SEARCH

SEARCH BY CITATION

Keywords:

  • Amazonia;
  • citations;
  • collaboration;
  • scientific productivity

Abstract

  1. Top of page
  2. AbstractResumo
  3. Methods
  4. Results
  5. Discussion
  6. Coda
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

The presence of researchers from Western (i.e., developed world) institutions in Amazonia has frequently been contentious due to fears of ‘scientific imperialism’ or suspicions that they may be exerting undue influence over research agendas and knowledge production to the detriment of local researchers. Such negative perceptions are widespread, but not well substantiated. A more nuanced understanding of these issues requires information on who is conducting research in Amazonia and how knowledge production has changed over time. We performed a bibliometric analysis on Thomson Reuters'ISI Web of Science of all research articles about the Amazon published in three time periods: 1986–1989, 1996–1999, and 2006–2009. We found that the number of articles published and the diversity of countries involved in Amazonian research increased dramatically over the three time periods. The representation of several Amazonian countries—especially Brazil—increased, while the proportion of articles without a single author from an Amazonian country also increased. The results indicate that the research capacity of Amazonian countries has increased, but that leadership of high-impact projects may still largely reside with researchers from developed countries.

Resumo

A presença de instituições de pesquisadores ocidentais (mundo desenvolvido) na Amazônia tem sido frequentemente controversa devido a receios sobre ‘imperialismo científico’ ou suspeitas de que estes pesquisadores possam exercer influências indevidas sobre pautas de pesquisa e produção de conhecimento em detrimento de pesquisadores locais. Essas percepções negativas são bem difundidas, mas não são fundamentadas. Uma maior compreensão sobre estas importantes questões requer informações detalhadas sobre quem está conduzindo pesquisas na Amazônia, e como a produção de conhecimento mudou ao longo do tempo. Para atingir este objetivo, realizamos uma análise bibliométrica em Thomson Reuters' ISI Web of Science de todos os artigos publicados sobre a Amazônia em três períodos: 1986–1989, 1996–1999 e 2006–2009. Observou-se que o número de artigos publicados e a diversidade de países envolvidos em pesquisa na Amazônia aumentaram drasticamente ao longo dos três períodos. Além disso, a representação de vários países Amazônicos, especialmente o Brasil, aumentou – embora a proporção de artigos sem um único autor dos países Amazônicos também tenha se elevado. Os resultados obtidos neste trabalho indicam que a capacidade de pesquisa dos países Amazônicos tem crescido ao longo das últimas três décadas, porém a liderança de projetos de alto impacto pode ainda, muitas vezes, estar nas mãos de pesquisadores dos países desenvolvidos.

Different countries vary enormously in their capacity to perform research. The most economically developed countries typically contain the strongest universities and research centers, can devote more resources to research, and produce the most publications (May 1997, King 2004). Depending on the research field, however, a researcher's geographic focus may be in a country thousands of kilometers from their office. This is especially the case for biotropical research, where many researchers in developed countries study ecosystems that exist in the predominantly less developed countries of the global South.

Historically, there have been several types of conflict that can result from presence of researchers from foreign institutes in the global South. In the 1960s and 1970s, such conflicts focused on supposed ‘scientific imperialism’, whereby ‘scientists from developed countries descend upon developing countries to collect, ‘protect’ or capture and take home flora, fauna and professional prestige’ (Budowski 1975, p. 354). Such behavior may have been widespread at one time, but researchers have generally responded to these criticisms by becoming more collaborative and responsive to national priorities (Watkins & Donnelly 2005).

With the increasing globalization of science (Paasi 2005), the focus has shifted to issues associated with the production of knowledge. An example of this is the often delicate matter of scientific direction, execution and publication, e.g., who decides the research agenda? Who conducts the field research? Ultimately, who merits authorship? These questions have taken on increasing importance with the global spread of bibliometric indices based on publication quantity and quality (Todd 2009). These indices are routinely used to assess not only researchers, but also their institutions and even the countries in which their institutions are situated (Moed et al. 2004). Indeed, publications have become a global indicator of research quality that can determine funding, career development or even the fate of entire departments or institutions (Smith 2000).

Publications also represent a major source of knowledge production, and their careful analysis can reveal what people, institutions, or countries are driving research agendas and, conversely, who is at the margin (Moed et al. 2004). Given the educational and resource advantages that Western researchers often have over their collaborators and competitors in the global South, it is unsurprising that concerns have been expressed that the former may exert too much control over knowledge production in the developing world (Acosta-Cazares et al. 2000, Carnoy & Castells 2000, Malhado 2011).

The Amazon forest of South America is a microcosm for the debate about the role, influence and obligations of foreign researchers in the tropics. The region has acted as a magnet for Western researchers and there has been a long and continuous history of foreign scientific study in the Amazon from the late 18th century onward (Maslow 1996). Amazonia continues to be a fertile area for research, especially in the context of the dominant environmental agendas of climate change (e.g., Malhi et al. 2008), biodiversity (e.g., Laurance et al. 2002), and the enormous gaps in knowledge regarding almost every aspect of this forest's ecosystem (Bush & Lovejoy 2007).

In this paper, our objective is to explore historical trends of scientific research in the Amazon region through bibliographic analysis(Hammarfelt 2011). The focus of our study is the national identity of author institutions (cf. van Leeuwen 2009, Persson 2010). Specifically, our aims are to quantify (1) temporal patterns in the diversity of countries involved in Amazonian research; (2) representation patterns at a country level in terms of proportion of publications, authorship of highly/poorly cited articles; (3) patterns of first/coauthorship. We focus on first authorship because it is commonly recognized as an indication of scientific leadership. Furthermore, first authorships resulting from international cooperation can identify to which extent countries are taking the lead in publishing of scientific results and can be ‘bibliometrically used as an indication of the leadership of a country’ (van Leeuwen 2009, p404).

Methods

  1. Top of page
  2. AbstractResumo
  3. Methods
  4. Results
  5. Discussion
  6. Coda
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

We used Thomson Reuters' ISI Web of Science to find all (English language) standard research articles that appeared for the search string ‘Amazon OR Amazonia’. Articles were sorted by number of citations and each article was checked individually to assess if the content was substantially associated with Amazonia. We removed articles whose focus was a global analysis, where the study was not clearly Amazonian, and in which Amazonia was only mentioned peripherally.

Articles used in the analysis were restricted to those published in English for several reasons: First, the vast majority of international peer-reviewed scientific journals only publish articles written in English, and this is almost universally the case for high-impact journals. Second, English is the lingua franca of science (Tardy 2004) and there is considerable pressure on non-native English speaking (Non-NES) researchers to publish in international journals (Alejandro 2012, see Ammon 2001, Montgomery 2004 for a detailed discussion of this issue). Finally, as knowledge that is codified and communicated in languages other than English is difficult for the international community to assess, non-English language articles are likely to have different citation dynamics and therefore would not be directly comparable to the other articles used in the analysis.

For each publication, we noted (1) year of publication; (2) number of citations accrued by the day of retrieval; (3) number of institutions involved; (4) the country of the first author's institution; (5) countries of coauthors' institutions; and (6) journal of publication. We collected these data for all articles published over three time periods: 1986–1989, 1996–1999, and 2006–2009. Articles were also divided into three citation classes (high, medium and low) for each decade. For the older time periods (1986–1989 and 1996–1999), articles were assumed to have reached (or nearly reached) citation equilibrium and were therefore treated as homogenous groups (May 1997). Articles for 2006–2009 were split into citation classes by year to account for differences in the time available for articles to accrue citations.

Citation groups were defined by the following procedure: first, all articles within a given time period (1986–1989, 1996–1999, 2006–2009) were sorted by number of citations. Second, the number of cites for the 33rd and 66th percentile were recorded. Third, articles that fell within these boundaries were classified as belonging to the medium-citation category, articles with more cites than the 66th percentile were classified in the high-citation category, and articles with fewer citations than the 33rd percentile were classified in the low-citation category. It should be noted that, due to overlap in the number of citations, the medium-citation category is larger than the high and low categories.

Data compilation and collation were completed within 60 d (in June/July 2012). Following van Leeuwen (2009), we assume that first authors are ‘leading’ the research. Correspondingly, coauthors were considered participants with a predominantly supporting role (e.g., data provision, technical support, data analysis, editing, etc.). These assumptions are relatively robust and temporally stable across disciplines (Weltzin et al. 2006, Tscharntke et al. 2007).

Results

  1. Top of page
  2. AbstractResumo
  3. Methods
  4. Results
  5. Discussion
  6. Coda
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

We assessed a total of 3354 articles after applying our filtering procedure, the majority of which (2225) were published between 2006 and 2009 (Table 1). From the 1980s to the 2000s, there was a strong positive trend in number and national diversity of institutions at which authors were based (Table 1). As would be anticipated, articles published in the 1980s and 1990s had accrued more citations on average than those published in the 2000s.

Table 1. Summary of the data set used, divided into the three different time periods
 1986–19891996–19992006–2009Total
Number of articles25087922253354
Mean nos. authors2.49 ± 1.813.23 ± 2.164.62 ± 3.664.09 ± 3.30
Mean number of addresses1.48 ± 0.902.04 ± 1.263.02 ± 2.322.65 ± 2.08
Mean number of citations44.60 ± 69.2631.43 ± 44.959.97 ± 16.1318.17 ± 3465
Number of countries21376868

We found strong evidence that Amazonian science is becoming more globalized (Fig. 1). In total, institutions from 68 countries contributed to knowledge about the Amazon and Amazonia over the three survey periods. These data were highly skewed, however, with only 21 countries represented in Amazonian articles in the 1980s, 37 in the 1990s, and 68 in the 2000s. The three most represented countries (proportion of total addresses) in each decade were Brazil, U.S.A., and U.K., with Brazil overtaking the U.S.A. as the most represented nation in the two most recent time periods. All Amazonian countries (Bolivia, Brazil, Colombia, Ecuador, French Guiana, Guyana, Peru and Venezuela) with the exception of Suriname had institutions that were represented in the data base.

image

Figure 1. Trends in representation as first author for articles in the high-cited and low-cited categories during the three study periods. (Color version available online as part of Supporting Information).

Download figure to PowerPoint

There were a total of 68 countries with one or more affiliated institutions that participated in articles about Amazonia, 14 of which appeared only in a collaborative (coauthor) capacity. Overall, researchers from institutions based in Amazonian countries were lead authors on 44.5 percent of articles across all three time periods. Moreover, there was a clear trend of increasing leadership by researchers from institutions based in Amazonian countries (χ2 = 15.62, < 0.001), from 17.7 percent of articles in 1986–1989 to 29.9 percent of articles in 2006–2009 (see Table S1). Over the same period, leadership of low-cited articles by Amazonian institutions also significantly increased (χ2 = 117.58, < 0.001) from 31.5 percent in 1986–1989 to 63.4 percent in 2006–2009.

For the articles led by non-Amazonian institutions, 42 percent had at least one Amazonian-coauthor. Based on this same sub-dataset, there is a proportional increase of articles through time (χ2 = 16.90, P < 0.001) led by non-Amazonian authors having at least one Amazonian coauthor (31%, 38% and 43% in the respective study periods).

At the country level, Brazil was the most represented country in terms of first authors. Researchers from Brazilian-based institutions led 40.8 percent of all the articles in the data base. The next most represented countries were the U.S.A. (29.3%), U.K. (5.9%), Germany (3.8%), France (3.6%), and Canada (2.7%). Researchers from all other countries led less than two percent of articles. The three top-cited countries show strongly divergent trends over the three time periods. Researchers from Brazilian institutions have a very strong pattern of increasing representation as first authors in low-cited articles (χ2 = 224.26, P < 0.001)—but this trend is much weaker for the subset of highly cited articles (χ2 = 11.79, P < 0.003), with a rise between 1986–1989 and 1996–1999 after which the percentage remains at approximately 27 percent.

In contrast, the first author representation of researchers from US-based institutions decreases in high (χ2 = 123.87, P < 0.001) cited articles (from 48% in 1986–1989 to 35.7% in 2006–2009; Fig. 1; Table S1), with a weaker but still significant decrease (χ2 = 8.17, P = 0.017) in first authorship for low-cited articles (Fig. 1, Table S1). Several other trends were also apparent for lead authorship among lower represented nations, such as the reduction of low-cited articles from researchers at German institutions (χ2 = 27.23, P < 0.001) and the increase in the representation of researchers from U.K. institutions in high-cited articles (χ2 = 38.77, P < 0.001)from 3.6 percent in 1986–1989 to 12.9 percent in 2006–2009 (Fig. 1; Table S1).

Seven out of the nine Amazonian countries were represented as first authors. There is a marked increase in the diversity of representation of Amazonian countries over time, with only three countries (Bolivia, Brazil and Venezuela) represented as first authors in articles published from 1986 to 1989 compared to seven countries in the 2006–2009 period. With the exception of Brazil, representation as first authors of researchers associated with institutions in Amazonian countries is low and shows no clear trend with respect to citation category. Temporal trends for collaboration are broadly similar to those observed in the first author analysis (Figs 2 and 3). The most represented countries are again Brazil, the U.S.A. and the U.K., with one or more researchers from Brazilian institutions appearing on 61.6 percent of articles published between 2006 and 2009. Brazil had a strongly increasing trend of collaborative research in both low (χ2 = 215.78, P < 0.001) and high-cited (χ2 = 57.26, P < 0.001) articles. As with lead authorship, the representation of researchers from US-based institutions showed different trends depending on citation category. Specifically, coauthors from US institutions strongly decreases in high (χ2 = 134.816, P < 0.001) cited articles (from 80.4% in 1986–1989 to 49.1% in 2006–2009; Fig. 2, Table S2), but remains approximately constant (χ2 = 3.800, P = 0.150) at around 25 percent for low-cited articles (Fig. 2; Table S2).

image

Figure 2. Representation (% of articles) of countries (one or more researchers associated with a national institution) as coauthors – the ten most represented countries.

Download figure to PowerPoint

image

Figure 3. Representation (% of articles) of countries (one or more researchers associated with a national institution) as coauthors – Amazonian countries.

Download figure to PowerPoint

Apart from Brazil, researchers from institutions based in other Amazonian countries had higher representation as collaborators than lead authors (Fig. 3; Table S2). This is especially noticeable for Peru, whose researchers were authors on more than 6 percent of articles published in the final time period, but usually not as first authors (only on 0.8% of the articles published in this same period). All Amazonian countries—with the exception of Suriname, which was not represented as either a lead institution or collaborating institution in our data base—increased their rate of collaboration over the three time periods.

There was no clear trend in coauthor representation of researchers from institutions based in Amazonian countries on high- or low-cited articles across the whole data base (Fig. 3). In the final time period (2006–2009), Brazil had greater coauthor representation on low-cited articles, but Bolivia, Colombia, Ecuador, French Guiana, Peru and Venezuela had greater representation in the subset of highly cited articles. Moreover, there was no obvious increase in the representation of researchers affiliated with institutions from Amazonian countries on articles with first authors from foreign institutes for either low- or high-cited subsets of articles.

Discussion

  1. Top of page
  2. AbstractResumo
  3. Methods
  4. Results
  5. Discussion
  6. Coda
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

From the mid-1960s onward, the dominating narratives regarding trends in scientific knowledge production in developing countries have been largely focused on the putatively corrosive effects of ‘scientific imperialism and colonialism’ (Varsavsky 1967, Lewis 1973, Budowski 1975). Such narratives arguably had various influences. First, they drew attention to the importance of sovereignty and the right of nations to exploit their own natural resources. This later became enshrined in legislation in Article 15 the 1992 Convention on Biological Diversity, which recognizes ‘the sovereign rights of States over their natural resources’, and that the ‘authority to determine access to genetic resources rests with the national governments and is subject to national legislation’. The CBD facilitated the drafting of legislation that carefully defines and limits commercial and research activities by representatives of foreign institutions (ten Kate 2002, Schindel 2010) – now widespread in Amazonian countries(Elisabetsky & Costa-Campos 1996, Neto 1998, Rull & Vegas-Vilarrúbia 2008). Second, scientific colonialism narratives undoubtedly raised awareness among developed world researchers about the need to contribute to national research agendas and raise capacity in host countries (Nobre et al. 2008).

Changes in legislation and attitudes have undoubtedly influenced the behavior of researchers working in the Amazon and simplistic explanations based on Western hegemony and Neocolonialism need to be rethought. For a start, it needs to be acknowledged that researchers from industrialized countries who work in less developed tropical countries can have both positive and negative aspects (Velho & Velho 1996, Grajal 1999, Watkins & Donnelly 2005). Positive consequences include training and capacity-raising among local researchers, an increase in resources for local research, economic benefits for host communities and governments, investment in infrastructure and, perhaps most importantly, the enhanced production of globally significant knowledge. Negative outcomes include foreign researchers removing valuable biological material, exploiting local researchers and, potentially, restricting their opportunities to lead important research projects within their own geographic borders. Trends in knowledge production also need to be viewed in the context of intrinsic and historical inequalities in global science that favor researchers from high-ranking institutions in industrialized countries (Dahdouh-Guebas et al. 2003, Parker et al. 2010). Indeed, many talented researchers from Amazonian countries work in these institutions—a factor that, however, cannot be assessed using Web of Science bibliometrics.

Our bibliographic analysis suggests that there are at least four linked trends in knowledge production in Amazonia, each of which could be viewed positively, neutrally or negatively depending upon nationality, politics and personal values. First, Amazonian research is becoming more globalized with an increasing range of international research networks operating within the basin. The number of countries with institutions represented in the data base rose from 21 in 1986–1989 to 67 in 2006–2009. This trend should be partly interpreted in the context of increases in the mean number of authors on scientific manuscripts, particularly apparent in ecological articles (cf. Weltzin et al. 2006) that make up the bulk of the articles in our study. Nevertheless, the increasing diversity of ‘institutional nationality’ clearly demonstrates the process of globalization of science within the Amazon, especially in the final study period (2006–2009). This latter trend may be partly attributable to an explosion of collaborative research projects, especially those associated with assessing the impacts of climate change on the Amazonian ecosystem. Pre-eminent among these is the large-scale biosphere–atmosphere experiment in Amazonia (LBA), the largest program in international cooperation ever focused in the Amazon region and which brought together a multitude of researchers from (predominantly) Brazilian, American and European institutions (Avissar et al. 2002, Lahsen & Nobre 2007). The first phase of the program (1998–2006) was co-funded by NASA and had an explicitly international focus, providing extensive training to support the more nationally focused second phase (Nobre et al. 2008). More generally, the last 50 yr have seen a strong trend of increasing globalization of science (Zitt & Bassecoulard 2004) and Amazonia appears to be part of this globalizing trend.

The relative representation of institutions from a given country within the data base clearly depends on a range of factors, including the flow of researchers between nation states, funding opportunities, linguistic and cultural relationships and the history of collection and research in the Amazon. This latter factor may contribute to the relatively high representation of smaller countries such as Finland and the Netherlands, which both have extensive biological collections of Amazonian material in the Herbarium of the University of Turku and the National Herbarium of the Netherlands. Germany, the U.K., and the U.S.A. also have large biological collections from the Amazonian with many specimens over 100 yr old.

Individual researchers may also have influenced contemporary patterns of knowledge production by both their contributions to biological collections and through the students they trained and research groups that they founded. This is especially true for pioneering researchers from the developed world such as Ghillean Prance (U.K.), Jürgen Haffer (Germany), Thomas Lovejoy (U.S.A), and Alwyn Gentry (U.S.A), among many others. Their immense academic legacy (and that of other western researchers in the Amazon) has continued to influence the geography of knowledge production and shape contemporary research agendas.

The second strong trend is the increasing research capacity of Amazonian nations. This is especially apparent in Brazil, which has invested strongly in research throughout the country (Regalado 2010) and was ranked 15th in the world for scientific output in 2012(SCImago 2012). Brazil has a large university system and there has been a continuous rise in the numbers of graduate students since the 1950s (Leite et al. 2011). Peru has also increased its research capacity, although the factors driving this trend are harder to pin point. They may be partly due to the introduction of regulations that require foreign researchers to work with researchers from a local scientific institution (Edison et al. 2011).

There are several Amazonian countries (Guyana, French Guyana and Suriname) with consistently very low institutional representation in all categories of research and all time periods. However, these values should not be taken at face value, given the small number of scientific institutions and the low research productivity of these countries, especially compared with Brazil. If the number of articles by Amazonian countries in the data base is adjusted in proportion to their overall scientific productivity (number of citable documents 1996–2011 derived from SCImago 2012), then a very different pattern emerges. The most productive countries (all citation categories, all years) are Bolivia, French Guiana, Peru, and Ecuador, demonstrating the relative importance of Amazonian research to the overall research productivity of these nations.

The increasing production of research from institutions based in Amazonian countries is particularly noticeable for articles with a low number of citations. Once again, this is consistent with more global studies, which indicate that Latin American articles, especially those originating from domestic production, typically have a very low impact that is considerably below world average values (Benevént-Perez et al. 2012). Such research output may have considerable value being tailored to local research needs (Ladle et al. 2012). Higher impact internationally orientated research, however, is also important (e.g., to attract international funding, for national prestige, etc.) and its lack may indicate a lack of capacity (see discussion below).

A third clear trend has been the relative diminution of North American representation in Amazonian research (Fig. 1). Our data demonstrate that as early as the late 1980s, almost half of all articles about the Amazon were led by researchers with affiliations to U.S. institutions, but this had fallen to one in four articles by the late 2000s. As with all the trends in our data base, the reasons behind this relative decline in representation are undoubtedly complex and multifaceted. Furthermore, the decrease is partly attributable, to an increase in scientific production from other countries. Nevertheless, the U.S.A. is still the highest producer of scientific literature in the World (SCImago 2012), and researchers associated with U.S. institutions remain among the most important collaborators and funders for many researchers from Amazonian countries. For example, the Gordon and Betty Moore Foundation Andes-Amazon initiative (www.moore.org/andes-amazon) is currently one of the largest in the region, providing substantial grants for environmental researchers throughout the Amazon basin.

Finally, our data suggest that, even though representation of researchers from some Amazonian countries in Amazon academic articles has increased, developed world researchers are still overrepresented as first authors in high-impact publications. This should not necessarily be interpreted as a negative trend as, as mentioned above, it is anticipated that researchers from Amazonian countries will engage in more geographically localized research that addresses national priorities, which often leads to publications in lower impact journals (Ladle et al. 2012). Even so, linguistic, cultural, and practical barriers (cf. Dahdouh-Guebas et al. 2003) may exist, making it less likely that a scientist from an Amazonian country will assume leadership of a collaborative paper. Moreover, there is evidence that articles published by researchers from Latin American institutions in high-impact journals accrue fewer citations than articles by authors affiliated with European/north American institutions or articles that are the result of collaboration (Meneghini et al. 2008). These inferences and the results on which they are based should be treated with considerable caution. Although authors are often assumed to be listed in order of their contribution, this norm of practice has never been codified and numerous exceptions exist (Weltzin et al. 2006, Wager 2009). Moreover, bibliometric studies cannot definitively assess who is driving research agendas, and who is genuinely contributing content, expertise, and material support. This may particularly be the case in the Amazon, given that legal requirements in some parts of the region make it an obligation to include authors from that country's institutions in any subsequent publications stemming from the research.

Assuming that a native Amazonian research leadership deficit does exist, training a new generation of Amazonian researchers with the skills, motivation, and opportunities to lead major research projects and to dictate research agendas should be a priority for policy makers and funders (Malhado 2011). Giving greater control over Amazonian knowledge production to researchers from Amazonian countries would have the added advantages of ensuring a better alignment between national and global research priorities, and may go some way to allaying a widespread, but unsubstantiated, conspiracy theory that ‘foreigners are trying to take over the Amazon’. Many Amazonian research projects already include a strong capacity-building element (Nobre et al. 2008) and new initiatives, such as the Brazilian Governments ‘Science Without Frontiers’ project (www.cienciasemfronteiras.gov.br) will further contribute to filling the ‘leadership deficit’.

Coda

  1. Top of page
  2. AbstractResumo
  3. Methods
  4. Results
  5. Discussion
  6. Coda
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

Mapping and assessing scientific knowledge production using bibliometrics can provide valuable insights that can inform research funding, fieldwork policies, and the governance of international research networks (Moed et al. 2004). Results of such analyses, however, need to be treated with caution and we would strongly advocate an approach that distinguishes between high- and low-impact research and which discriminates between research leaders and those providing data and technical support.

The debate regarding knowledge production in the Amazon now focuses on how to facilitate the continued development of researchers based in Amazonian countries (Stocks et al. 2008) and how to fill important knowledge gaps. We strongly endorse Pitman et al.'s (2011) recent suggestions for how this might be done: (1) strengthen local institutions; (2) invest in local talent; (3) facilitate responsible research by foreigners; (4) offer grants for field research;(5) carry out landscape studies; and (6) continue to monitor volume and geographic distribution of scientific publications. Many of these strategies are already being implemented, and participation of researchers based in Amazonian countries in both high- and (especially) low-cited articles is increasing. Brazilian-based researchers, in particular, appear well on the road to establishing a genuinely independent research culture. Finally, it is important to re-emphasize that the underlying drivers of the trends identified in this research are undoubtedly complex and should be cautiously inferred. Further research would be needed to tease out local and regional factors from global trends in scientific funding, practice, and culture. Likewise, research on the national origins of Amazonian researchers would help disentangle the influence of factors such as educational background, linguistic competence, and international networks on research productivity.

Acknowledgments

  1. Top of page
  2. AbstractResumo
  3. Methods
  4. Results
  5. Discussion
  6. Coda
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information

We thank Matias Batista (Federal University of Campina Grande) and Dr Lynette Loke (National University of Singapore) for assistance in the data analysis. Thanks also to two anonymous referees and Professor Emilio Bruna whose comments and criticisms significantly improved the manuscript.

Literature Cited

  1. Top of page
  2. AbstractResumo
  3. Methods
  4. Results
  5. Discussion
  6. Coda
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information
  • Acosta-Cazares, B., E. Browne, R. E. Laporte, D. Neuvians, K. Rochel de Camargo, R. Tapiaconyer, and Y. Ze. 2000. Scientific Colonialism and Safari Research. BMJ Publishing Group, London.
  • Alejandro, B. 2012. Running like Alice and losing good ideas: On the quasi-compulsive use of english by non-native english speaking researchers. Ambio 41: 769772.
  • Ammon, U. 2001. The Dominance of English as a Language of Science: Effects on Other Languages and Language Communities. Mouton the Gruyter, New York.
  • Avissar, R., P. L. Silva Dias, M. A. F. Silva Dias, and C. Nobre. 2002. The large-scale biosphere-atmosphere experiment in Amazonia (LBA): Insights and future research needs. J. Geophys. Res. 107: 8086.
  • Benevént-Perez, M., J. Gorraiz, C. Gumpenberger, and F. de Moya-Anegón. 2012. The different flavors of research collaboration: A case study of their influence on university excellence in four world regions. Scientometrics 93: 4158.
  • Budowski, G. 1975. Scientific imperialism. Science and Public Policy 2: 354360.
  • Bush, M. B., and T. E. Lovejoy. 2007. Amazonian conservation: Pushing the limits of biogeographical knowledge. J. Biogeogr. 34: 12911293.
  • Carnoy, M., and M. Castells. 2000. Globalization, the knowledge society, and the Network State: Poulantzas at the millennium. Global Networks 1: 118.
    Direct Link:
  • Dahdouh-Guebas, F., J. Ahimbisibwe, R. van Moll, and N. Koedam. 2003. Neo-colonial science by the most industrialised upon the least developed countries in peer-reviewed publishing. Scientometrics 56: 329343.
  • Edison, A. S., E. Cosio, S. Halloy, and J. Vivanco. 2011. So, you want to do research in the rainforest? Chemical biology of the tropics. In J. M. Vivanco, and T. Weir (Eds.). Chemical Biology of the Tropics, pp. 97111. Springer, Berlin, Heidelberg.
  • Elisabetsky, E., and L. Costa-Campos. 1996. Medicinal plant genetic resources and international cooperation: The Brazilian perspective. J. Ethnopharmacol. 51: 111120.
  • Grajal, A. 1999. Biodiversity and the nation state: Regulating access to genetic resources limits biodiversity research in developing countries. Conserv. Biol. 13: 610.
  • Hammarfelt, B. 2011. Interdisciplinarity and the intellectual base of literature studies: Citation analysis of highly cited monographs. Scientometrics 86: 705725.
  • ten Kate, K. 2002. Science and the convention on biological diversity. Science 295: 23712372.
  • King, D. A. 2004. The scientific impact of nations. Nature 430: 311316.
  • Ladle, R., P. Todd, and A. C. M. Malhado. 2012. Assessing insularity in global science. Scientometrics 93: 745750.
  • Lahsen, M., and C. A. Nobre. 2007. Challenges of connecting international science and local level sustainability efforts: The case of the Large-Scale Biosphere–Atmosphere Experiment in Amazonia. Environ. Sci. Policy 10: 6274.
  • Laurance, W. F., T. E. Lovejoy, H. L. Vasconcelos, E. M. Bruna, R. K. Didham, P. C. Stouffer, C. Gascon, R. O. Bierregaard, S. G. Laurance, and E. Sampaio. 2002. Ecosystem decay of Amazonian forest fragments: A 22-year investigation. Conserv. Biol. 16: 605618.
  • van Leeuwen, T. 2009. Strength and weakness of national science systems: A bibliometric analysis through cooperation patterns. Scientometrics 79: 389408.
  • Leite, P., R. Mugnaini, and J. Leta. 2011. A new indicator for international visibility: Exploring Brazilian scientific community. Scientometrics 88: 311319.
  • Lewis, D. 1973. Anthropology and colonialism. Current Anthropology 14: 581602.
  • Malhado, A. C. M. 2011. Amazon science needs Brazilian leadership. Science 331: 857.
  • Malhi, Y., J. T. Roberts, R. A. Betts, T. J. Killeen, W. Li, and C. A. Nobre. 2008. Climate change, deforestation, and the fate of the Amazon. Science 319: 169172.
  • Maslow, J. 1996. Footsteps in the jungle: Adventures in the scientific exploration of the American tropics. Ivan R. Dee, Chicago.
  • May, R. M. 1997. The scientific wealth of nations. Science 275: 793796.
  • Meneghini, R., A. L. Packer, and A. N. D. L. Nassi-Calo. 2008. Articles by Latin American authors in prestigious journals have fewer citations. PLoS ONE 3: e3804.
  • Moed, H. F., W. Glänzel, and U. Schmoch (Eds.). 2004. Handbook of Quantitative Science and Technology Research: The Use of Publication and Patent Statistics in Studies of S &T Systems. Kluwer Academic Publishers, Dordrecht.
  • Montgomery, S. 2004. Of towers, walls, and fields: Perspectives on language in science. Science 303: 13331335.
  • Neto, R. B. 1998. Brazil's researchers warn against ‘nationalistic’ restrictions. Nature 392: 538.
  • Nobre, C. A., M. Lahsen, and J. P. H. B. Ometto. 2008. Global environmental change research: Empowering developing countries. Anais da Academia Brasileira de Ciências 80: 523529.
  • Paasi, A. 2005. Globalisation, academic capitalism, and the uneven geographies of international journal publishing spaces. Environment and Planning A, 37: 769789.
  • Parker, J. N., C. Lortie, and S. Allesina. 2010. Characterizing a scientific elite: The social characteristics of the most highly cited researchers in environmental science and ecology. Scientometrics 85: 129143.
  • Persson, O. 2010. Are highly cited articles more international? Scientometrics 83: 397401.
  • Pitman, N. C. A., J. Widmer, C. A. Jenkins, G. Stocks, L. Seales, F. Paniagua, and E. M. Bruna. 2011. Volume and geographical distribution of ecological research in the Andes and the Amazon, 1995-2008. Tropical Conservation Science 4: 6481.
  • Regalado, A. 2010. Brazilian science: Riding a gusher. Science 330: 13061312.
  • Rull, V., and T. Vegas-Vilarrúbia. 2008. Biopiracy rules hinder conservation efforts. Nature 453: 26.
  • Schindel, D. E. 2010. Biology without borders. Nature 467: 779781.
  • SCImago (2012) SJR — SCImago Journal & Country Rank. Available at: http://www.scimagojr.com (accessed 20 May 2013).
  • Smith, N. 2000. Afterword: Who rules this sausage factory? Antipode 32: 330339.
  • Stocks, G., L. Seales, F. Paniagua, E. Maehr, and E. M. Bruna. 2008. The geographical and institutional distribution of ecological research in the tropics. Biotropica 40: 397404.
  • Tardy, C. 2004. The role of English in scientific communication: Lingua franca or Tyrannosaurus rex? Journal of English for Academic Purposes 3: 247269.
  • Todd, P. A. 2009. Ambiguity, bias, and compromise: An abc of bibliometric-based performance indicators. Environment and Planning A. 41: 765771.
  • Tscharntke, T., M. E. Hochberg, T. A. Rand, V. H. Resh, and J. Krauss. 2007. Author sequence and credit for contributions in multiauthored publications. PLoS Biol. 5(1): e18.
  • Varsavsky, O. 1967. Scientific colonialism in the hard sciences. American Behavioral Scientist 10: 2223.
  • Velho, L., and P. Velho. 1996. Scientific collaboration of developing/advanced countries in biological sciences: The case of the Maraca Rainforest Project. Cadernos de Ciência & Tecnologia, Brasília 13: 920.
  • Wager, E. 2009. Recognition, reward and responsibility: Why the authorship of scientific articles matters. Maturitas 62: 109112.
  • Watkins, G. G., and M. A. Donnelly. 2005. Biodiversity research in the Neotropics: From conflict to collaboration. Proceedings of the Academy of Natural Sciences of Philadelphia 154: 127136.
  • Weltzin, J. F., R. T. Belote, L. T. Williams, J. K. Keller, and E. Cayenne Engel. 2006. Authorship in ecology: Attribution, accountability, and responsibility. Front. Ecol. Environ. 4: 435441.
  • Zitt, M., and E. Bassecoulard. 2004. Internationalisation in science in the prism of bibliometric indicators. In F. Moed, W. Glänzel, and U. Schmoch (Eds.). Handbook of Quantitative Science and Technology Research: The Use of Publication and Patent Statistics in Studies of S&T Systems, pp. 407436. Kluwer Academic Publishers, Dordrecht.

Supporting Information

  1. Top of page
  2. AbstractResumo
  3. Methods
  4. Results
  5. Discussion
  6. Coda
  7. Acknowledgments
  8. Literature Cited
  9. Supporting Information
FilenameFormatSizeDescription
btp12079-sup-0001-TableS1-S2.docxWord document22K

TABLE S1. Representation of countries with scientists publishing as first authors, divided according to the number of citations.

TABLE S2. Representation of countries (one or more scientists associated with a national institution) as coauthors.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.