What does the field of International Relations look like in South America ?

This article provides a comprehensive picture of IR in South America by applying content analysis to 7,857 articles published in 35 journals from six South American countries from 2006 to 2014 in order to discover what the predominant theories, methods and research areas in this field are, how scholars tend to combine them in their research designs, and what the profiles of regional journals are, regarding their epistemological, methodological and subject preferences. The findings reveal a predominantly Positivist and largely Qualitative discipline, resembling North American and European IR.


Introduction
S ome authors consider that the first Department of International Relations was created in 1917 at the University of Wales at Aberystwyth (Nogueira and Messari 2005, 3).Therefore, in a couple of years the academic community should be celebrating the centenary of a discipline that has since then evoked lively discussions.One of the main debates concerns what ideas and actors have shaped International Relations (IR) as a subject through the years.That is to say, whether the origins and main practices of this young North American/European-born discipline make it epistemologically limited.To what extent have normative criteria been restricting IR's general explanatory power?Even though one cannot conceive South America 1 outside Cold War, prioritizing security in a Realist logic, Latin America began to see the world in a vertical perspective, bringing out another kind of conflict, North versus South, denuding the asymmetries that characterize international relations.This world-view brought with it another set of concepts (e.g.center and periphery, development and underdevelopment, exploitation, inequality) and actors (e.g.social classes, unions, multinational companies, among others) to the research of international phenomena.Though this, research agenda is more connected to sociology of development than IR per se, overall structuralism remained a central paradigm in the Latin American conception of the international realm (González 2006;Fróio 2012).As any theory, Dependency Theory selects some aspects of reality and bypasses others.Of course, it offers another approach to international studies, but it also presents limits, such as its inability to analyze the interdependence processes that characterize the world nowadays beyond the center-periphery division.
The diversity of IR theories imposes classificatory difficulties, especially in the southern continent, where international reality has been an object of study through many disciplinary branches.We turn now to the institutional set-up of IR in South American academia in order to understand the discipline's regional underpinnings.
South American thought on IR mostly began in institutions other than universities.It was present in academia as well, but with little curricular differentiation from other subjects, and overall guided by more political, prescriptive and practical concerns.As an example, the study of international issues in Brazil was carried out by areas as different as Law, Economics, History and Political Science (Lessa 2005).Though these studies diversified and enriched those particular disciplines, they contributed little to the outlining of an autonomous IR field of research.The first Brazilian inquiries in IR began in institutions such as Itamaraty and the Armed Forces, which were guided chiefly by practical concerns, prioritizing historic and geopolitical approaches.The earliest courses in Argentina were connected to a diplomatic law tradition.Thus, they predated the onset of Political Science courses in some universities, and prioritized the formation of elite bureaucrats, politicians and also researchers (Bulcourf 2008).As early as 1927, the Universidad Nacional del Litoral (UNL) created courses on diplomacy and political science (Fernandéz and Guardamagna 2011).In Chile, the first courses in Political Science that emerged in the 1950s were also more concerned with forming statesmen, therefore emphasizing public administration and law (Gatica 2012).Chile was also the original headquarters to the Facultad Latinoamericana de Ciencias Sociales (FLACSO), created in 1957 by a UNESCO initiative to advance social sciences in Latin America and which now has 16 participating countries.
In Brazilian academia, an important early step towards the differentiation of IR was the opening of the Instituto Brasileiro de Relações Internacionais (IBRI) in 1954 in Rio de Janeiro.It would only be in the 1970s and 1980s that universities, such as the University of Brasilia (UnB) and the Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio), began to consolidate this field of research.The relation between universities and the state also moved forward with the creation of the Instituto de Pesquisa em Relações Internacionais (IPRI) in 1987 to deepen the dialogue between the Foreign Relations Ministry and academia, while the regional integration processes of the 1990s also produced a great mass of research.
However, IR's true boom in South America would arrive in the 2000s, with a significant increase in the number of undergraduate and graduate courses, as well as scientific journals in the region.In Chile and other Andean countries, though Political Science and IR courses grew a lot in the 2000s, they did not become as popular as those focused on public policies and administration (Fuentes and Santana 2005).The latter are also prevalent in Colombia and Venezuela, where Political Science and Law are generally fused into one single faculty (Diaz and Antar 2005) 3 .
In most South American universities, IR courses tend to be placed in Political Science and Law departments, with some exceptions in institutions that have specific focuses (e.g. military academies).
The progressive differentiation of the IR field and specialization of its scholars was accompanied by the creation of academic and professional associations, think tanks and other institutions gathering researchers and incentivizing scientific production in the region.The first centers were created in Argentina (Consejo Argentino para las Relaciones Internacionales, 1978), Chile (Consejo Chileno para las Relaciones Internacionales, 1989), Uruguay (Colegio de Licenciados en Relaciones Internacionales del Uruguay, 1989; Centro de Formación para la Integración Regional, 1993), andBrazil (Centro Brasileiro de Relações Internacionais, 1998).The majority of South American associations was created in the 2000s, still concentrated in the same countries: Associação Brasileira de Relações Internacionais (2005)

Theoretical trends
As stated by Bedin (2000, 62), IR theory is "a vision, an interpretation, a perspective of international or global phenomena, supported by some method, whose claim is to explain and give meaning to the events that are unfolding in the international arena".As the main IR paradigms had in their origin a great influence from Western thought, especially the United States, some authors argue 5 theoretical choices are prone to political implications, 3 It is also worth pointing out that though there are not as many undergraduate IR courses at Colombian universities, IR is the second most relevant subject within the curricula of political science undergraduate courses in the country (Fortou et al. 2013).
4 According to the Global Go To Think Tank Index (2014) published by the University of Pennsylvania 5 For instance, Cox (1981) and Linklater (1996).
What does the field of International Relations look like in South America?Rev. Bras. Polít. Int.,59(1): e004, 2016 Medeiros; Barnabé; Albuquerque; Lima 5 usually stated as limits and ideologies.Stressing the shortcomings of theories, Cervo (2008, 8) says that: The universal explanatory reach of theories is forged, since it is bounded to the interests, values and standards of conduct of countries or group of countries in which are developed and for which are useful, contradictorily to the concepts, that exposes the national or regional roots which are based on and refuse themselves to be invested of a global explanatory reach.
It is true that, with exception of Dependency Theory (André Gunder Frank, Fernando Henrique Cardoso, Enzo Faletto, Raúl Prebisch, among others), IR theories were developed by researchers from the West and, of course, they reflect a point of view from their places of origin.However, that should not mean that they are, automatically, unable to explain international phenomena in other parts of the world, as we discuss in this section.

Universalism versus Western World: inertial mimetic process and the search for autonomy
In the beginning of last century, Woodrow Wilson took the first steps towards the theoretical construction of IR.His 14 points paved the way for the building of Liberal Idealist Theory, and inspired the research and the organizations aiming a peaceful future built upon the principles of the capitalist market.
It was, however, after World War II that IR theories indeed began to take shape: the Realism of Morgenthau, Carr, Aron and Kissinger represented not only a reaction to the romantic Wilsonian idealism of the interwar period, but it has established itself as the classic IR paradigm.Along with Realism, the other major paradigm, Liberalism, sought, in its own way, to justify international reality, pretentiously presenting themselves as objective and universal theories.The influence of these currents stems not just from academia, but also from the international weight of their countries of origin.It can be argued that these theories are subjective in that they explain phenomena from those nations' viewpoint.But is this not the essence of all theories in the field of Humanities?As Tickner (2002, 10) said, "(...) a paradigm fulfills the function to set up those aspects that a scientific community will judge worthy of being investigated at the same time that it marginalizes the issues that (...) in any way contradicts the central ideas of the paradigm".
The question then is whether South American research has been guided by those influential Western currents, adapted them according to the reality investigated, or -given their insufficiency to provide answers to specific international issues raised by the periphery-sought autonomy of thought and offered an alternative analytical framework.
What does the field of International Relations look like in South America?Rev. Bras. Polít. Int.,59(1): e004, 2016 Medeiros; Barnabé; Albuquerque; Lima 6 Amado Cervo (2008,24) makes a call for researchers, mainly Latin Americans: To contribute to the end of international relations theories and their replacement by concepts applied to relations proposes itself as a way of transitioning from an international system that services the interests, values and patterns of action of capitalist old structures to another that welcomes interests, values and patterns of action of emerging countries.It proposes mental evolution corresponding to the material evolution in progress.
Nonetheless, simply discarding Western theories implies refuting potentially useful (though limited) interpretations, thus presenting a challenging choice for South American researchers.

Ideology is not Science
A common feature in the development of social sciences in South America is the marked presence of ideology in its theoretical foundations.As a region constantly seeking autonomy and impregnated by the discourse of colonial powers' distance, some South American scholars resist theories from the North, as if they were resisting new interferences on their internal affairs (Bernal-Meza 2005).Der Derian (1989) says that predominant narratives in IR studies tend not only to ignore divergent voices, but also to stimulate domination over historical practices, reinforcing dominant discourses.
Ideological bias is linked to an epistemological commitment which can be identified by answering a simple question: is it possible to make objective social science?If the answer is affirmative, we are looking at researchers who are concerned with explaining the real world through causal links between causes and effects, in a positivist epistemological approach; on the other side, if the answer is negative, we are assuming the impossibility of making social science in a value-free way, in which we do not explain, but are only able to understand the real world (Hollis and Smith 1990;Hollis 1994).The latest, interpretivist approach became known in IR as the alternative approach to positivist theories, called post-positivist approaches (Lapid 1989).
It may be useful to recall Weber (1973) in his commentary about the axiological neutrality of social sciences.According to him, normative judgments influence decisions regarding what to study, as well as presenting, at the end, the importance of the research.The fundamental issue is to assure that, while in the intermediary phases, the research is not conducted by normative values.Thus, the adoption of the scientific method is not an option, it is a necessity.
Theories may have research agendas, and legitimize epistemic communities, just as ideologies have corollary agendas and interest groups.Confounding science and ideology is, therefore, to mystify objectives, dominating groups, dominated groups and power structures, usually superposing ideological groups and epistemic communities.
It must be considered that, if it is possible to accept that an ideology-driven science is damaging to science advancement, there are many researchers who point to the impossibility of What does the field of International Relations look like in South America?Rev. Bras. Polít. Int.,59(1): e004, 2016 Medeiros; Barnabé; Albuquerque; Lima 7 completely separating science from ideology 6 .It is necessary to notice that though irrevocably connected, they can and must be subject to an effort of keeping each other apart, in order to preserve both of them as they were meant to be.Science is linked to formal processes of learning and knowledge production, while ideology is attached to social action and the ethics of conviction.
As we demonstrate in the next section, these trends are very present in an extensive body of articles from South American IR literature.

Empirical evidences
An accurate and up-to-date picture of the IR research field in South America can only be obtained by means of empirical verification.In order to grasp what is the main academic profile of the region, we have analyzed a very large corpus of scientific articles published from 2006 to 2014 following the methodology described in this section.

Methodology Corpus selection
We have analyzed the academic production of South American countries included in the Scientific Electronic Library Online (SciELO), a library and database for journals in all fields of knowledge founded in 1998.This library was chosen because: it is a repository created in Brazil and focused on the scientific production of developing countries, thus increasing the likelihood that our sample of articles would provide a reliable picture of IR in South America, and because it has a rigorous set of admission and permanence criteria to ensure that the journals it hosts meet high academic standards 7 .
Choosing only journals listed in SciELO implied the exclusion of some IR titles that, as of 2015, had not yet met their publishing requirements.Though this meant a reduction in the number of journals analyzed, it avoided ad hoc inclusions and it ensured relevance -as listed journals have to observe superior quality practices 8 .
SciELO has journals from the following South American countries: Argentina, Bolivia, Brazil, Colombia, Chile, Paraguay, Peru, Uruguay and Venezuela.Due to (1) the scarcity of strictly IR journals in those countries (only Brazil, Colombia and Chile have exclusive publications), (2) the fact that IR researchers publish in journals from different fields regularly, and (3) the interdisciplinarity that characterizes this area of study, our scope was enlarged so as to include journals from IR-related fields, such as Political Science, Sociology, History, Economics, Anthropology and Law.
7 For the full list of admission standards of SciELO Brazil, see SCIELO (2014).
8 The high quality of SciELO journals can be verified via a cross-reference with the Qualis index (available at http://qualis.capes.gov.br).
The majority of the journals that make it to SciELO are ranked B2 and above (upper half of eight possible grades).
What does the field of International Relations look like in South America?
Rev. Bras.Polít.Int., 59(1): e004, 2016 We have taken into consideration possible biases on our corpus selection.First, SciELO is an online database founded and maintained by a Brazilian research foundation.Consequently, it has a greater number of Brazilian journals.Though this could be a potential bias, it is nonetheless important to underscore that Brazil, having a larger number of academic centers, can be expected to account for the greatest share of IR research in the region.
Second, SciELO seeks to standardize scientific procedures in order to guarantee higherquality research.We chose it instead of other databases which could have more South American titles (e.g.Latindex) because it was the only one employing criteria of quality for hosting journals.
The corpus had articles written in six different languages: English, Spanish, Portuguese, French, German and Italian.The dominant languages are Portuguese (4,024 articles) and Spanish (3,303 articles), which account for 93% of all material.English is the third language used (6%), and though the distance from the two leading languages is great, its usage has risen steadily over the years, from only 26 articles (in 2006) to 87 (in 2014), indicating a growing tendency of researchers to publish in English.

Content analysis
Our interest was to verify the predominant (i) theories, (ii) methods and (iii) research areas in South American IR today.Focusing on those three domains allows us to get a concise picture of the main influences and interests in the field, specifically to (iv) discover prevailing research templates used by researchers and (v) the profile of regional journals.
The Teaching, Research, and International Policy (TRIP) Around the World survey has already provided some evidences of how the area looks like today, via their cross-national surveys with IR professors (Maliniak, Peterson and Tierney 2012).
Researchers attempting to investigate the state of the art of a given scientific field have some available courses of action.Traditionally, they can elaborate a comprehensive description of the history and trends of the area, drawing upon key developments, institutional history, and can more or less systematically attempt to qualify and quantify academic production -as done by Miyamoto (1999), Lessa (2005) and Almeida (2006) for Brazilian IR.Analysis of university course curricula is also employed to characterize the teaching tendencies of the field (Merke 2005;Froio 2012).
A more systematic approach is possible through meta-analysis, in which bibliometrics is a means of measuring science.Meta-analyses consist in manually analyzing a sample of articles, classifying each piece according to its theory, methodology, results, or other variable of interest 10 .To the extent of our knowledge, such analyses are non-existent in South American IR.Journals on IR have tangentially been included in Political Science meta-analyses 11 .
10 For a brief review of what is meta-analysis and its procedures, see Figueiredo Filho et al. (2014).
What does the field of International Relations look like in South America?Rev. Bras. Polít. Int.,59(1): e004, 2016 As we wanted to obtain the most representative picture of IR production in the region, we tried an alternative method which would allow us to handle a large corpus such as ours.Analyzing the 7,857 downloaded articles, and then manually classifying each piece would be humanly impossible.Therefore, we resorted to content analysis as a research method.By using this approach, we relied on a software to count word frequencies and to locate the occurrence of specific key terms predetermined by the user in thematic dictionaries.Therefore, we built a classificatory dictionary containing the key terms and names of authors that were most likely to be found in an article adhering to a certain theory, method or area.For instance, the terms Constructivism and Wendt were added under the Constructivism theoretical category, as we expect an article that uses a constructivist perspective to employ those words often.Naturally, not all matches can be interpreted as meaningful, only those pertaining to IR articles.This can be problematic due to the interdisciplinarity of our corpus and also because just counting a term can be misleading when the context is not taken into consideration.The term Liberalism, for instance, could be mentioned just as often by Liberal and Marxist authors, though with very different connotations.Thus, we also embedded context rules in our dictionary to avoid negative matches, especially for terms that are polysemic and tend to be frequent in non-IR articles (i.e.only count the term "Marxism" if it is present in the same document as the phrase "International Relations").It is worth underscoring that the usage of content analysis presupposes that authors explicitly name the theories, methods and areas used in the article.
We used the software QDA Miner v.4.1.21and WordStat 7, both by Provalis Research.First, we classified each individual article at QDA Miner according to four variables: (i) journal title, (ii) date, (iii) country and (iv) language.Afterwards, we sent the corpus to WordStat, which performed the content analysis.In WordStat, we elaborated a dictionary containing the following categories and subcategories 12 : From our original corpus of 7,857 articles, 4,209 had no positive matches with the terms listed in the dictionary and were therefore excluded.The discussion in the following section refers to the analysis performed on the remaining 3,648 articles.The last two columns of Table 1 indicate the new distribution of articles per journal.

Theory
Liberalism was the predominant theory for all studied years, being present in 31.22% of the articles, followed by Realism (26.92%).The third most popular theory, Constructivism, scored significantly less (12.8%) and was not very distant from other second-tiers theories -being actually less popular than Marxism in 2007 and 2008, and Post-Modernism/Post-Colonialism in 2007 and 2014.Feminism was the least frequent theory.The results are summarized in Table 2. Liberalism and Realism were prevalent in all countries, except Uruguay.Second-tier theories, in turn, displayed greater cross-country variation as shown in Table 3 and Figure 1.Liberalism and Realism also had the highest co-occurrence, appearing jointly in 622 cases (Jaccard Index 14 : 0.415).This leads us to believe that the prevalence of these theories is due to the fact that IR articles tend to follow an argument structure which opens up with a contrast between Liberalism and Realism, even when the author intends to employ an alternative approach.
14 Jaccard's Index between two categories is calculated by WordStat as a/(a + b + c), in which a represents the number of cases where both items occur, and b and c represent cases where one item is found but not the other.Perfect similarity between two categories would yield 1, while no intersection between them would yield 0. This is a measure that allows one to interpret how similar are two groups of articles.
What does the field of International Relations look like in South America?

13
The 0.415 JI, the highest ranked in the whole study, means that 41.5% of the appearances of the Liberalism category in the corpus are accompanied by the Realism category, and vice-versa.
A similar interplay might be in place between Marxism and Liberalism (JI: 0.267), reproducing a thesis-antithesis essay style.Constructivism was most strongly associated with Realism (JI: 0.225).Post-Modernism/Post-Colonialism and Critical Theory were both more associated with Marxism, evidencing their indebtedness to this original paradigm (JI: 0.206 and 0.146, respectively).English School was most strongly associated with Realism (JI: 0.156), probably due to the international system/society debate.Feminism seemed to be more isolated, as its association with other approaches was comparatively weaker, being most significantly associated with Post-Modernism/Post-Colonialism (JI: 0.105).
We also had the concern of monitoring the status of Dependency Theory.As this is considered an original South American contribution to the study of IR, we wanted to know whether the maturing of this field of research in the region was accompanied by a comparable development of this particular theory.
By isolating Dependency Theory from the Marxism category, we could see that it has declined in importance over the years, as show in Figure 2. The fact that in total it was present in only 184 articles from a total of 424 Marxist ones indicates that, even within this theoretical frame, Dependency Theory is not a dominant approach.What does the field of International Relations look like in South America?Rev. Bras. Polít. Int.,59(1): e004, 2016 Medeiros; Barnabé; Albuquerque; Lima 14 By performing a correspondence analysis, it was possible to place all analyzed countries in a continuum that showed where each one was located between both epistemic poles.As seen on Figure 3, apart from Uruguay, all countries are placed on the more positivist end of the spectrum and do not have great distance between them, indicating epistemological homogeneity in most South American countries.

Method
Concerning the methods used, Qualitative Analysis was prevalent, being present in 226 articles, while Quantitative Analysis came in second, registered in 176 articles.Only 55 articles made explicit mention of using Mixed Methods.Formal Models were the least common method, occurring on a mere 0.41% of all cases.  of the IR field in South America 17 , and those figures allow us to perceive the magnitude of this oblivion.Anyhow, placing the data in time-series shows us that the number of articles explicitly mentioning methods has grown in the past years, as displayed in Figure 4. Most importantly, the series indicates a steady growth in Quantitative Analysis, which surpassed Qualitative Analysis for the first time in 2014.This could perhaps be just a temporary peak which will recede in the next year, or it could be the turning point for South American IR, signaling a departure from its original humanities-related approaches and a growing preference for statistical methods, as it has been happening with Brazilian Political Science 18 .

Qualitative Techniques
The Qualitative Analysis category was composed of five different techniques.Breaking down this category, we could see that the most popular approach was Historical Analysis, being present in 152 of the 226 articles that used Qualitative Analysis (67.26%).This hierarchy was similar in all countries (except for Venezuela, which had zero qualitative cases), with Historical Analysis accounting for over half of the Qualitative Analysis articles.Uruguay was the only divergent case, which had 33.3% Historical Analysis, followed by 25% Hermeneutics.

Quantitative Techniques
As for Quantitative Analysis, we broke it down into Statistical Methods and Content Analysis.Statistical Methods were the most common technique, accounting for a quarter of the 176 articles using Quantitative Analysis, while Content Analysis only accounted for 7.39%, as shown in Table 8.Even so, on the years 2011 and 2013, Content Analysis was used more often than Statistical Methods.Regarding individual countries, Argentina, Chile and Venezuela did not use Content Analysis at all, while Uruguay had no record of using Statistical Methods.

Area
The predominant area of interest was "Institutions and International Regimes", which was present in over half of the 3,648 articles, followed by "Regional Integration" (43.35% of the cases) and "International Security, Strategic Studies and Defense Policy" (34.81%), as shown in Table 9.For the time period considered, most of the research areas continued to display the same hierarchical importance, with the exception of "Foreign Policy Analysis" and "International Law", which often switched between fourth and fifth place.
The prevalence of "Institutions and International Regimes" can be partly explained by the broad definition we have employed for this area.As Keohane (1984) puts it, institutions are a pattern of activity or a human-constructed arrangement; Krasner (1982) adds that international regimes are principles, norms, rules and decision procedures shared among international actors.Together, these concepts encompass almost every object of research in IR.Source: Elaborated by the authors "Institutions and International Regimes" was prevalent for all countries, with Uruguay and Venezuela showing a larger concentration of their production in that single area."International Security, Strategic Studies and Defense Policy" was the second most important area for Colombia, and third for Brazil, Chile and Uruguay."Foreign Policy Analysis" was third for Argentina and Venezuela.There was a high proportion of co-occurrences between areas, which was expected given the fact that a single article is likely to address one main area in combination with other secondary fields.In particular, "International Security, Strategic Studies and Defense Policy" showed high similarity with a number of other areas: "International Law" (JI: 0.436), "Foreign Policy Analysis" (0.405), and "Institutions and International Regimes" (0.373).Moreover, "Regional Integration" was also close to the areas of "Institutions and International Regimes" (0.356) and "Foreign Policy Analysis" (0.335).

South American IR research templates: commonly associated theories, methods and areas
Though measuring the importance of individual theories, methods and areas is in itself a significant contribution to understanding the IR field in South America, it remains incomplete if no attempt is made to discover how they are combined by publishing authors in their research designs (Lebow and Lichbach 2007;Brady and Collier 2010).
Hence, we verified how often each theory, method and area co-occurred across the corpus.The resulting co-occurrence matrix, displayed in Table 11, shows how frequently all theories, methods and areas appear in combination with one another for all studied articles.We will discuss this absolute frequency in comparison with the similarity measured by the JI.
What does the field of International Relations look like in South America?Rev. Bras. Polít. Int.,59(1): e004, 2016 As shown, all eight theories had high co-occurrences with Liberalism and Realism, followed by Marxism.Methodologically, they overwhelmingly favored Qualitative Analysis.Liberalism and Realism were the theories with more co-occurrences with Quantitative Analysis.Considering the association between theories and research areas, it was possible to see that, in absolute terms, "International Security, Strategic Studies and Defense Policy" is the area of research that co-occurs more often with theories in general, suggesting that articles on this subject are less a-theoretical.
Focusing on individual theories, we can perceive that Constructivism is frequently associated, in absolute terms, with the research areas "Institutions and International Regimes", "International Security, Strategic Studies and Defense Policy" and "Regional Integration".Considering similarity, as measured by the JI, Constructivism is actually more closely related to "International Security, Strategic Studies and Defense Policy" (0.208) and "Foreign Policy Analysis" (0.197).This probably occurs because the two most popular concepts developed in IR Constructivist theory are securitization (Buzan and Waever 2003) and identity (Wendt 1999).
A similar reversal takes place for Critical Theory, which is more associated, in absolute terms, with "Institutions and International Regimes", but according to the JI, to "International Security, Strategic Studies and Defense Policy" (0.105).This is understandable, given Critical Theory's normative concern with how problem-solving theories such as Realism portray war and peace, and therefore proposed alternative, non-dichotomous readings of these phenomena.
Concerning Feminism, it had few co-occurrences since it was overall a marginal theory.It was, nonetheless, related to "International Security, Strategic Studies and Defense Policy", "Institutions and International Regimes" and "International Law".The same hierarchy regarding research areas was present for Post-Modernism/Post-Colonialism, indicating a shared interest structure for authors from those Post-Positivist perspectives.
As the prevalent theory of the corpus, Liberalism had the most co-occurrences across methods and areas.It registered the highest co-occurrences with all methods, including the unpopular Mixed Methods and Formal Models.It was more associated with the area of "Institutions and International Regimes", followed by "International Security, Strategic Studies and Defense Policy" and "Regional Integration".It is important to underscore that "Regional Integration" had the most co-occurrences with Liberalism, and with the accompanying area "Institutions and International Regimes".This is indicative of how Liberalism focuses on economic issues, which is the main variable in most regional integration studies.Also, regional integration is usually associated with institution-building that entangles countries' economies and trade (Deutsch et al. 1957;Haas 1958).
Marxism's JI ranked higher with all methods, suggesting that, although it mentioned methods less frequently than Liberalism and Realism in absolute terms, it had a smaller proportion of articles lacking methodology.
Realism was most strongly associated with "International Security, Strategic Studies and Defense Policy", which was expected, given the former's emphasis on issues of war, peace and What does the field of International Relations look like in South America?Rev. Bras. Polít. Int., 59(1): e004, 2016 Medeiros; Barnabé; Albuquerque; Lima 22 power.It is noteworthy that, from all theories, Realism had the highest co-occurrences with the area "History of International Relations and History of Foreign Policy", which was overall a neglected area.
Concerning individual areas of research, most of them tended to associate more often with Liberalism, Realism, Constructivism and Marxism, in that order of importance."Foreign Policy Analysis" and "International Law" displayed an interesting harmony in that aspect, as they had a close number of co-occurrences with those theories.This is compatible with the behavior of these areas of research across time.They switched between fourth and fifth places often, indicating that they have received the same attention from all theoretical perspectives.
"International Security, Strategic Studies and Defense Policy" had the highest number of cases co-occurring with the Formal, Qualitative and Quantitative Methods; and likewise for theories, suggesting that articles dealing with this topic tend to be more method and theory-aware.
The area of "Institutions and International Regime" was, by a very large lead, most commonly associated with Liberalism.This was expected as those were the overall most popular area and theory, but also given the above mentioned broad conceptualization of institutions and international regimes.
South American IR journals: many areas, some methods, and few theories Having investigated the prevalence of specific theories, methods and areas across countries, we now turn our analysis to the journals.In this section we present a classification of all 35 journals included in the study, so as to arrive at an overview of what are the main epistemological and methodological affinities of each publication, preferred areas of interest, and if any clusters can be spotted.
We classified the journals according to the percentage of Positivist vs. Post-Positivist articles (Y-axis) and to the percentage of Qualitative vs. other methodologies (X-axis).The result is displayed in Figure 8, which creates four possible profiles for publications.It is clear that the majority of IR journals in South America employs some form of Positivist theory.Only two journals from the 35 were predominantly Post-Positivist.Method-wise, we can see greater diversity, though Qualitative Methods are still dominant (21 vs. 14 journals with more than 50% of their articles using Qualitative Analysis).If we focus on IR-only titles (Colombia Internacional, Contexto Internacional, Estudios Internacionales, Rev. Bras.Pol.Int., Rev. Rel.Int.Estr.y Seg.), we will see that they cluster on the upper, center-to-left part of plot, indicating they are predominantly Positivist and tend to have a greater share of non-qualitative articles.Therefore, the most populated quadrant is that of Qualitative and Positivist journals.It is noteworthy that the Post-Positivist and Quantitative quadrant is completely empty, which could be explained by the primacy given by Post-Positivist theories to discourse, representation and ideas -variables more promptly accessible by qualitative research.In order to verify what the journals' main areas of interest are, we inserted the information from this scatter plot, plus data on the prevalence of each area of research for all journals, into a radial coordinates plot, producing Figure 9 below.The six research areas are placed in the outer circumference of the graph, while the 35 journals are represented by the bubbles plotted inside the circle.Journals which have the majority of their articles focused on one peculiar area will be more distant from the center and closer to the spot on the circumference corresponding to that respective area, while journals that have a more balanced distribution of articles will be closer to the center.The size of the bubble indicates epistemology: the more positivist a journal, larger the bubble area.Shading, in turn, indicates method: the more qualitative a journal, the darker the shade.Most of the journals tend to be concentrated in the center of the circle, meaning that they have a rather balanced interest in different areas.From the six research areas, "International Institutions and Regimes", "International Security, Strategic Studies and Defense Policy" and "Regional Integration" form a triad which draws the majority of journals.
A rather balanced range of interest seems to characterize South American IR journals.The ones which are farther apart from the center, and are therefore more mono-thematic, are from IR-related areas: International Law, Civilizar, and Ciclos en la Historia, la Economía y la Sociedad.Many of the plotted journals publish on Social Sciences or Humanities in general, which might explain the overall balance of interest.If we focus again on the five IR-exclusive titles, we can see that they too are close to the center, and slightly more drawn to the areas of "International Security, Strategic Studies and Defense Policy" and "Foreign Policy".This is perhaps indicative that those areas are less accessible to more interdisciplinary journals, which can instead focus on integration and institutions, which are phenomena more open to neighboring disciplines.
Regarding methodology, qualitative and quantitative journals are scattered, though there is a slightly greater concentration of quantitative-oriented journals in the center and lower-half of the circle.

25
As most journals use Positivist theories, the area of the bubbles in the plot does not vary greatly.Nonetheless, one finds smaller bubbles, indicating more Post-Positivism, in the center of the circle and in the direction of the areas "International Law" and "Institutions and International Regimes".Some of the smaller bubbles are also lighter in color, indicating that journals that have a more balanced methodological preference will also be more receptive to different epistemological viewpoints.
Country-wise, we can see some affinities between countries and areas.In comparison to other countries, Argentinean titles tended to emphasize on the "International Security, Strategic Studies and Defense Policy", "Foreign Policy Analysis" and "History of International Relations and History of Foreign Policy" triad.Colombian journals were more drawn to the "International Institutions and Regimes", "International Security, Strategic Studies and Defense Policy" and "International Law" portion.Chilean journals, to "International Security, Strategic Studies and Defense Policy" and "Foreign Policy".

Conclusion
Under the Lakatosian approach proposed by this article, in general, the IR field in South America looks like a North American mimetic ersatz, i.e., it has been built under the Northern/ Western influence despite its own idiosyncrasies.Maybe because South American authors understand that those idiosyncrasies are not relevant and in the majority of cases the Realism and Liberal paradigms are able to explain how nations behave (Henkin 1979).
As a matter of fact, considering the three dimensions of analysis this article proposedtheory, method and area -one cannot identify, apart from a positivist and qualitative preference, any major genuine trend of South American IR, nor any significant attempt to contribute to the new "Global IR" ideas (Hurrell 2015).
Concerning theories, the study of our sample demonstrated that Liberalism was predominant, followed by Realism and Constructivism, the latter being close to second-tiers theories: Marxism and Post-Modernism/Post-Colonialism.By isolating Dependency Theory from Marxism, we could see that its importance has declined over the years, and that even within the Marxist tradition it was not a dominant approach.
Regarding methods, Qualitative Analysis was prevalent, although it was surpassed by Quantitative Analysis for the first time in 2014.Whether this is a momentary spike or a turning point in South American IR remains to be observed in future assessments.Only 12.94% of all articles mentioned methods, indicating a widespread methodological unawareness.
The predominant areas of interest for the studied period were "Institutions and International Regimes", "Regional Integration" and "International Security, Strategic Studies and Defense Policy", which was already expected, as these areas, in fact, are the main ones studied by the three most used theories mentioned above.
We have also found common patterns which South American IR researchers use when combining theories, methods and areas.All eight theories had high co-occurrences with Liberalism and Realism, followed by Marxism.Methodologically, they favored Qualitative Analysis.In absolute terms, "International Security, Strategic Studies and Defense Policy" is the area of research that most cites theories in general, meaning the production on this topic is the least a-theoretical.
From our classification of all the 35 journals studied -according to epistemologies and methodologies -, it is clear that most publications lean towards Positivism.There is some methodological diversity, though Qualitative Methods are prevalent, and examples of Quantitative/ Post-Positivist journals are inexistent.IR-exclusive titles, in turn, are predominantly Positivist and have a comparatively larger share of non-qualitative articles.Most journals have a balanced interest in different areas.From the six research areas, "International Institutions and Regimes", "International Security, Strategic Studies and Defense Policy" and "Regional Integration" attract the majority of titles.
Our findings are complementary to that of previous empirical researches that have also tried to characterize the IR field, although using other approaches, such as surveys and analyses of IR courses curricula.Froio (2012, 16) indicated an increase in "Security and Defense" analyses in Latin America.Additionally, Tickner (2013) identified Realism and Liberalism as the predominant paradigms studied in IR undergraduate courses in Argentina, Brazil, Colombia and Mexico, and Marxism as a minor current in Latin America (a second-tier theory, in our results); and notices a clear preference for qualitative over quantitative methods.She also underlined that "Regional Studies" are the most cited area of interest in the aforementioned countries (idem, 22), which is similar to our finding that "Regional Integration" is the second most frequent area in number of articles and the most frequent in number of matching terms.Though the author indicated a certain balance concerning the option on a positivist or post-positivist approach (idem, 26), our results show a clear predominance of positivist epistemology.Other studies on the IR field in individual South American countries also corroborate our findings (Merke 2005).Since we approached the IR field via the publication side, and not the education one, differences in the figures are expected and are actually quite illuminating on the congruities and discrepancies between what is taught and what gets published.
The findings of this paper do not reveal why South American IR looks like European and North American IR, as its aim was to offer a descriptive view, not an analytical one.Nevertheless, they show this similarity is real.Whether this likeness shows some improvement of IR theories developed in the region, it remains to be seen.It is not possible to evaluate what the data collected means for theory-building in a Lakatosian way (Elman and Elman 2003).A future research agenda should seek to investigate if this situation is a conscious option or, on the contrary, is the result of the incapacity to produce and maintain new epistemologies.A broader analysis, with a greater sample of journals from countries that had few titles matching our criteria in this study, such as Uruguay and Venezuela, could also increase the accuracy of the picture presented here.

Figure 1 -
Figure 1 -Theory frequency by country

Figure 2 -
Figure 2 -Number of articles citing Dependency Theory

Figure 6 -
Figure 6 -Qualitative technique frequency per country

Figure 8 -
Figure 8 -Scatter plot of 35 South American journals classified according to their epistemology (% of positivist articles) and method (% of qualitative methods)

Table 1 -List of all countries, journals and number of articles in the corpus
The time span analyzed went from January 2006 to December 2014.We arrived at a corpus of 35 journals from six countries: Argentina, Brazil, Colombia, Chile, Uruguay and Venezuela.The total number of journals and downloaded articles are summarized in Table1below. 9 Medeiros; Barnabé; Albuquerque; Lima 8 Source: Elaborated by the authors.9 The complete description of the reasons for choosing this period, the selection procedure applied to the journals, and method of analysis can be found on ResearchGate, "Methodological appendix to paper accepted for publication in RBPI -Revista Brasileira de Política Internacional, July 2016 edition", DOI: 10.13140/RG.2.1.4441.2561.What does the field of International Relations look like in South America?Rev. Bras.Polít.Int., 59(1): e004, 2016 Medeiros; Barnabé; Albuquerque; Lima 9

Table 2 -Overall frequency of theories
Source: Elaborated by the authors 13 The full list of terms and context rules used for each category of this dictionary can be found in ResearchGate, DOI: 10.13140/ RG.2.1.4441.2561.What does the field of International Relations look like in South America?Rev. Bras.Polít.Int., 59(1): e004, 2016 Medeiros; Barnabé; Albuquerque; Lima 12

Table 3 -Theory frequency by country
Source: Elaborated by the authors

Table 4
allows us to see that the total number of articles mentioning any method at all is rather small if compared, for instance, to theories.Even the most frequent method is present in only 6.2% of all considered articles.The lack of method-awareness is an enduring shortcoming What does the field of International Relations look like in South America?Rev.Bras.Polít.Int., 59(1): e004, 2016 Medeiros; Barnabé; Albuquerque; Lima 15

Table 5 -Method frequency by country
Source: Elaborated by the authors 17 Not only in South America.As Sprinz andWohlinsky-Nahmias (2004)argue, the history of quantitative methodology in IR is very similar to its history in political science, with very few adepts at the beginning, but with a consistent and growing body of knowledge on the long run.18Onthisdebate,although Soares (2005) sees a critical condition when he examines political science development and methodological training in Brazil,Barboza and Godoy (2014)show a much more optimistic ten-year scenario evolution, noting that several graduate programs have been teaching specific courses on methodology, most of them on its quantitative branch.What does the field of International Relations look like in South America?Rev. Bras.Polít.Int., 59(1): e004, 2016 Medeiros; Barnabé; Albuquerque; Lima 16 Figure 5 -Method frequency by country Source: Elaborated by the authors

Table 10 -Area frequency per country ARGENTINA BRAZIL CHILE COLOMBIA URUGUAY VENEZUELA
Source: Elaborated by the authors What does the field of International Relations look like in South America?Rev. Bras.Polít.Int., 59(1): e004, 2016 Medeiros; Barnabé; Albuquerque; Lima 19 Figure 7 -Area frequency per country Source: Elaborated by the authors

Table 11 -Co-occurrence between theories, methods and areas for all articles
Source: Elaborated by the authorsWhat does the field of International Relations look like in South America?