Local and global communications in Chilean social science: Inequality and relative autonomy

Building ‘central science’, the mechanisms for its production and its effects in Latin America

Science, in all its areas, has had a transnational orientation since its beginnings with modernity, gathering knowledge from diverse parts of the world and it has claimed the universality of the knowledge generated – although in practice there has been a clear predominance of the knowledge originating in the central countries, without a proper validation in other places of the world. This predominance is strengthened by the peculiar characteristics of the devices employed in recent decades to select and to regulate scientific production.

Together with the advances made in defining methodological procedures for evaluating and proving its hypothetical statements, science has had to design institutional mechanisms for communicating, evaluating and selecting scientific communications. The first printed bibliography –a register of about 10,000 books– was made in 1545, and putting it together took years of work (Burke, 2002). The publication in journals of the results of scientific undertakings began in 1665. The scientific societies and academies that arose during the 17th century were crucial for the invention of the scientific journal, the use of which began to expand as a communication medium in replacement of the letters, treatises and books which were the common way of communication until then. The first two journals appeared in 1665, one in England – the famous Philosophical Transactions – and another in France (Merton, 1973). The generalized practice of using footnotes comes from the 17th century (Burke, 2002) and only in the 19th century did the format of scientific papers become more or less established and generally adopted, including the peer review system and the standardization and general usage of the academic apparatus of notes and quotes (Merton, 1973). Scientific societies, congresses, academic journals, the peer review system and bibliographic references are some of the procedures taken for granted today but that have only gradually been refined and stabilized over four centuries.

Following the Second World War, at a time of tremendous growth in scientific activity, a last great device appeared that would have powerful effects on the structure of global science in the decades to come: the bibliometric register of articles and authors, and of the quotes that refer to them in a body of journals selected as being the best known. It is a device supported by the mechanisms already operative in a scientific journal, particularly the bibliographical references and the peer review procedure, and by a prestige structure that had taken form and was recognized in some fields of science. There were several other trials attempted previously, but the Scientific Citation Index (SCI), developed by Eugene Garfield in the US, finally imposed itself around 1961. It was based on an automated procedure that does away with human classification, and was managed by a private institution with commercial aims, the Institute for Scientific Information (ISI). Originally concentrating on the biomedical field, it rapidly expanded to involve other disciplines, and in 1972 the Social Science Citation Index (SSCI) appeared. Garfield’s institute would be bought in 1992 by Thomson Business Information (now Thomson Reuters), another private and multinational corporation, based in Canada and the US, that strengthened the services provided by the index and moved its operation onto the Internet, multiplying its earnings (De Bellis, 2009). These indexes became the main tools for regulating science around the world, being the main ways to manage the relevance of the articles and the levels of prestige of scientists, enjoying an undisputed preeminence through the end of the 20th century. They are an answer to the enormous proliferation of publications, which would become unmanageable both for the scientific community and other users of scientific information if they could not rely on these filtering and ranking mechanisms of published literature.

When these indexes were put together, there was no great effort made to achieve within them a representation of the various regions of the world. Selection was made with a view towards the central countries, and particularly from an Anglo-Saxon point of view. This is reflected in the predominance of the English language: between 1998 and 2007, 94.5% of the articles in the SSCI were written English, and only 0.4% in Spanish, for example (Gingras and Mosbah-Natanson, 2011). This involves the benefit of having a lingua franca for science but it has an asymmetrical cost for access (Ammon, 2011). As regards regional representativeness, during the 1970s there was next to no presence of Latin America or Asia. There was only one journal from Chile among all of the scientific disciplines. In later years there has been an effort to increase the diversity of origins, but despite all of that, there is still a great concentration. In 2010, the US and Europe were the source of 84.3% of all of the journals included in the SCCI, with 49.5% from the US, 23.7% from the United Kingdom and fewer than 3% from Latin America (Rodríguez, 2010).

In 2004, the Elsevier publishing company based in Europe (Amsterdam) launched a new database and citations index, Scopus, which broke the monopoly of the North American Thomson Scientific. ¹ At the same time, other more open search engines like Google Scholar gained relevance. This caused an overall increase of the number of publications included in these indexes, although Scopus, with its commercial outlook, the greatest change that produced was the increase in the proportion of European publications (Guédon, 2011).

These indexes and citation databases that mix the selectivity characteristic of pre-existent scientific recognition with the selectivity criteria that arise from the geopolitical and sociocultural positions of those who construct the index, provide orientation for the searching and reading of scientists and for the making of institutional decisions, particularly those of librarians in purchasing scientific publications (Vessuri, 2008). They generate a collective ‘Matthew effect’, not only referring to the most visible authors who are recognized as prestigious, as described by Merton (1973), but also referring to visible journals recognized as prestigious. As a result of their selection in the index they concentrate the preferences for publication and reading, to the detriment of others that are not included. This produces a massive ratification of the privileged status of these journals. The initial qualification of the journals is self-validated. Those that are included, and thereby made visible, are more frequently cited, and those that are excluded and therefore less visible do not attract submissions for publication or citations, at least not as a consequence of the effect propagated worldwide that the index produces. Even more, the exclusion extends to entire regional clusters of journals and the sub-representation becomes consolidated, resulting in a configuration of a global science that is markedly associated with the central countries, particularly the Anglo-Saxon countries, as a performative effect of the mechanisms registering publications and citations. SSCI and Scopus shape this central science; the very make-up of the devices and the way they operate cause this shaping to stabilize and reproduce itself.

Since the 1990s, as a reaction against this scientific marginalization, several indexes and journal databases have been created in Latin America: Redalyc, Latindex and Scielo. They try to articulate and increase the visibility of the regional production in the social sciences. Scielo, also including Spain and Portugal, promoted by the Brazilian government, is the index that has achieved the greatest recognition from the academic world. In second place is Latindex, which operates from Mexico and is chosen above all for its most demanding version: Latindex Catálogo (Guédon, 2011; Rodríguez, 2010).

The scientific institutions in Chile, in the area of the social sciences, have assigned full validity to the central indexes, especially to the SSCI, commonly referred to as ISI, and to Scopus, assuming them as objective and unquestionable standards of quality. Scielo is recognized as belonging to a secondary category and Latindex to a third rank category. The National Council for Science and Technology (Conicyt), which is the primary source of government support for research in the social sciences, currently defines that a requirement for the approval of a research project proposal is the achievement of at least one ISI publication. In 2007 Conicyt imposed this condition, whereas before it had only been a recommendation. ² Furthermore, within the competition among social science research projects, operated by Conicyt, researchers receives almost three times as many qualifying points for an ISI or a Scopus publication than for a Scielo publication, points that are often decisive factors in the overall approval decision. And it is worthwhile pointing out that Chile has a very limited number of ISI journals: none in sociology, one in political science and two in anthropology, which undoubtedly puts pressure on people to publish abroad. On the other hand, some of the better recognized and most read journals in Chile are neither ISI nor Scielo indexed. This is the case of Estudios Públicos, which in a survey that I applied among 37 outstanding researchers of the national academic world, was considered to be the most highly valued journal. In spite of this fact, in the competition for project funding, it receives a third or fewer points than a publication in an ISI or Scopus journal.

Something similar occurs with the evaluations made in the country of the institutional productivity at universities or other scientific centers. ³ The most frequently used indicator for ranking and making comparisons is the number of ISI publications. As a consequence, an increasing number of universities, particularly new private universities, founded since the 1980s, are now offering economic incentives for achieving that type of publications, with amounts of money that, in some cases, are close to a month’s salary, being extremely attractive, given the tight economic situation of scholars in Chile. Some universities include this kind of incentive for Scielo publications, repeating the stratified pattern, by offering sums that are between half and one-third of those offered for ISI articles. Furthermore, these publications, especially those in ISI, are considered to be a privileged indicator of individual productivity and so they influence professional advancement. In that way, Chilean scientific institutions themselves, wherein the very researchers participate, promote the reproduction of this stratified structure of a central social science, more highly valued, and a peripheral science.

What, then, are the effects of this structuring of global science, and the way of embedding Chilean social sciences in it? It is clearly a type of communication that has been established on a foundation of inequality, for which the rules for the selection of journals are set by scientific regulatory corporations located in the central countries and the selection of articles is defined by evaluators who are principally from those same countries. Local social science competes at a disadvantage, and nor do its local academic institutions stimulate, as we have seen, publication in local journals, no matter how well qualified they are.

When we look at scientific communications from the point of view of the databases and indexes of centralized science, the following results emerge (see Table 1). Scientific production in Latin America barely exists for the central countries: Europe and the US make fewer than 1% of their citations to authors from that region. For its part, Latin America orientates 90.1% of its references to those countries, a situation that is repeated in Asia and Oceania. This is to say, if we make an evaluation on the basis of the exchange, then we have a total imbalance.

OPEN IN VIEWER

Table 1.

The orientation of the social science citations worldwide, 2003–2005 (in the 200 most cited journals, in percentages).

		Country or region making the reference
		USA	Europe	Latin America
Country or region referred to	USA	78.1	47.9	56.2
	Europe	20.4	50.3	33.9
	Latin America	0.2	0	6.9
	Rest of the world	1.3	1.8	3.0
		100.0	100.0	100.0

Source: SSCI, Gingras and Mosbah-Natanson (2011).

Can it be concluded from this that there is a cognitive or intellectual dependence? Is this a matter of coloniality of knowledge? Over the past decades, a thread of critical thought has appeared regarding epistemic or cognitive dependency (Mignolo, 2004), the colonial nature of knowledge (Lander, 2004) and, in general, about the problem of the relationships between the central and the peripheral countries as regards the generation of knowledge (Alatas, 2003; Burawoy et al., 2010; Connell, 2007; Mignolo, 2003, 2004; UNESCO, 2011). From this point of view, the problem is the excessive and inappropriate use of theoretical and conceptual categories generated in the central countries, without giving the necessary attention to their coincidence with local realities in the peripheral countries, and without a local production able to generate its own stream of knowledge, which could develop a cognitive relation, on an equal standing, with the core countries, and, furthermore, which could overcome the marginality of autochthonous currents of thought.

Although science supports itself upon a global accumulation of knowledge and has an inherent pretension of universality, in the social sciences, local references are extremely important, in a way that is not present in other scientific disciplines. In this regard, all the social science disciplines involve important and meaningful configurations of knowledge associated to local areas (e.g., sociological knowledge about France, political science knowledge of certain realities of the United States, and so on). Even in the field of economics, which certainly has a self-image of being global, some of its theoretical constructs and statements contain marked national differences, as shown with conspicuous precision in the study made by Fourcade (2009) that compares the USA, the UK and France; the flags of the universal nature of knowledge, so ardently waved by economists, cause those differences to pass unnoticed. Such cognitive differences are associated with the peculiar characteristics of the institutions that deal with the production and transmission of knowledge, and with their ways of entanglement with the rest of society.

To analyze the situation of cognitive dependence in a particular country, like Chile, my assertion is that it is not adequate to judge the direction of scientific communications and to ponder the possible cognitive dependence with regard to the central countries by limiting our observation to what happens in the central indexes and citation databases. It needs to be taken into account that this Latin American social science, which makes fewer than 10% of its references to Latin American countries (Gingras and Mosbah-Natanson, 2011), represents but a small fraction of the social science production in this region, and that the authors of this group, selected by central indexes, are precisely those involved in establishing a dialogue with the production of the central countries. So it is necessary that we analyze what is going on with the rest of the national production.

That is what I shall do here with regard to Chile. I choose this country to examine ideas about cognitive dependence. In global economic and power relations, Chile, although subordinate to core countries, has an intermediate position and has been considered a semi-peripheral country (Babones and Alvarez-Rivadulla, 2007). It can be suggested that the results found regarding Chile could be hypothetically extended to other semi-peripheral countries, at least from the region, such as Brazil, Mexico and Uruguay.

I will analyze, then, what direction scientific communications take when we base our analysis on the universe of social scientific production being generated in the country, and how strong the orientation is towards the central countries as opposed to the orientation towards Chile itself and the Latin American region.

As part of this analysis and by paying attention to the assertions about cognitive dependence, I also question what is happening regarding theory. Theory is at the very center of the making of scientific observation and of the way reality is interpreted and explained. So I seek to identify the sources of theory used in the country – in terms of the theorists referred to by national researchers and with whom they establish a dialogue in their publications – and who are the most central theorists in the field. At the same time, I ask if there are local theorists who are relevant within the network of theoretical connections. Once that has been analyzed, the question is what the findings have to say about the eventuality of cognitive dependence.

Methodology

For this analysis of the scientific communications as observed from the Chilean field itself, I have taken three basic disciplines of the social sciences: sociology, which has achieved notable progress in becoming established institutionally in this country during the 1960s; political science, at first with strong links with and dependence upon sociology but gradually consolidating its own institutionality since the 1980s; and anthropology, which made early progress in the country, at the beginning of the 20th century, although its institutionalization and development have been slower (Fuentes and Santana, 2005; Garretón, 2005; Palestini et al., 2010; Ramos, 2005; Ramos and Canales, 2009; Rehren, 2005). Economics was not included because for Chilean scientists, at least in the period studied, it is conceived as a separated field, whereas sociology, political science and anthropology are viewed as part of the same field. This is similar in the social sciences of other Latin American countries (Trindade, 2007). The discipline of economics, with all the complexities involved, would require its own research.

Research into scientific communications has habitually considered only articles published in journals, because of their accessibility. This is valid in sciences like chemistry or biology, where most of the results of research effectively appear in journal articles, but in the social sciences, the pattern of publications is different. In particular, sociology has been a culture of the book, as indicated by Clemens et al. (1995) writing about the US, and there are estimates that suggest that ‘between 40% and 60% of the literature of the social sciences is comprised of books’ (Archambault and Larivière, 2011: 264). So, in order to avoid distortion, I have included all the formats that are relevant to social scientific knowledge dissemination in the country: books, book chapters, journal articles and publicly available working papers.

I concentrated attention on publications that make up the core of scientific activity: publications that report research results, which is to say those that involve the generation of new knowledge through systematic research, empirical or theoretical. Therefore, I have excluded book reviews and opinion pieces, and texts that do not meet the minimal requirements for selection in a social science journal of recognized quality. To select a text, its author – or at least one of the authors, in the cases of multiple authors – should have undergraduate or graduate education in one of the three aforementioned disciplines. I included not only researchers who were Chilean citizens, but also those who were residents in Chile during the period of study and who were active participants in their disciplines.

The research tried to cover the universe of texts that fulfilled the requirements mentioned and that were published between 2000 and 2006. The research team gave special attention to gathering material produced outside of the central metropolitan region, and we traveled to regional centers looking for texts. We also reviewed a variety of series of working papers and institutional publications, both private and public, and from international organizations located in Chile, and we requested texts from those researchers who might have copies of those that we were unable to locate. Regarding publications made in foreign journals, we reviewed ISI and Scielo social science journals, and looked for Chilean authors who wrote in them. All of this was a demanding effort made to achieve a maximum coverage, so that those texts that were finally not identified nor found would have to represent a very small fraction of the whole body of work, and there are no reasons for thinking that there was any bias involved in their being excluded; their distribution must be random. The final corpus comprised 479 texts.

With two other researchers, we read each text, reviewed it and made its characterization based on a variety of features, one of which, pertinent to this article, was the intended destination of knowledge. We determined the principal destination for the knowledge generated considering what was said explicitly in the texts themselves and through inferences based on available data.

Furthermore, we recorded all the bibliographical references contained in each text. After long and very time-consuming work, we obtained a total of 21,787 bibliographic references, and we specified all of the characteristics of each one.

Among the references we distinguished those that cite theorists, understood as authors whose work reaches sufficient levels of abstraction and generalization articulating a hypothetical argumentation with sufficient coherence, consistency and originality. To study the connections with theorists and the networks that involve them we applied social network analysis (Degenne and Forsé, 2004; Scott, 2000; Wasserman and Faust, 1994). Using the logic of co-citation analysis we considered that a connection between two theoretical authors occurs when they are both cited in the same text of the corpus (Gmür, 2003). ⁴

Local and global dimensions of social science in Chile

Table 2 presents the distribution of the publications effectively found and studied, according to discipline of the authors and format of the publication. These figures confirm the importance of books in Chile as media for transmitting knowledge in the social sciences, especially in sociology. This reveals the degree of distortion that may result from an analysis of publications in the social sciences that does not take this format into account. The book is a cognitive product with manifestly different characteristics and with capabilities that an article has not, that are outstanding in the field of the social sciences. In order to include descriptive stories and explanations of some complexity in society, with an integrative aim, supported by an adequately developed line of argument and backed up by a store of empirical material, a journal article is insufficient. There are numerous works in the social sciences that have made considerable impact in Chile that have been published as books. I can mention: Chile: The Anatomy of a Myth, by Tomás Moulian; Chilean Identity, by Jorge Larraín; The Paradoxes of Modernization, by the UNDP; Culture and Modernization in Latin America, by Pedro Morandé; and The Shadows of the Future, by Norbert Lechner. The impact achieved by such works would be unimaginable in the format of a journal article. One cannot find journal articles of comparable levels of impact. On the one hand, those formulations are not condensable into the tight space allowed for an article; and on the other, academic journals in Chile are of a very restricted level of circulation within the academy and they are not generally able to attract a public from outside academia.

OPEN IN VIEWER

Table 2.

Distribution of publications by format and discipline (in percentages).

Format	TOTAL	Disciplines
		Sociology	Political science	Anthropology
Journal article	42.4	34.0	52.1	53.8
Book or book chapter	35.9	45.2	23.6	25.5
Working paper or similar format	21.7	20.8	24.3	20.8
	100.0(479)	100.0(268)	100.0(165)	100.0(106)

Note: In some cases, there have been multiple authors and a text was classified in more than one discipline, so the total of the articles by discipline is greater than the sum of the articles.

Of the total of the texts found, only 7.6% had been published abroad. One might suppose that it is basically over this small group that the central countries would have exercised an influence in defining their research agenda. Of these texts, located in the sphere of central science, a few also occupied positions of relevance in national publications and citations, whereas the others were located in a second line of citations and recognition. On the other hand, nearly half of the most frequently cited authors in the country have not been published abroad. That is to say, in the social sciences, international prestige does not coincide with national prestige. Something peculiar to those who are better players of the international game is that most of them have relationships with institutions in the central countries, which facilitates their maintenance of personal contacts and a physical presence in those countries, and not merely a connection by way of their published writing.

To appraise the relative importance of local and global dimensions of Chilean social science, the analysis of the distribution of bibliographic references provides us with a measure of the degree of attention that social scientists give to local production of knowledge – either to the national or to Latin American production – in contrast with the attention given to knowledge generated in the central countries; this allows us to determine the direction textual communications take. Thus, we have classified the 21,787 identified references according to the cited author’s country. The categories applied are: (1) Chile; (2) other Latin American country; (3) the USA (and Canada, although there are few references to this country); (4) the UK (mainly England ); (5) France; (6) Germany; (7) Spain; (8) other European country (Belgium, Italy, Poland, Austria, Sweden, etc.); (9) other country (India, Japan, Israel, China, Australia, Singapore, etc.); (10) several countries for the same reference or unidentifiable country. The resulting distribution of the references is presented in Table 3, which also groups them according to the discipline of the researcher making the citation.

OPEN IN VIEWER

Table 3.

Distribution by country of the bibliographic references, according to the discipline of the citing author (distribution of the cited authors, in percentages).

Country of origin of author cited	Discipline of the citing Chilean author			TOTAL
	Sociology	Political science	Anthropology
Chile	44.6	39.2	43.2	42.6
Other Latin American country	10.7	10.2	12.7	10.8
USA/Canada	14.1	27.8	11.9	18.5
UK	4.6	6.0	5.3	5.2
France	5.6	2.7	7.0	4.8
Germany	6.3	4.0	4.2	5.3
Spain	5.6	4.5	7.4	5.5
Other European country	3.0	3.8	2.6	3.2
Other country	0.4	0.8	0.6	0.6
Several or unidentified countries	5.0	1.0	5.1	3.6
TOTAL	100.0(11,582)	100.0(7369)	100.0(2836)	100.0(21,787)

It can be observed that Chilean social science shows a clear local vector: 42.6% of the references cite Chilean authors and allude to discussions about the country. There is a clear orientation to the social problems that are of local concern: social inequality, poverty, educational problems, the evaluation of public policy, social movements, gender, etc. and the knowledge generated has a significant orientation towards local audiences.

If we consider the destination for which the work was intended, apart from the academic community itself, the other large destination for knowledge is the state. About 40% of all the production of the three disciplines studied has that destination, whether because of demands coming from the state itself, or because they are the initiative of other institutions – universities, non-governmental organizations or international organizations – which seek to influence in the definition of policies, programs or other governmental decisions. Sociology, particularly, demonstrates a strong connection with the state apparatus: nearly half of its production is interconnected with the state or oriented in its direction. Governmental organisms such as the National Institute for Youth (INJUV) or the National Service for Women (SERNAM) are frequently demanding social science research. On the other hand, 18% of the production is oriented towards civil society entities (social movements, NGOs, political parties, etc.).

As we see in Table 3, the international or global direction is also very important in the communications Chilean social scientists made. Particularly important are those communications that have to do with the production of the central countries (42.5% of all the references), the USA appearing as the main pole of attraction.

Comparing disciplines, political science appears as the most focused on the global orientation: 48.8% of its references are made to central countries (vs. 39.2% from sociology and 38.2% from social anthropology). In this discipline also noteworthy is the greater relative importance assumed by the US. It is, moreover, the discipline in which the members have a greater number of publications in foreign journals. Thus, political science is the discipline that now appears to be the most internationalized. ⁵

The regional dimension – references to knowledge generated within Latin America – has a reduced presence: only 10.8%. The field of social science in Chile shows little interest in Latin American production. We do not have comparative information for periods in the past for Chile, but I would think that the proportion of references made to Latin America has been decreasing since 1970s, when there was much more attention to the region. In fact, according to an analysis for the entire region, using the SSCI database, the citations from Latin American authors to other Latin Americans have been declining in number: in the period 1993–1995 they corresponded to 11.7% of total citations, while between 2003 and 2005 this figure dropped to 6.9%, giving way to a greater proportion of connections with the US and Europe (Gingras and Mosbah-Natanson, 2011).

Regarding the inequality between central science and regional science derived from the operation of indexing and ranking systems, the variety of forms of pressure and institutional incentives that place greater value on the central standards determined by ISI and Scopus have been turning researchers towards the publications that are blessed by said indexes, and that seems to be setting the future tendency. In the case of sociology, during the decade of the 1990s the average of ISI publications was 1.5 articles per year (Farías, 2004); in the period 2010–2012 the rate is 15.3 by year (Web of Science). The Scielo and Latindex Catálogo publications have also increased in number but well below that rate of growth.

The strong institutional preference for prioritizing ISI publications can increasingly tilt the balance towards global connections, especially Anglo-Saxon. In fact, in ISI articles the references to national authors are substantially lower, still much lower than those in books, book chapters and articles with Scielo or Latindex indexing, or articles without indexing.

This value given to the central standards, which leads scientists to seek publication at the international level, especially in English, causing negative impacts on national journals and the publication of books, is constantly being criticized by scholars and local authorities, but universities and scientific institutions apply those standards of evaluation because they are broadly legitimized and have become part of the rules of the game.

The structure of communications with theorists

To investigate the connections with theorists, I began by reviewing the existence of references to a list of 120 internationally recognized theorists, and found that, from the total of 21,787 bibliographic references, 10% were referred to them: 11.4% in sociology, 7.5% in political science and 10.5% in anthropology. ⁶ There were 50 theorists who got more than 10 references in the whole field, but in each discipline about 40% of references were concentrated on five main authors (52.4% in anthropology, 39.9% in sociology and 37.8% in political science). See Table 4, which presents the 25 most cited authors in the field.

OPEN IN VIEWER

Table 4.

Internationally renowned theorists most cited in the field (the top 25, in percentages).

	Author	Country	Discipline of the citing researcher			TOTAL
			Sociology	Political science	Anthropology
1	Luhmann, Niklas	Germany	11.9 (1)	0.5	13.2 (2)	9.4
2	Bourdieu, Pierre	France	7.0 (3)	1.0	16.5 (1)	6.9
3	Giddens, Anthony	England	7.1 (2)	2.7	7.4 (4)	6.1
4	Habermas, Jürgen	Germany	6.6 (4)	4.4	6.6 (5)	6.1
5	Beck, Ulrich	Germany	6.3 (5)	1.7	3.3	4.8
6	Foucault, Michel	France	2.7	2.4	8.7 (3)	3.5
7	Touraine, Alain	France	4.5	1.5	0.8	3.3
8	García Canclini, Néstor	Argentina	3.1	0.2	5.4	2.8
9	Mainwaring, Scott	USA	0.1	11.2 (1)	0.0	2.7
10	Weber, Max	Germany	2.7	3.7	0.4	2.6
11	Sartori, Giovanni	Italy	1.0	5.9	0.0	2.0
12	Bauman, Zygmunt	Poland / England	2.4	1.5	0.8	2.0
13	Huntington, Samuel	USA	0.5	6.8 (3)	0.0	2.0
14	Parsons, Talcott	USA	2.9	0.2	0.0	1.9
15	Lipset, Seymour	USA	0.0	7.8 (2)	0.0	1.8
16	Maturana, Humberto	Chile	2.0	0.0	3.7	1.8
17	O’Donnell, Guillermo	Argentina	0.7	5.6	0.0	1.8
18	Castel, Robert	France	1.6	1.0	2.1	1.5
19	Nohlan, Dieter	Germany	0.2	6.1 (4)	0.0	1.5
20	Geertz, Clifford	USA	1.0	0.0	5.8	1.4
21	Lijphart, Arend	Holland	0.1	5.9 (5)	0.0	1.4
22	Martín-Barbero, Jesús	Spain / Colombia	2.2	0.0	0.0	1.4
23	Marx, Karl	Germany	1.7	0.7	0.0	1.3
24	Castoriadies, Cornelius	Greece / France	1.5	0.5	1.7	1.3
25	Willke, Helmut	Germany	1.9	0.0	0.0	1.2
			(1092)	(409)	(242)	(1743)

There are strong similarities between sociology and anthropology regarding the authors with the highest quantities of citations; of the top five in each discipline, four are matched. Political science, in contrast, markedly differs from that pattern of authors. Sociology and anthropology are dominated by the ‘big theorists’ (Bourdieu, Luhmann, Habermas, Giddens), while in political science middle-range theorists are emphasized, authors that are more closely tied to empirical research on the basis of which they establish abstract generalizations (Mainwaring, Lipset, Huntington, etc.). Moreover, while in sociology and anthropology the major countries of origin of the theorists are European – Germany, France and England – in political science, the USA dominates: the three most cited authors in this discipline are of US origin.

Among the most cited theorists only 12% are of Latin American origin or have had prolonged stays in the region and show a close relationship with it: García Canclini, Maturana, O’Donnell, Martín-Barbero, Germani, Cardoso, Laclau, and Hinkelammert. One of them – Maturana – additionally, is not properly a social scientist and does not make a reflection focused on the social reality of Latin America. Laclau, meanwhile, despite being born in Argentina and concerned with the analysis of his native country, has developed most of his work primarily in the central countries. This low figure of 12% would ratify the pattern of a division of work in which the theory is basically produced in the central countries. The theoretical production from the region is secondarily reported, so that the theoretical connection is primarily established with the central countries, in what could be understood as a pattern of cognitive dependency.

All of the above notwithstanding, and although the dominance of the central countries in terms of the development of theory in the field of social science is quite clear, a review of the national production permits us to identify a group of Chilean authors who are frequently cited; and even though they are not recognized as international-level theorists, they have developed abstract arguments of some generality that have achieved a significant level of diffusion within national borders and additionally extending their ideas, in some cases, to other Latin American countries. We could say that they are ‘local theorists’ or, more precisely, authors doing theoretical work, since they do not focus entirely on theory and are not generally recognized as theorists. In this group we considered, due to being the most cited: Eugenio Tironi, José Joaquín Brunner, Norbet Lechner, Tomás Moulian, Jorge Larraín, Sonia Montecinos, Pedro Morandé, Fernando Robles and Oscar Godoy, as ranked according to the number of citations.

If we include these authors in the total group of theorists, they receive 16.8% of all theoretical references. Although a low proportion, it is not negligible and may have influence in the national process of generation of knowledge. Whether it actually has this influence or not is the next question, whose answer I seek to advance using network analysis. I ask, then, how these producers of knowledge, these local theorists, connect with theorists from the North and how central or peripheral are their positions in the resulting theoretical network of the field. For this I appeal to the logic of co-citation analysis. I consider 73 authors, a sum of 64 international and those nine local authors.

The network analysis applied allows identifying a grouping of central authors such as Habermas, some others at a clearly peripheral position such as Jon Elster, Harold Garfinkel and Chantal Mouffe, and a third set in an intermediate position, like Norberto Bobbio or Richard Sennett. At the center are the greatest theorists: along with one of the classics – Weber – there is a group of authors of the second half of the 20th century: Habermas, Bourdieu, Luhmann, Giddens and Foucault. But also, closely intertwined with them, are several of the ‘local theorists’, prominently Moulian, Larraín, Brunner, Lechner and Tironi. Analyzing the networks separately by discipline, this situation is repeated: in sociology, Moulian occupies a position of great centrality in the theoretical network; in political science, Lechner and Tironi have such a position; and in the field of anthropology Montecinos and Lechner are the central authors, constituting Lechner an exceptional case of a political scientist cited by the anthropologists, making him one of the most prominent brokers in the whole field.

Conclusions

When observed with a global perspective, considering the results of central indexing platforms, the data reveal a strong divide between central and peripheral social sciences. However, when the relations between local social science and global science are observed from the point of view of the local space of a particular dependent and semi-peripheral country, Chile, the picture is more balanced than when observed from the perspective of the central indexing devices. One sees a body of social science that deals intensively with local situations and requirements, with extensive connections among local authors; although, on the other hand, paying special attention to what central scientific production could provide.

The great relevance of the references made to the theorists of central countries is unquestionable and, in that sense, Chile is part, to a large degree, of the international division of labor, in which theoretical production is being realized principally in the core countries. Nevertheless, there is also a group of local authors who do creative work elaborating abstract propositions, of a theoretical nature, referring fundamentally to local reality, who are frequently cited, and who are strongly intermingled with the international theorists (as confirmed by the co-citation analysis). One might say that the use and assimilation of international theories happens, to a large degree, within this mediating and translating network. Therefore, to conclude, on the basis of the relevance that the use of knowledge from the central countries has, that there is a situation of cognitive dependence would be exaggerated or at least an excessively simplified statement. If a central characteristic of cognitive dependence is to use knowledge that is inadequate for the local reality, which has the imprint from the central countries, without a filter, translation or criticism (Alatas, 2001, 2003), it could be argued that within this local circle of theoretical development such a creative and adaptive process is occurring. As a matter of fact, a review of the works of said authors leads us to the conclusion that among them such translation work – critical and adaptive – is really taking place. These results in the Chilean case contradict some of the statements made under the idea of cognitive dependence, allowing us to criticize the generalizations made about it and could further discussion of this notion.

An historical review of social science in Chile allows us to confirm that from the beginning, together with the institutionalization of this way of generating knowledge, there have been such receptive, adaptive and creative processes (Barrios and Brunner, 1988; Beigel, 2010, 2011; Brunner, 1988; Franco, 2007). The period between 1960 and 1973 is a conspicuous example in this regard, with a significant production of social science knowledge, focused on the local reality and with a profuse use of theories coming from the central countries (Beigel, 2010). ⁷ Around 1980, also, there is a significant theoretical elaboration and debate trying to develop ways to analyze the social reality of Chile under dictatorship. The field of local social science has conserved endogenous capabilities for generating knowledge, although it is situated within an international framework of inequality regarding the flow and appreciation of that knowledge.

A matter of concern regarding the observed pattern of global connections is the fact that the local production is not sufficiently valued and projected on a global level and that the local institutions do not help to increase its value and international projection, but rather they place obstacles in the path of its achievement. The problem is this and not the pronounced utilization of international publications. The central indexes, like SSCI and Scopus, despite the fact that they began with a definitely local character, referring to countries like the USA and UK, have defined themselves from the start as global, as representing a science authentically universal, covering the whole world, whereas, in contrast, the regional indexes of Latin America, like Scielo and Latindex, are conceived and projected as local, and are used by the countries of the region – definitely so in Chile – as second class indexes. In such a way, unintentionally, and despite all of the reiterated public discourse against the situation of scientific inequality, academic institutions become an accomplice in maintaining and reproducing the distinction between central science (coincidental with what is produced in the central countries) and peripheral science (correlated with what is produced by the peripheral countries).

Consequently, if there is no change in the central and regional indexation procedures or in the ranking criteria employed by scientific institutions in Chile, the inequality between central and peripheral science will remain. On the other hand, the strong emphasis given to global science by universities and public funding institutions is a threat to the current local focus and relative autonomy of the social sciences in this country.