Open Access Journals in the Developing World Laura Wimberley San José State University November 20, 2008

This paper examines the use of open access journals by academic libraries in the developing world: are open source journals a good choice for universities in the developing world, and to what extent are they currently being used? On one hand, as journal subscription prices rapidly increase for both print and electronic access, it seems obvious for research libraries with the least funding to turn to alternative publication methods for both distribution of their faculty members’ research and access to the research of others. On the other hand, this may ghettoize developing world research, cutting it off from scholarly communication with institutions that can afford to maintain traditional journal subscriptions. Any method of publication, whether open access or toll access or some other system, is a means to an end. That end is the free flow of information among researchers to allow the scholarly community to approach the truth. So far, the developing world has been held back from participating in that flow by three blockages: the costs of purchasing journals to read, the costs of publishing researching in journals, and censorship. I argue that open access journals are the best choice for the developing world, because only open access removes all three of these roadblocks. The concept of “the developing world" is quite broad, and encompasses a wide range of countries with very different conditions for scholarship. For the purposes of this paper, I propose thinking of the developing world in three categories. The first category is what the United Nations calls the least developed countries, such as Mozambique or Bangladesh. Many of these states have endured years of war, and libraries are typically destroyed early in any violent conflict (Rubin, 2004, p. 260). The second category is rapidly developing countries such as India and Brazil: these states are similar in that they are large federal democracies, with very uneven levels of regional development.

The final category is typified by China. China is so large and changing so swiftly as to be almost sui generis, but I would argue that its level of governmental censorship puts it in a distinct category when considering its universities’ options for journal access. Iran might also belong in this category. For example, China blocks nearly all blogs hosted by wordpress.com, an open source software platform that some faculty might use for self-archiving (“Thomson Reuters”, 2008). China also demands nationwide filters on all search engines, intended to eliminate everything from pornography to Islamic fundamentalism to discussion of democracy and human rights (Meiss & Menczer, 2007). Both China and Iran have blocked Wikipedia (Rask, 2008). This kind of central control is incompatible with the distributed, “bazaar” model of knowledge embedded in open access.

1

A search of the Directory of Open Access Journals supports this hypothesis, as shown in Table 1. Moreover, the Chinese and Iranian journals are mostly in uncontroversial topics: gastroenterology, physics, or agriculture. The Indian and Brazilian journals are much widerranging, covering both those normal science topics and potentially explosive topics like globalization, bioethics, minority ethnic groups, the sociology of education, gender, race, and international relations. These observations support the hypothesis that censoring regimes inhibit open access scholarship.

Table 1: Number of DOAJ from Sample of Developing Nations China 13 Iran 38 India 107 Brazil 356 This finding is entirely consistent with what Berkman Center scholar Ethan Zuckerman calls “the cute cat theory of digital activism” (Zuckerman, 2008). Zuckerman argues that any functional Web 2.0 tool will attract first pornography, and secondly activists: when people can effectively create, share, and distribute content, so far we have seen such systems used first for pornography, and, if they work well, for dissident political activity. Then, later adopters use them to share the trivia of their daily lives (like cute pictures of cats, hence the name of the theory). The implication of this general rule is that any tool that can be used for open access scholarship - which requires the creation, sharing, and distribution of content – can also be used for things some governments want to censor. Repressive governments, in order to maintain their censorship, must shut down either entire media – such as when Bahrain such down access to Google Earth to hide the extent of the royal family’s land ownership (Zuckerman, 2008) – or filter by keyword, as the government of Myanmar does for sites that name the country “Burma” (OpenNet, 2007). The richest countries, such as Japan, European Union members, and the U.S. and Canada, are relevant to our discussion here as the creators and potential donors of much of the most expensive research products. These are the home countries of publishers such as Elsevier, Thomson Reuters, EBSCO, etc. Middle-income states, such as South Korea or Kazakhstan, are not the focus of this paper: they can largely afford access to subscription journals, at least at their premiere institutions, but are not home to the most expensive research output. Historically, these richest countries have produced the majority of measured scholarly output. Countries representing 80% of the world’s population contribute only 13% of the titles listed in Ulrich’s Directory of Scientific Serials (Chan, Kirsop, & Arunachalam, 2005). Researchers in eight countries produce almost 85% of cited scientific articles, while researchers from 163 countries (almost all in the developing world) produced fewer than 3% (Chan, Kirsop, 2

& Arunachalam, 2005). This limitation on who produces, and is cited as, “science”, is reinforced by the definition of “core journals” by the Science Citation Index, a list which is necessarily truncated to a bibliometrically feasibly number, and which then must justify itself by denigrating those journals it cannot accommodate (Guédon, 2008). Exclusion from citation indices renders not just journals, but also individual scholars and entire institutions, invisible and unrecognized on the basis of their impact factors (Guédon, 2008). At first, this appears to be a self-perpetuating cycle with two effects on researchers in the developing world: 1) it requires them to spend unaffordable sums on journals from the global North to participate in “world science”; and 2) it forces them to publish their output in marginal “local” journals, which are not read or cited by researchers in the global North (Sahu, 2008). If these assumptions were still true, then toll access journals would be vital to keep the developing world in the global scholarly community.

Fortunately, there are a number of reasons why this is no longer the case. First of all, ever increasing computer power has made it increasingly easy to include journals of all kinds in citation indices. ISI’s Science Citation Index has grown significantly over time, and now includes over 8,000 titles. Better yet, it now faces competition from another subscription service, Elsevier’s Scopus, which indexes over 15,000 titles and conference proceedings (Guédon, 2008). Even better, both of these services now face competition from SCImago, a freely accessible journal ranking service, built on the Scopus list of journals, that is searchable by 233 countries and indexes journals from the physical and the social sciences (including library and information science) (SCImago, 2008). To distinguish themselves in this now competitive market, these indices will have to compete on the basis of comprehensive coverage, which means expansion to include both developing world journals and open access journals. To give examples from just one source, the open access family of journals at BioMed Central has thirteen notable journals not indexed in SCI that receive prominent rankings in SCImago (Cockerill, 2008). Once the impact of open access journals is measured, then reliance on them will not diminish the contributions of the researchers who publish in them.

Secondly, even universities in wealthier countries have reached the breaking point with skyrocketing journal costs. Theoretical physics, for example, operates almost entirely through the open access arχιv - which may be part of why the prestigious Max Planck Society in Germany cancelled its subscription to over 1,200 Springer journals last year, citing excess prices (Schimmer, 2007). Harvard University has the largest endowment in the United States, but its faculty have voluntarily committed to participation in an open access repository. So commitment to open access is increasingly unlikely to cut readers in the developing world off from the output of researchers in the global North.

The global South is not just a place of readers, but also of researchers. It is important to avoid the false and patronizing assumption that universities in the developing world are merely the passive recipients of knowledge charitably distributed by publishers or researchers from wealthier countries. In fact, developing countries have led the way to make open access possible. For example, Brazil has, according to the CEO of Sun, one of the most vibrant open source software development communities in the world (Schwartz, 2006). South African universities, 3

led by the University of Pretoria, began last year to integrate their open access repositories using D-Space (Van Deventer & Pienaar, 2008). The University of Pretoria has required submission of theses and dissertations to its institutional archive since 2004, and is now helping nine other institutions create and share their own open archives of a wider variety of material (Van Deventer & Pienaar, 2008). So the question of open access in the developing world is a question of traffic on a two-way street.

The online marketing manager of Elsevier Science UK takes exactly this kind of condescending tone while congratulating himself and his colleagues for their participation in the HINARI program (Silver, 2002). HINARI - Health Internetwork Access to Research Initiative under the leadership of the United Nations World Health Organization, subsidizes journals in medicine for institutions in lower-income countries, and provides them for free for the very poorest nations. Silver questions the need for these journals in the developing world, musing whether research on the latest drugs is merely taunting those without access, but concludes that a gesture of inclusion, even if merely symbolic, is good for morale. At no point does he consider that scholars in the developing world might be contributors to these Western-published journals, even as collaborators with better funded researchers, or in areas of particular local concern (such as malaria or HIV). Silver continues on to say that there are no guarantees about free or subsidized electronic access, even for those who buy the print journal (as if this were no doing of his, as the marketing manager at a major publisher), and frankly admits that the whole HINARI program is a marketing scheme to make librarians and researchers in rich countries feel better about themselves and the journal publishers. This is not a basis for a sustainable future for scientific inquiry in the fastest growing countries in the world.

So far, the HINARI program has been a disappointment. As of 2007, to use Peru as an example, researchers were unable to access 57% of the 150 HINARI journals with the highest impact factors; the other 43% were largely either journals that had been already individually subsidized for the developing world, or were fully open access for all, regardless of country (Villafuerte-Gálvez, Curioso, & Gayoso, 2007). In-depth, qualitative interviews with doctors in five African countries - Cameroon, Nigeria, Tanzania and Uganda and The Gambia - found similar results: only 47% of respondents had heard of HINARI, and of those who had, many reported difficulty with searching the interface and gaining access to passwords (Smith et al., 2007).

The results, however, were much more encouraging for PubMed - 90% awareness - and BMJ (British Medical Journal) (78% awareness) (Smith et al., 2007). These sources are also easier than subscription databases to use at Internet cafès, which in some countries are the most reliable access points. PubMed, operated by the U.S. National Institutes of Health, indexes both open access and toll access journals, but it is also now the repository of choice for the new requirement that NIH grant recipients provide an OA copy of any published work based on government grants. This means that as of 2007, PubMed archives an increasingly large body of cutting edge medical research. BMJ operates on a more traditional journal editorial model, and is

4

entirely open access. Again, the move to open access is not cutting the developing world off from, but rather integrating it with, wealthier states.

Open access production models, by lowering costs of publication, also make it easier for researchers in the developing world to publish their own findings. Medknow, based in Chennai, India, is an excellent example. Medknow publishes 73 fully open access journals, at no charge to the authors. Instead, the publications are supported by advertising and by charges for print copies (a useful supplement where Internet connectivity is sporadic). While many Medknow titles are Indian-focused, they also publish journals such as the African Journal of Paediatric Surgery, the Oman Journal of Ophthalmology, and the Saudi Journal of Gastroenterology. As of September 2006, readers downloaded 80,000 per month from Medknow, demonstrating this research is clearly valuable (Kirsop, Arunachalam, & Chan, 2007). Medknow is not alone; Bioline International, which also distributes OA publications created in the developing world, hit 2.5 million articles downloaded by the end of 2006 (Kirsop, Arunachalam, & Chan, 2007). This heavy extensive use of OA journals from the developing world indicates that the choice to focus away from subscription journals will not isolate scholars in the global South. On the contrary, as journal prices continue to skyrocket, even major, wealthy institutions like the University of California system are canceling subscriptions. Moving directly to open access is joining the developed world, not being ghettoized away from it.

This paper has largely focused on the applied sciences, because the majority of open access journals are in the science, technology, and medicine (STM) fields. This is partly because necessity is the mother of invention, and social sciences and humanities journals have not gotten as expensive quite as rapidly as scientific journals. In 2008, the cost of the average journal in the Science Citation Index was $1,267, while the cost of the average journal in the Social Sciences Citation Index was $448, and the cost of the average title in the Arts and Humanities Citation index was only $131 (Van Orsdel & Born, 2008). It is also because these fields are more focused on the book as a unit of scholarship: qualitative evidence takes more pages to present than quantitative data, and is so not easily amenable to the compactness of a journal article. Open access e-books are feasible, but present reading difficulty, and raise other questions, such as print-on-demand. For now, the most pressing need and the most creative work in open access serials is largely in the applied sciences.

However, some of the trends have been similar in the social sciences. For example, the American Anthropological Association just last month opened up its flagship journal and newsletter to open access - after a thirty-five year embargo (Jaschik, 2008). This is particularly galling from a developing nation perspective, since most of anthropology has historically been built on the study of the developing world by scholars from “more advanced" societies. Despite the differences in the disciplines, this pattern is similar to the HINARI program: after much criticism, the publisher, with great fanfare, made some contribution accessible to the developing world, but with so many restrictions as to be very difficult to use.

5

In the words of the editors of Public Library of Science Medicine (2006), “championing open access is arguably the most effective way that journals can help to lift people out of the extremes of poverty." For research to work, for epistemic communities to function, then scholars from all over the world must be able to read each other’s work. For researchers in the developing world, there are currently three barriers to access to this exchange of ideas: the costs of purchasing journals to read, censorship of journals, and the costs of publishing or disseminating research results. The donation system, as epitomized by HINARI, only addresses the first of these barriers, and it does not even do that effectively. The second of these barriers might not get enough attention in the library science literature: “Information wants to be free" is not an empty catch-phrase but rather a substantive political claim that some regimes find legitimately threatening. If China and other authoritarian states cannot allow the free flow of ideas, they will eventually limit their intellectual and economic growth. Open access journals are the only model that puts all the intellectual resources of the world to work on the all the world’s problems.

6

References Chan, L., Kirsop, B., & Arunachalam, S. (2005). Open access archiving: The fast track to building research capacity in developing countries. SciDev.Net. Retrieved November 10, 2008 from http://hdl.handle.net/1807/4415 Cockerill, M. (2008, June 4). SCImago – a new source of journal metrics offering a wealth of free data on open access journals. BioMed Central Blog. Retrieved November 10, 2008 from http://blogs.openaccesscentral.com/blogs/bmcblog/entry/scimago_a_new_source_of Guédon, J. C. (2008, March 7). Mixing and matching the green and gold roads to open access: Take 2. Serials Review. Retrieved November 10, 2008 from http://eprints.rclis.org/archive/00013863/01/Take-2.pdf Kirsop, B., Arunachalam, S. & Chan, L. (2007, July). Access to scientific knowledge for sustainable development: options for developing countries. Ariadne, 52. Retrieved November 3, 2008 from http://www.ariadne.ac.uk/issue52/kirsop-et-al/ Jaschik, S. (2008, October 7). Open access or faux access? . Inside Higher Ed. Retrieved November 3, 2008 from http://www.insidehighered.com/news/2008/10/07/anthro Medknow Publications. (2008). Retrieved November 3, 2008 from http://www.medknow.com/ Meiss, M. & Menczer, F. (2007, July). Visual comparison of results: A censorship case study. First Monday, 13, 7. Retrieved November 3, 2008 from http://www.uic.edu/htbin/cgiwrap/bin/ojs/index.php/fm/issue/view/267 OpenNet Initiative. (2007, May 10). Burma. Retrieved November 20, 2008 from http://opennet.net/research/profiles/burma PLoS Medicine Editors. (2006, August). How can biomedical journals help to tackle global poverty? . PLoS Medicine, 3, 8: e380. Retrieved November 3, 2008 from http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1564295 Rask, M. (2008, June). The reach and richness of Wikipedia: Is Wikinomics only for rich countries? . First Monday, 13, 8. Retrieved November 3, 2008 from http://www.uic.edu/htbin/cgiwrap/bin/ojs/index.php/fm/issue/view/266 Rubin, R.E. (2004). Foundations of Library and Information Science, 2nd ed. New York: NealSchuman.

7

Sahu, D.K. (2008, April 26). Gaining impact, readers and authors through fee-less-free dissemination: an experiment with open access. in Brainstorming Meet on Open Access, FLOSS and Copyright Law for Scholarly Communication and Literary Work, New Delhi, India. Retrieved November 10, 2008 from http://openmed.nic.in/2911/ Schimmer, R. (2007, October 18). Max Planck Society cancels licensing agreement with Springer. Retrieved November 10, 2008 from http://www.mpg.de/english/illustrationsDocumentation/documentation/pressReleases/200 7/pressRelease20071022/index.html Schwartz, J. (2006, April 16). The Brazilian effect. Jonathan’s Blog. Retrieved November 2, 2008 from http://blogs.sun.com/jonathan/entry/brazil SCImago. (2007). SJR Ñ SCImago journal & country rank. Retrieved November 10, 2008 from http://www.scimagojr.com Silver, K. (2002). Pressing the “send" key: Preferential journal access in developing countries. Learned Publishing, 15, 91-98. Retrieved November 2, 2008 from http://www.ingentaconnect.com/content/alpsp/lp/2002/00000015/00000002/art00003 Smith, H. et al. (2007). Access to electronic health knowledge in five countries in Africa: A descriptive study. BMC Health Services Research 2007, 7. Retrieved November 3, 2008 from http://www.biomedcentral.com/1472-6963/7/72 Thomson Reuters, China, and “regional" journals: Of gifts and knowledge production. (2008, May 29). GlobalHigherEd. Retrieved November 3, 2008 from http://globalhighered.wordpress.com/2008/05/29/thomson-scientific-china/ Van Deventer, M. & Pienaar, H. (2008, April). South African repositories: Bridging knowledge divides. Ariadne, 55. Retrieved November 2, 2008 from http://www.ariadne.ac.uk/issue55/vandeventer-pienaar/ Van Orsdel, L. & Born, K. (2008, April 15). Periodicals price survey 2008: Embracing openness. Library Journal. Retrieved November 10, 2008 from http://www.libraryjournal.com/article/CA6547086.html Villafuerte-Gálvez, J., Curioso, W.H. & Gayoso, O. (2007 June). Biomedical journals and global poverty: Is HINARI a step backwards? . PLoS Med., 4, 6: 220. Retrieved November 3, 2008 from http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1896213

8

Zuckerman, E. (2008, March 8). The cute cat theory talk at ETech. My Heart’s in Accra. Retrieved November 20, 2008 from http://www.ethanzuckerman.com/blog/2008/03/08/the-cute-cat-theory-talk-at-etech/

9