andHomer-Dixon thus constitute the secondintellectual base of the research on envir-onmental security. They are often co-citedbecause they represent the driving forcesof the research.

*WORLDBANK

HOMERDIXON-TF

WALKER-R

HAUGE-W

MIGUEL-E

THEISEN-OM

BARNETT-J

ESTY-DC

GLEICK-PH

SWART-R

COLLIER-P

DALBY-S

FEARON-JD

DEUDNEY-D

KAHL-CH

NORDAS-R

DESOYSA-I

MATHEWS-JT

SCHWARTZ-P

RENNER-M

BUHAUG-H

WOLF-ATLEBILLON-P

GLEDITSCH-NP

SMIL-V

HENDRIX-CS

MYERS-N

BEAUMONT-PLONERGAN-S

KLARE-MT

REUVENY-R

BROWN-L

*UNDP

KAPLAN-RD

*IPCC

HEGRE-H

DABELKO-GD

BECK-N

PAGE-E

CONCA-K

BAECHLER-G

SWAIN-A

OLEAR-S

LOWI-M

LEVY-MA

GOLDSTONE-JA

BROWN-N

MATTHEW-RA

BUZAN-BMATTHEW-R

LIPSCHUTZ-R

LIPSCHUTZ-RD

AUTY-RM

LUJALA-P

PERCIVAL-V

MARSHALL-MG

KING-G

TIR-J

*UN

SACHS-JD

BULLOCH-J

GLEICK-P

SUHRKE-A LEVY-M

ULLMAN-RH

CLINTON-W

GIZEWSKI-P

ROWLANDS-I

DIEHL-PF

HOWARD-P

Figure 2 A growing field of research 2000–2007.Note: Only relationships of four co-citations or more are shown.


2008–2014: THE ES FIELD TODAY

Figure 3 is again, a simplified map of thefield. In fact, the network becomes sodense that it is hard to pick up authors’names.19 At the core, we find three con-temporary groups of research oftenreferred to as the ‘Oslo group’, the ‘Berke-ley group’ and the ‘CCAPS group’.20 The‘CCAPS group’ is notably composed ofBusby, Raleigh, Salehyan and Hendrix.Quantitative methodology is at the heartof the project.21 One of the main objec-tives of this group is to know where andhow climate change poses threats to sta-bility in Africa (Hendrix and Saleyhan,2012). On the one hand, the data showsthe dominance of Gleditsch, Buhaugand Theisen (Buhaug’s Ph.D. student)from PRIO. On the other hand, it showsa close relationship between Hsiang,Burke and Miguel from the University ofCalifornia, Berkeley.22 In fact, the co-cita-tion analysis highlights the ongoing quar-rel identified in interviews between thesetwo groups. Hsiang et al start publishingon climate and conflict in 2010.23 Since

then, they have been engaged in metho-dological debates with Buhaug et alHsiang et al are not war and peaceresearchers contrary to PRIO researchers.They tend to be closer to econometricsand use mathematical models to test theinfluence of environmental events on con-flict (Hsiang et al, 2013). In fact, they aretrained in climate physics and economics.

The analysis catches the intenseexchanges occurring between these tworival groups from 2010 until the 2014 Cli-matic Change commentary ‘One effect torule them all? A comment on climate andconflict’. Twenty-six authors co-signed acomment directed against the work of the‘Berkeley group’.24 Several authors of thiscomment appear on the map: Buhaug,Bernauer, Koubi, Brzoska, Busby, Fjelde,Gartzke, Gleditsch, Hegre, O’Loughlin,25

Raleigh, Scheffran, Theisen and Urdal.These scholars belong to four differentgroups that formed a coalition againstthe ‘Berkeley group’. There is the ‘Oslogroup’ (Buhaug, Fjelde, Gleditsch, Hegre,Theisen and Urdal), the Swiss FederalInstitute of Technology – the ‘ETH Zurich

DESOYSA-I

HOMERDIXON-TF

BERGHOLT-D

THEISEN-OM

URDAL-H

SALEHYAN-I

BUHAUG-H

DALBY-S

RALEIGH-C

CICCONE-A

BUSBY-JW

OLOUGHLIN-J

GLEICK-PH

KEVANE-M

ZHANG-DD

MIGUEL-E

BARNETT-J

STERN-N

HEGRE-H

SLETTEBAK-RT

SCHWARTZ-P

COLLIER-P

BUZAN-B

SMITH-D

HENDRIX-CS

ADANO-WR

NEL-P

TOL-RSJ

JENSEN-PS

*IPCC

*CNACORP

GLEDITSCH-NP

HSIANG-SM

KAHL-CH

BURKE-MB

BENJAMINSEN-TA

REUVENY-R

BERNAUER-T

CAMPBELL-KM

FAIRHEAD-J

DEUDNEY-D

WITSENBURG-K

*UNDP

MEIER-P

SCHEFFRAN-J

BAECHLER-G

PELUSO-NL

*WBGU

GURR-TR

LEBILLON-P

KAHL-C

KAPLAN-RDCONCA-K

NORDAS-R

BACHLER-G

ADGER-WN

FJELDE-H

SUHRKE-A

LEVY-MA

BRANCATI-D

FEARON-JD

MYERS-N

GOLDSTONE-JA

HAUGE-W

MATTHEW-RA

KOUBI-V

*UNEP

*UN

Figure 3 Recent research 2008–2014.Note: Only relationships of eight co-citations or more are shown.


group’ – (Bernauer and Koubi), the ‘CCAPSgroup’ (Busby and Raleigh), the ‘CLISECgroup’ from Hamburg, Germany (Brzoskaand Scheffran). We noticed that theresearch has developed outside NorthAmerica and Norway and that there is agrowing field in Europe, especially in Ger-many (CLISEC, Max Planck Institute).CLISEC was launched in 2009 as a multi-disciplinary research group, working withthe Institute of Geography at HamburgUniversity. CLISEC members use a vastarray of methods including qualitativemethods as well as data and modellingtools (Scheffran et al, 2012). Scheffran isoften co-cited with Gleditsch and Buhaugas CLISEC research programme is quitesimilar to PRIO’s. CLISEC emphasizes theneed to answer why there is a relationshipbetween climate and conflict if data say so.CLISECmembers also conduct fieldwork totake local complexities as a major variablein the relationship (Scheffran et al, 2014).Furthermore, on the right side of the

map we see the original group withHomer-Dixonwho is still an essential refer-ence in the field even though he is not anactive member in terms of publications.At the periphery, we find critical voiceslike Dalby and Barnett, as well as Adgerand Peluso, of whom Barnett, Adger andPeluso are geographers. By contrast,Dalby is more oriented towards criticalgeopolitics. The network represents theempirical research on climate and con-flict and as such, the data we compileddoes not give a lot of weight to publica-tions about human security, peace build-ing or adaptation.26 This can explain whywe find these authors at the periphery. Itseems like the research on environmen-tal security in the climate conflict aspectis mainly a methodological discussionbetween two quantitative groups, thatis, the ‘Berkeley group’ and the ‘Oslogroup’. In the meantime, the field seemsto continue a typical configuration withHomer-Dixon and Gleditsch as twounshakeable pillars.

1990–2014: A GENERALDEPICTION OF THE ES FIELD

Figure 4 is a general snapshot of the co-citation network of authors cited by paperson environmental security. It sums upFigures 1, 2 and 3. It is more comprehen-sive. It is interesting to see that the networkis centralized around two authors although,again, no single researcher is dominant.Homer-Dixon andGleditschmaintain a cen-tral position through time, as leaders of twodistinct research agendas. On the left sideof the network, we find the original group(Ullman, Brown, Mathews and Myers),empirical validations of Homer-Dixon’swork (Renner, Gleick, Kahl and Brauch),and early critics of his work (Deudney andLevy). On the right side, we look at themid-1990s research first led by Gleditschthrough PRIO and perpetuated by his pupils(Buhaug, Urdal and Hauge) through the2000s. Figure 4 also picks up the actualmethodological debate between Buhaugand the ‘Berkeley group’ composed byHsiang, Burke andMiguel. Aswewill discussthis later, we find CCAPS people (Hendrix,Raleigh, Salehyan and Busby) ‘navigating’between PRIOpeople as they sustain strongcollaborations with them. The same remarkapplies to other European scholars like Ber-nauer (ETH, Zurich), Scheffran (CLISEC,Hamburg) andAdono (Max Planck Institute,Munich). The left side of the network ismore case study oriented whereas the rightside is more oriented towards statisticalmethods to assess the relationship betweenthe environment and conflicts. Both of themuse deductive and positivist approaches.The divide is foremost an epistemologicaldivide. What counts as knowledge? At theperiphery, we find scholars who use differ-ent frameworks. For most of them, politicalscience and war theories are not satisfyingto think about the ways the environmentimpacts social behaviours. Dalby, Hart-mann, Peluso and Adger share roots inpolitical geography and geopolitics. Theycontribute to the research as critical voices.


BETWEEN COLLABORATION,RIVALRY AND CONFLICT

Our analysis unveiled some of the coreresearchers of the ES field. Some of themare not active any more (Homer-Dixon,Myers and Kahl), and a new generation ofscholars is emerging (PRIO). Focusing onthe temporal dimensions of the field, weexpected to witness the changing config-uration of the research, that is, that newactors would have replaced the ‘foundingfathers’ in terms of centrality. It is not thecase. Instead, we find a relatively staticfield in terms of co-citations. In the nexttwo paragraphs, we discuss the organiza-tion of the field using qualitative informa-tion. It has emerged and evolved throughpatterns of conflict but also throughcollaboration.

THE ‘ENVIRO-SECURITY’ FIELD: ACLIMATE OF WAR?

Co-citation patterns do not say anythingabout what is really happening betweengroups or individuals. We go beyond the

formal analysis to bring out the socialdimensions of the research network.To do that, we conduct interviews withacademics. We face a contested field ofresearch in which, as Collins and Bour-dieu put it, struggles take place toacquire scientific authority. Conflict ispart of the everyday practice in a scien-tific field. Personal animosity is notsomething new, and it is not unique tothis field. Notwithstanding, we arguethat struggles within the ES field are notonly about empirical and methodologicalquestions. Scholars’ disputes also illus-trate a struggle for the recognition oftheir legitimacy as scientific actors. It isnot only about the objects of knowledge.Sharing his experience as a Ph.D. candi-date in the early 1990s, one professorsays: ‘It was a blood sport, and I got totell you people were mean to shit’.27

Interviews unveil knowledge aboutscholars’ assessments of their peers‘seriousness’ or their ‘lack of discipline’.The implicit question is thus: Who is legit-imate to speak in the field? Most of ourinterviewees took a critical eye on their

Figure 4 The ES field 1990–2014.Note: Only relationships of ten co-citations or more are shown.


peers’ practices. We therefore asked themabout the state of the field to understandhow contentious it could be:

There has been a lack of discipline Ithink in its field with people makingstatements (laughs) making claims,using data claiming that the data wassupporting their statements.28

People have not been careful aboutbroad sweeping claims that their stu-dies weren’t designed to address.29

He has veered off in a direction that isinsane. I mean his attempt to dismissthe recent work that is trying to eluci-date the evidence of causal connectionbetween climate stress and political vio-lence which for me is exactly the kind ofwork that is important to do.30

I think it (the empirical research) ispretty much sterile. Part of the problemthere is that people are trying to gener-ate something that looks different andthen hiking up the claims because theytry to justify the work.31

The ES field has been polarized aroundthree questions: Does environmentaldegradation (because of climate change)lead to conflicts? What is the best way toanswer this question? Who does it best?Researchers have engaged in, some-times, very heated debates. Since the1990s there have been different stakes.First, there was the question of redefiningthe concept of security after the Cold waras the containment philosophy becameobsolete. There were, then, opportunitiesto be grasped by scientists. ‘People didn’thave a clue about how to change secur-ity’.32 Researchers tell us about howearly works (referring notably to ThomasHomer-Dixon’s work) had been ‘undersiege’ from the beginning; not only frompeople whowere questioning the researchbut also from people who had difficultieswith others’ successes. Several intervie-wees insist on the fact that the large

amount of research money engenderedprofessional jealousy. The early 1990shave been described as really intense.One of our interviewees has started hiscareer in the think tank community. Con-sequently, he has witnessed the emula-tion around the theme of environmentalsecurity, both in the policy community andin academia. He remembered about whatmade the field a contested one. He told usabout the consequences of handing newresearch to policymakers without partici-pating in the discussions and thus riskingto be instrumentalized. At the sametime, the publicity that comes with newresearch in the security realm also begetsjealousy. As our interviewee puts it:‘There is a combination of that and “wedon’t like people who succeed, I’m jealousand I’m too big for your purchase” right?So a lot of this is personal’.33

Second, the research has becomemore and more sophisticated in order toexplain the relationships between theenvironment and security. The debateshave focused more on methodologicalaspects of the research. This situationmaintains two kinds of tensions withinthe field. The first one deals with somescholars’ frustrations to see the researchgoing ‘out of steam’. Some express theirimpatience to see the research move on:‘Ok! We got the message, we need to payattention’.34

This frustration is often discredited ascoming from the ‘critical theorists (who)have an epistemological bone to pick withthe whole literature’.35 It reinforces theidea that scholars who keep discussingmethodological aspects of the researchrender the debate sterile. As such, theyseem less legitimate.

The second tension lies in the ‘big fight’between those who continue to search forgeneralizations thanks to more and moretechnical mathematical tools and thosewho think they are missing the point.Interviews uncover a conflict about thescope of the research. A young academic


defines his research as doing ‘the best jobpossible’ referring to the idea of goodscience. He says:

In reality policymakers are interested ingood science. Every day I try to teachmy student to do good science, not totalk about ideology. My interest is inproducing good science to design goodpolicy.36

While other respondents do not denythe possibilities that there exist importantlinkages between environmental factorsand conflicts, their reaction relates to thedesign of the research. Its sophisticationis not uncontroversial, as one researcherexpresses it:

Our reaction is primarily a reaction tothose few studies we see as overstatingrelationships or grim results that aremade for publications that receive a lotof attention.37

When we address those issues withresearchers that are not part of theimmediate debate, we face discontentverbalized with humour and irony. Oneinterviewee puts it in a provocative way:

They like to go back and forth (Laughs)About quantitative people, honest toGod, they just love to blow each otherup.38

We identify the recurring use of thelexical field of battle. Exchanges betweenscholars have been defined as ‘attacks’between ‘camps’, and conversations havebeen called ‘disputes’, ‘shoutings’ or ‘yell-ings’. Critics have looked like personalattacks or ‘snipping’. This is how actorsdefine some of the exchanges they havewith each other:

Dispute is a word that also often comesup when people describe the conversa-tion. Conversation is more to talkabout a formal communication. It hasbeen somewhat nastier than thatunfortunately.39

Contentious? (Laughs). I stay out ofthe personal fights (pause) he is a niceguy (pause) he had published in blogs(sigh) hum you know maybe it wasdone on a spirit of academic debatebut it seemed more like pointed(pause) snipping.40

Scholars tend to accuse each other forbeing ‘aggressive’ or ‘nasty’. ‘Those peo-ple are not particularly my buddies’, saidone of them.

You know there have been a lot of con-troversies. People have had troublesmoving on.41

The particular flavour of the debate.It has become unusual. The subjectarea is something fascinating to thepublic, so I think a lot of folks in thepress have covered this. That got peo-ple excited. It is good for people to beengaged. On the other hand, if peoplebecome upset and angry in partbecause of the number of people watch-ing the discussion, it does not make forthe most productive discussions.42

One interviewee has shared his visionon the way ‘people do their research’.According to him, they often ‘mischarac-terize’, ‘downplay’ other people researchto ‘catapult’ and ‘emphasize’ their ownresearch. Again these excerpts point outto the struggle that aims to distinguishcompetent actors from others. This isalso visible through scholars’ assess-ment of the maps. Do the maps mirrorthe state of the field? We found thatscholars reacted to their own position inthe field either by being surprised, flat-tered or satisfied. In fact, their reactionwas twofold. First, they said our analysis‘overemphasizes’ one’s importance inthe network, or ‘underestimates’ some-one’s contribution. Second, they alsotried to justify their position and identifythemselves in relation to other by label-ling their work. We asked them to define


their position in the field by showing theresults of the co-citation analysis:

I am a little island to myself. That fitsmy intuition. If you look at the publica-tions, they came to a halt a pretty longtime ago. Broker role is definitely therole that I’ve taken.43

There is no one person that has figuredstuff out. Everybody is highly limited orflawed.44

When challenged about their promi-nence, researchers say:

I am a generous soul. I’m not doing thisfor any kind of gratification. People cansay whatever they want.45

I notice there is a big node around me(chuckles).46

Since our paper, a dozen of papers havecome out where people use our statis-tical framework and obtain results thatare almost consistent with what wepublished. So there has been lots ofvalidation of that result.47

The use of labels to identify groups ofresearch is another marker of the field’sheterogeneity. They are geographicallyconstrained. We used them to map thefield, but they are used by people fromthe field as a way to define either colla-borators or rival camps, as the ‘Berkeleygang’ label shows. We asked whethertalking about the ‘Oslo group’ is mean-ingful for PRIO researchers. Gleditsch andBuhaug responded negatively. They saw itas an exaggeration, a label put on byothers. As Gleditsch reflected, there is a‘bunch of individuals’ working togetherfrom different countries. The ‘Berkeleygroup’ seems also to be misleading as thework by Hsiang et al has been produced atColumbia and Princeton, before he movedto Berkeley. Nevertheless, the ‘Berkeleygroup’ was the target of the commentarywe mentioned in this paper, signed off bytwenty-six authors of the field. While they

saw this comment as a needed statementto denounce controversial research, theircounterparts saw it as a real ‘petition’(Buhaug et al, 2014). We view it as theculmination of a ‘climate of war’.

A COLLABORATIVE FIELD

Using co-citations we looked at the cogni-tive relations within groups of researchers.In the midst of what seems to be a con-flictual space, we found that the view of thefield as being composed of antagonisticgroups was only one part of the story.Setting formal considerations aside, wepresent a few examples of collaborativeties expressed mainly through the ideas of‘friendship’, and ‘support’. 48

Interviews unveil the existence of infor-mal collaborative groups and relationshipsin the early 1990s. For instance, scholarsremember about the time they weregraduates and research fellows as the‘Cambridge days’. There would be meet-ings gathering scholars from Harvard andthe MIT. Thomas Homer-Dixon was one ofthem. Homer-Dixon had an informal inter-disciplinary groups of peers with whom heshared his first thoughts on environmentalconflicts. There was also a growing com-munity with the American Association forthe Advancement of Science with whomHomer-Dixon became associated. He wasalso associated with the American Acad-emy of Art and Science at Cambridge aswell as the MacArthur Foundation where hewould meet Peter Gleick. In the Washing-ton area, the Council on Foreign Relations(Mathews and Dabelko), Georgetown Uni-versity (Matthew) and the University ofMaryland (Dabelko, Conca) have gatheredimportant contributors. All of them haveworked closely together and were suppor-tive of Homer-Dixon’s preliminary results.Although the debate that has occurredbetween Homer-Dixon and Gleditsch hasmade history, it overshadows the interestPRIO had in Homer-Dixon’s findings. DanSmith directed PRIO from 1993 to 2001


and he has maintained intellectual tieswith Homer-Dixon. During Smith’s termat PRIO, there was a strong interestin Homer-Dixon’s findings. They weredescribed as an interesting set of questionsat a moment where environment, conflictand development were not addressed in aserious way. As one researcher puts it:

You look at PRIO from an academicstandpoint and you see PRIO equalsGleditsch equals critique of TAD(Homer-Dixon) but no! PRIO also equalsSmith also working with TAD but not in a‘I agree with everything you say’.49

The actors of the debate remember the‘intellectual differences’ between PRIOGleditsch and Homer-Dixon but they seethem as a ‘genuine stimulation’. Homer-Dixon and Val Percival went to PRIO toshare their thoughts on the linkagesbetween the environment, developmentand conflict. Asking PRIO researchers’opinion on Homer-Dixon’s work, we findthat there is very much sympathy for theperspectives he has presented. However,choosing different angles of research theyhaven’t been able to find empirical sup-port for them.On the same note, co-citation patterns

do not show that Homer-Dixon maintainsintellectual relationships with most of itscritics. As one interviewee noted:

Dan Deudney is often cited as a criticof my early work. Over the years, espe-cially in the early days, we had reallyvaluable intellectual exchanges aboutenvironmental security issues. As itturns out, Dan and I see almost eye toeye on most environmental securityissues. That’s not at all apparent in thepublic documentation.50

Although themainstream story opposesHomer-Dixon and Deudney, they bothinsisted that they have been ‘long-timefriends’. Deudney recalled that hereviewed Homer-Dixon’s Environment,

Scarcity and Violence book for PrincetonUniversity Press and he recommendedits publication. Moreover, Deudney addsthat Homer-Dixon drew on his 1990Millennium paper, which was highlycritical of the securitization of theenvironment:

In my article, I focus on five scenarios.He only focuses on one. He conceivesthe others as not likely to conduct toconflict. And the one he focuses on, it’sthe one that I conceive the most likely(chuckles).51

Scholars spend time abroad and thistranscends traditional institutional fron-tiers. PRIO has been the academic hubfor many scientists in the field, as theexample above shows.52 They talk about‘personal connections’ and ‘friendships’.Those relationships are not visiblethrough the co-citation analysis. Alsonot visible is the informal relationshipbetween Hsiang and Levy. The latterwas been one of Hsiang’s supportersduring his years as a Ph.D. candidateand later as a post-doctoral researchfellow at Columbia University and Prin-ceton University.53 In line with this,Hsiang remembers receiving an emailfrom Homer-Dixon after the publicationof his Nature article (2011) about how‘Civil conflicts are associated with globalclimate’. It came as a recognition froman early pioneer although Hsiang hadlittle knowledge about the field in itself.This shows how intellectual proximityand collaboration work from a very infor-mal standpoint. Although the story tellsthat Levy and Homer-Dixon were at oddson environmental security during the1990s, they seem to share commongrounds on the linkages between the envir-onment and security. In fact, Levy hasrecently launched Columbia University’sEnvironment, Peace, and Security Certifica-tion of Professional Achievement, aiming atpractitioners to develop skills to better


understand the connections between theenvironment and security. This implies,along the fact he supports Hsiang’s viewon climate conflicts, that he also sharesHomer-Dixon’s view on environmentalconflicts.To sum up, we have described the

organization of the field looking at co-citation relationships. This formal analy-sis helps to identify the main actors ofthe ‘enviro-security’ field research. Ithighlights groups of research systemati-cally. We have argued that the map ofthe field is incomplete if we do not givemeaning to the social content of thoserelationships. This widens our under-standing of the competitive nature ofthe field. We also nuance this view,as the story told by scholars is alsomade of informal and collaborativeexchanges.

CONCLUSION: ANIMPERFECT IMAGE ANDAVENUES FOR RESEARCH

That climate change is a security threatseems to be evident in the general dis-course. In this paper, we performed astructural analysis of the enviro-securityfield as we face a multiplicity of scientificarguments and evidences that challengethe idea that there is a consensus. Wechose two standpoints to understandhow the field is organized. First, wetraced the contours of the research byidentifying core researchers. Second, wefocused on the social dimensions ofknowledge production to show how thefield came about. Relying on co-citationanalysis and researchers’ narratives, weargued that behind formal relationshipslie patterns of conflict, rivalry and colla-boration. Those patterns relate to anessential feature of academic practice,that is, its competitiveness. The ES fieldis composed of six visible groups andperipheral authors that partake in the

advancement of knowledge at the sametime that they struggle for their recogni-tion as competent scientific actors. Ouranalysis contributes to the empiricalwork on academics as ‘competitor-peers’.

Other standpoints are needed to fullyunderstand how the research on theenvironment and security has evolved.A lot of what is produced on this subjectcomes from the grey literature. Scholarshave personal blogs where they continuehaving interposed conversations. Theyalso write on institutional blogs, as it isthe case for New Security Beat, The blogof the Environmental Change and Secur-ity Program. They encapsulate theirarguments in brief forms such as videos.They participate to other activities likewriting reports and policy briefs. Theworld of NGOs and think tanks does havean important role in the way we talkabout the environment as a securitythreat. For instance, the four maps showhow a certain number of non-academicreports pervade the ES field. It isthe case for the World Bank, theIPCC, the Center for Naval AnalysisCooperation, the UNEP and so on. Thisdraws attention to the circulation ofknowledge. It would be interesting toanalyse how researchers carry theirideas and expertise in other fields thuscontributing to make the environment amajor concern.

‘The ES field is composedof six visible groups andperipheral authors thatpartake in the advance-

ment of knowledge at thesame time that they

struggle for theirrecognition as competent

scientific actors’.


Notes

1 In early July 2011, the 6587th meeting of the SC was devoted to the maintenance of internationalsecurity and peace and the impacts of climate change. Available at http://www.securitycouncilreport.org/monthly-forecast/2011/07/lookup_c_glKWLeMTIsG_b_7535735.php.2 The 2014 IPCC report reflects that. The chapter on climate change and security takes a careful positionreflecting the ongoing controversies within the academic sphere.3 We refer to groups of research as (a) groups of researchers working on the same question using similarmethods (CCAPS group), (b) interdisciplinary groups working under the general theme of environmentalsecurity using different frameworks but located at one institution (CLISEC group), (c) research group inwhich the environment is only one factor among others (Oslo group).4 Securitization is a concept developed by the Copenhagen School that intends to explain how generalissues are transformed in security problems (see Waever, 1995).5 We refer to evidences that either link the environment to the occurrence of conflict or empirical data thatinfirm the relationship.6 Co-authorship is one way to assess cooperation. We do not engage in this analysis here.7 The co-citation analysis captures works published before 1990.8 The WoS indexes articles, research notes and review articles published in peer-reviewed scientificjournals. Despite the fact that books or book chapters are not indexed as source items, the citations theyreceive in articles are included and, hence, it is possible to assess the role of their authors in the co-citationnetwork. Although the WoS is the most comprehensive and systematic database, we face somelimitations. Notably, the indexation is unbalanced across disciplines (Archambault et al, 2006; Larivièreet al, 2006), with a lower coverage for social sciences and, especially, humanities.9 As Figure 1 shows, the WoS has been indexing articles on the topic since 1954.10 Facing the impossibility of distinguishing ourselves from our object of study, we choose to protectrespondents’ identity. We tell the story using actors’ narratives anonymously as much as possible. Thisway we hope not to recreate tensions within the field. Respondents have been contacted by email and wehave exposed our guidelines to them. They have been informed by a follow-up email that their names willnot appear with a few exceptions.11 This allows to visualize more co-citation relationships (three or more are represented).12 There was the project for the Environment, Population and Scarcity (1994–1996) and the project onEnvironmental Scarcities, State Capacity and Civil Violence (1994–1998)13 Michael Renner is not an academic. Working at the WorldWatch Institute, he has published a bookuntitled Fighting for Survival: Environmental Decline, Social Conflict, and the New Age of Insecuritymeantfor general audiences. He has also published work through WorldWatch papers. The WoS shows that hisbook has been cited in scientific articles.14 Most of Baechler’s work has been published as monographs or as parts of edited volumes. But,citations received by such volumes in papers that form the corpus are counted. He may has been co-citeda couple of time though but the map shows only relationships of three co-citations.15 It can also be explained by the ‘Atlantic divide’ that often comes up during interviews. Scholars refer tothe tendency to cite less European literature.16 We will discuss these labels later in the paper.17 Kelly, Howard, Percival.18 Here we propose to look only at relationships of four co-citations or more.19 The map shows relationships of eight co-citations or more.20 It is funded by the U.S. Department of Defense’s Minerva Initiative, a university-based, social scienceresearch programme focused on areas of strategic importance to national security policy.21 Raleigh also leads the Armed Conflict Location and Event Data Project supported by Minerva projectthrough CCAPS. It collects data on political violence in developing countries.22 Burke now holds a position at Stanford University but completed his Ph.D. at Berkeley.23 Hsiang alone or as a lead author starts publishing in 2010. Burke actually published an article in 2009in PNAS but the keywords we use could not match the title ‘Warming increases the risk of civil war in Africa’as they tend to narrow the possibility of false positives results. So this paper is not in our corpus, and wecan’t count co-citations in its bibliography.24 This example can be interpreted as a case of rivalry in the field: Twenty-six authors form a ‘coalition’against three other authors’ practices of research. This commentary denounces the failure to complywith basic research rules thus looking for its authors’ credibility as opposed to their competitors-peers(the political strategy of scientific choices)


http://www.securitycouncilreport.org/monthly-forecast/2011/07/lookup_c_glKWLeMTIsG_b_7535735.php

http://www.securitycouncilreport.org/monthly-forecast/2011/07/lookup_c_glKWLeMTIsG_b_7535735.php

25 O’loughlin does not belong to any of these groups. But he shares the samemethodologies. We can saythat he is close to Raleigh as two geographers using quantitative methods.26 However, as we use different combination of keywords to build our corpus, it is logical to find theseauthors in the network.27 Author interview, 12 March 201528 Author interview, 12 May 201529 Author interview, 17 February 201530 Author interview, 21 January 201531 Author interview, 23 February 201532 Author interview, 12 March 201533 Author interview, 12 March 201534 Author interview, 23 February 201535 Author interview, 23 February 201536 Author interview, 12 May 201537 Author interview, 4 May 201538 Author interview, 12 March 201539 Author interview, 3 December 201440 Author interview, 17 February 201541 Author interview, 12 May 201542 Author interview, 12 May 201543 Author interview, 21 January 201544 Author interview, 28 November 201445 Author interview, 15 December 201446 Author interview, 7 November 201447 Author interview, 12 May 201548 These examples are not representative of the wide collaborative network of research that only aco-authorship network analysis could systematically capture.49 Author interview, 6 March 201550 Author interview, 4 October 201351 Author interview, 16 June 201552 For example, Idean Salehyan (University of Texas at Dallas, CCAPS group) has been a guestresearcher at PRIO in 200553 Levy is deputy director of the Center for International Earth Science Information Network at the EarthInstitute within Columbia University

References

Archambault, É., Vignola-Gagné, É., Côté, G., Larivière, V. and Gingras, Y. (2006) ‘Benchmarking scientificoutput in the social sciences and humanities: The limits of existing databases’, Scientometrics 68(3):329–342.

Bourdieu, P. (1975) ‘The specificity of the scientific field and the social conditions of the progress ofreason’, Social Science Information 14(6): 19–47.

Brown, L. (1977) Redefining National Security, Washington: Worldwatch Institute.

Buhaug, H., Nordkvelle, J., Bernauer, T., Böhmelt, T., Brzoska, M., Busby, J., Ciccone, A., Fjelde, H.,Gartzke, E., Gleditsch, N.P., Goldstone, J.A., Hegre, H., Holtermann, H., Koubi, V., Link, J.S.A., Link, P.M.,Lujala, P., O’Loughlin, J., Raleigh, C., Scheffran, J., Schilling, J., Smith, T.G., Theisen, O.M., Tol, R.S.J.,Urdal, H. and Von Uexkull, N. (2014) ‘One effect to rule them all? A comment on climate and conflict’,Climatic Change 127(3–4): 391–397.

Buzan, B. and Hansen, L. (2009) The Evolution of International Security Studies, Cambridge: CambridgeUniversity Press.

Buzan, B. (1991) People, States and Fear: An Agenda for International Security Studies in the Post-ColdWar Era, Boulder: Lynne Rienner Publishers.

Carrington, P.J. and Scott, J. (eds.) (2011) ‘Introduction’, in The Sage Handbook of Social NetworkAnalysis, London: Sage, pp. 11–25.

Cole, J. and Cole, S. (1973) Social Stratification in Science, Chicago: University of Chicago Press.Collins, R. (1998) The Sociology of Philosophies: A Global Theory of Intellectual Change, Cambridge:

Harvard University Press.


Dalby, S. (1992) ‘Ecopolitical discourse: “Environmental security” and political geography’, Progress inHuman Geography 16(4): 503–522.

Dalby, S., Brauch, H.G. and Oswald Spring, U. (2009) ‘Environmental Security Concepts Revisited Duringthe First Three Phases (1983-2006)’, in H.G. Brauch, U. Oswald Spring, I. Grin, C. Mesjasz, P. Kameri-Mbote, N.C. Behera, B. Chourou and H. Krummenacher (eds.) Facing Global Environmental Change:Environmental, Human, Energy, Food, Health and Water Security Concepts, New York: Springer, pp.165–172.

Deligiannis, T. (2012) ‘The evolution of environment-conflict research: Toward a livelihood framework’,Global Environmental Politics 12(1): 78–100.

Deudney, D. (1990) ‘The case against linking environmental degradation and national security’,Millennium 19(3): 461–476.

Deudney, D. (1991) ‘Environment and security: Muddled thinking’, The Bulletin of the Atomic Scientists47(3): 22–28.

Floyd, R. and Matthew, R. (2013) Environmental Security: Approaches and Issues, Abingdon: Routledge.Fountain, H (2015) ‘Researchers Link Syrian Conflict to a Drought Made Worse by Climate Change’. New

York Times, 2 March 2015.Freeman, L.C. (1979) ‘Centrality in social networks conceptual clarification’, Social Networks 1(3):

139–215.Gleditsch, N.P. (1998) ‘Armed conflict and the environment: A critique of the literature’, Journal of Peace

Research 35(3): 381–400.Gleditsch, N.P. (1997) Conflict and the Environment, Dordrecht: Kluwer Academic Publishers.Gleick, P.H. (1991) ‘Environment and security: The clear connections’, Bulletin of the Atomic Scientists

3(47): 16–21.Hanneman, R.A. and Riddle, M. (2011) ‘Concepts and Measures for Basic Network Analysis’, in J. Scott

(ed.) The Sage Handbook of Social Network Analysis, London: Sage, pp. 340–369.Hauge, W. and Ellingsen, T. (1998) ‘Beyond environmental scarcities: Causal pathways to conflict’,

Journal of Peace Research 35(3): 299–317.Hendrix, C.S. and Saleyhan, I. (2012) ‘Climate change, rainfall, and social conflict in Africa’, Journal of

Peace Research 49(1): 35–50.Homer-Dixon, T. (1991) ‘On the threshold: Environmental changes as causes of acute conflict’,

International Security 16(2): 76–116.Homer-Dixon, T. (1994) ‘Environmental scarcities and violent conflicts. Evidence from cases’,

International Security 19(1): 5–40.Homer-Dixon, T. and Levy, M. (1996) ‘Debate’, Environmental Change and Security Project Report 2:

49–57, 58–60.Hsiang, S.M., Burke, M. and Miguel, E. (2013) ‘Quantifying the influence of climate on human conflict’,

Science 341(6151) online publication september 13: doi:10.1126/science.1235367.IPCC. (2014) ‘Summary for Policymakers’, in C.B. Field, V.R. Barros, D.J. Dokken, K.J. Mach, M.D.

Mastrandrea, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N.Levy, S. MacCracken, P.R. Mastrandrea and L.L.White (eds.) Climate Change 2014: Impacts,Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II tothe Fifth Assessment Report of the Intergovernmental Panel on Climate Change, New York andCambridge: Cambridge University Press, pp. 1–32.

Kaplan, R. (1998) ‘The coming anarchy’, in G. Tuathail, S. Dalby and P. Routledge (eds.) The GeopoliticsReader, London: Routledge, pp. 188–196.

Knoke, D. and Yang, S. (2008) Social Network Analysis, 2nd edn. Los Angeles: Sage.Koubi, V., Spilker, G., Böhmelt, T. and Bernauer, T. (2014) ‘Do natural resources matter for inter – And

intrastate armed conflict?’ Journal of Peace Research 51(1): 1–17.Larivière, V., Archambault, É., Gingras, Y. and Vignola-Gagné, É. (2006) ‘The place of serials in referencing

practices: Comparing natural sciences and engineering with social sciences and humanities’, Journal ofthe American Society for Information Science and Technology 57(8): 997–1004.

Lee, J.R. (2009) Climate Change and Armed Conflict. Hot and Cold Wars, New York: Routledge.Levy, M.A. (1995) ‘Is the environment a national security issue?’ International Security 20(2): 35–62.Mathews, T.J. (1989) ‘Redefining security’, Foreign Affairs 68(2): 162–177.McCain, K.W. (1990) ‘Mapping authors in intellectual space: A technical overview’, Journal of the American

Society for Information Science 41(6): 433–443.Merton, R.K. (1973) The Sociology of Science: Theoretical and Empirical Investigations, Chicago:

University of Chicago press.


http://dx.doi.org/10.1126/science.1235367

Myers, N. (1989) ‘Environment and security’, Foreign Policy 74: 23–41.Reid, E. and Chen, H. (2007) ‘Mapping the contemporary terrorism research domain’, International

Journal of Human-Computer Studies 65(1): 42–56.Renner, M. (1996) Fighting for Survival: Environmental Decline, Social Conflict, and the New Age of

Insecurity. The Worldwatch Environmental Alert Series New York: W. W. Norton & Company.Ronnfeldt, C.F. (1997) ‘Three generations of environment and security research’, Journal of Peace

Research 34(4): 473–482.Scheffran, J., Brzoska, M., Kominek, J., Link, P.M. and Schilling, J. (2012) ‘Disentangling the climate-

conflict nexus: Empirical and Theoretical assessment of vulnerabilities and pathways’, Review ofEuropean Studies 4(5): 1–13.

Scheffran, J., Ide, T. and Schilling, J. (2014) ‘Violent climate or climate of violence? Concepts and relationswith focus on Kenya and Sudan’, The International Journal of Human Rights 18(3): 369–390.

Small, H. and Griffith, B.C. (1974) ‘The structure of scientific literatures I: Identifying and graphingspecialties’, Science Studies 4(1): 17–40.

Ullman, R. (1983) ‘Redefining security’, International Security 8(1): 129–153.Waever, O. (1995) ‘On Security’, in R. Lipschutz (ed.) Sectorisation and Desecuritization, New York:

Columbia University Press, pp. 46–87.Westing, A.H. (1989) ‘The environment component of comprehensive security’, Bulletin of Peace

Proposals 20(2): 129–134.White, H.D. (2011) ‘Scientific and Scholarly Networks’, in J. Scott (ed.) The Sage Handbook of Social

Network Analysis, London: Sage, pp. 271–285.

About the Authors

Sarah Saublet is a Ph.D. candidate at the University of Montreal. She is affiliated to the MontrealCenter for International Studies. She has a Master’s degree in International and EuropeanLaw of the Université Paris-Sud, a Master’s degree in Political Science at the University ofLyon Lumière II and a research Master’s degree in International Relations, Security andDefense at the University of Lyon Jean Moulin III. She is supervised by Frédéric Mérand.Her thesis deals with the social construction of the ‘environmental threat’. She is particularlyinterested in the role played by academic actors in this process.

Vincent Larivière received his Ph.D. in Information Studies at the University of McGill in 2010.He is Associate Professor at the School of Library and Information Science at the University ofMontreal. He is Associate Scientific Director at the Observatory of Science and Technologyof Montreal. He holds the Canada Research Chair on the Transformations of ScholarlyCommunication.

Supplementary information accompanies this article on the European Political Sciencewebsite (www.palgrave-journals.com/eps)


www.palgrave-journals.com/eps

Documents

Mapping the enviro security field