Tag Archives: developing world

Open access in Africa – green and gold, the impact factor, ‘mainstream’ and ‘local’ research

I have been following the debate raging in the UK and beyond about whether the Finch Commission and the Research Councils UK  – and then the EC with a slightly different emphasis – were right in opting for support for the ‘gold route’ of open access publishing rather than prioritizing only the ‘green route’ of open access repositories. There seems to have been a general consensus in the commentaries that I have read that this will disadvantage the developing world, which will be faced with the barrier of high article processing fees and become increasingly excluded. The green route, through continuing creation of institutional repositories, would be better for us, we are told.

I don’t agree. The reasons are complex, but at heart this takes us back to the question of whether we are seeking access to or participation in the production of global literature. Which policy path would most effectively give voice to research from Africa, largely silenced in the current system? Access to world literature is also important, but is inadequate on its own, risking perpetuating a neo-colonial dispensation that casts the dominant North as the producer and the developing world as the consumer of knowledge.

I have come to think that the green/gold debate is in fact a distraction from dealing with more insidious issues in our research publishing systems. These include the dominance of journals at the expense of other forms of publication; the almost universal adoption of the ISI and its Impact Factor as the basis for recognition and reward; and, most insidious of all, the marginalization of great swathes of global research through the implementation of this commercialized ranking system.

Another related but under-recognised issue is the extent to which there is an assumption that scholarly publishing is a commercial business, built around profit creation. This has led to a free rider syndrome, allowing senior administrators to remain oblivious to the need to address support for research communication as a policy issue. In this regard, Finch sets a good precedent, in making it clear that getting beyond the issues that block effective research communication requires government investment.

Once the argument moves to discussion of what the impact of these new open access policies would be on the participation of the developing world, then the nature of the debate changes. I would argue that the either/or dichotomy between green and gold is in fact a distraction –  the wrong question, generating the wrong answers. Whether green or gold is not that relevant in the African context unless one understands the mechanisms of exclusion that consigns our research publications to the margins. Even then there is unlikely to be a clean either/or solution. What is important is an understanding of the contextual issues and the power dynamics that are at play.

 History and context – a neo colonial system

African universities outside of South Africa tend to be very young, with most institutions dating from the post-colonial period in the second half of the 20th century. Only a few decades later, the World Bank and the IMF implemented structural adjustment programmes that marginalized higher education in favour of primary schooling. Just as the higher education system was expanding in the global North, universities were starved of funds and thus of the continuing growth that they really needed to function effectively. In the 1980s and 90s, if you appreciated diatribes that could sear your eyeballs, you had only to listen to African intellectuals at conferences berating structural adjustment, the World Bank and the IMF.

In around 2000 came the solemn revisionist discovery that in fact higher education was very important and should be supported. This was not long after South Africa emerged from apartheid, so the 21st century in our region has been a period of reconstruction of battered higher education systems. In these circumstances, there were two conflicting ambitions in the minds of university leaders. One was to be able to [re]join the global academic community, building prestige and recognition in the competitive terrain of publication impact and university rankings. The other was to demonstrate a contribution to national and regional development in countries that faced overwhelming challenges.  Or, as a university administrator put it to us recently, the biggest battle southern African universities face is to combine the achievement of both prestige and relevance.

One has to ask why this is so difficult. The answers are to be found in the ideology and global hierarchies created by the corporations that had come to dominate the journal publishing system. Open access developed in good part in reaction to the dysfunctional nature of this bloated system and its vertiginous subscription rates. In the global North, the debate has largely been about how to get visibility for journal articles locked behind paywalls in these journals and less about hierarchies of knowledge that marginalized a great deal of world research.

 The impact factor

The reason that the two goals of local and international impact are irreconcilable in our region has a lot to do with the universal adoption of the ISI Impact Factor as the dominant metric used as the preferred route globally for achieving research recognition in a managerialist and competitive higher education system. The IF is also, in the developing world, the biggest barrier to the achievement of recognition.

I have enjoyed the targeted and tough attacks on the rigour and credibility of the Impact Factor by Stephen Curry and Bjoern Brembs in recent blogs. As someone who flinches when I hear the term ‘bibliometrics’, I appreciate Brembs’s comparison of the IF with ‘homeopathy, creationism or divining’ and would readily encourage the adoption of Curry’s series of declarations:

  • If you include journal impact factors in the list of publications in your cv, you are statistically illiterate. 
  • If you are judging grant or promotion applications and find yourself scanning the applicant’s publications, checking off the impact factors, you are statistically illiterate

The trouble is that the IF is taken with deadly seriousness in southern African universities and is all too often used as a proxy by international agencies for the evaluation of national research effectiveness. Lesser known is the extent to which the criteria for inclusion in the ISI are skewed to actively exclude the developing world and its interests from this dominant system. As Guèdon describes, by an extraordinary sleight of hand a distinction was made by the ISI in 1982 between ‘local’ and ‘international’ or ‘mainstream’ science. ‘Local’ or national research was relegated to a lower level – or irrelevance – and only ‘Third World contributions to mainstream science’ would be considered for inclusion in the ISI.  In other words, a research article gets into the ISI if it addresses the interests of readers in the English-speaking North.

In this way, the research interests of three quarters of the world are relegated to irrelevance in the dominant global scholarly publishing system and downgraded to the ‘local’, while whatever is included in the Science Citation Index constitutes mainstream science. Effectively, Nelson Mandela becomes ‘local’ and George Bush ‘international’.

We could perhaps learn from Groucho Marx giving the finger when he was refused membership at a country club: ‘I don’t care to belong to a club that accepts people like me as members.’

 Institutional repositories

What this system has inevitably meant is that there are very low rates of publication in the ISI from the countries consigned to the periphery. This is particularly the case in the smaller African countries outside of South Africa, which dominates African production in the ISI. Does it make sense to maintain an OA institutional repository to profile ISI journal articles when many institutions are producing between 25 and 100 articles a year? And in institutions that are battling with issues of capacity and infrastructure in the wake of the policy turbulence that I have described, should these repositories be institutional or subject-based? All in all, the role of repositories – regional, institutional, subject-based, archival – may well be different to the common assumptions about journal article deposit and mandates.

 Beyond journals – local relevance

While the road to prestige and rankings is delivered largely through the commercial journal system, the way to social and development impact is surely through open access, with the ‘grey literature’ that Finch recognized at its heart. In our context, in Africa, access to the non-journal literature is not, as Stevan Harnad averredmerely providing access to research data and grey literature’ that the Finch Commission recommended for inclusion in repositories (my emphasis). Rather, in the developing world, access to this kind of research output is a very important part of ensuring the relevance of research to local and regional needs – or indeed ensuring the wider impact of research outputs in ways that have recently been recognised in the World Bank’s OA policy.

What this does raise is the question of peer review. Universities are anxious to ensure the quality of the research that gets placed online and are exploring alternative quality evaluation approaches to deliver this goal.

 The question is, therefore, what open access policy should look like in the developing world and in southern Africa in particular. This is likely to  include open access publication; the recognition of a wider range of  research outputs; repository and communication strategies that recognize this and which take account of the realities of available capacity and infrastructure. And – a challenging issue – how to change reward and recognition systems to bring them into line with the real strategies of governments and institutions?

Through the looking glass? Scholarly publishing seen from the South-eastern frontier

I have given a couple of workshop papers on scholarly publishing in the last ten days or so. Sounds dry, doesn’t it? That might apply to one of those ‘How to crack the system and get published in an accredited scholarly journal’ papers that I think my audiences were expecting (and dreading). But if one casts a steely eye over the system that we all take so unquestioningly for granted, then things can get a lot livelier. What would the proverbial woman from Mars make of it? The basis of the academic accreditation system is that our scholars are assessed and promoted primarily according to their ability to get published in journals in other countries, whose systems are patently weighted to exclude them – and to exclude many of the burning issues that might be of national relevance.

Looking at the scholarly publishing system from the perspective of a scholar in the humanities and social sciences in a South African university, one thing is for sure – we really are on the margins of the world. This is not a system conceived of for the benefit of our developing world, but designed to suit the needs of powerful institutional and hard-nosed commercial interests in big first-world economies. If one looks dispassionately at what the universities put into this system and what they get out of it, it is patently dysfunctional. Scholars pay for conducting the research, writing the articles, for acting as peer reviewers, then pay page charges, as often as not, to get published. Then the universities buy back that information, often at vastly inflated prices, from near-monopoly conglomerates, operating a commercial system in which market forces don’t work.

The way we relate to the system has its absurdities. We bind ourselves into trying to publish primarily in journals selected by indexing systems that explicitly marginalise contributions from the peripheries of the world, where we live. Also, as a CSIR researcher pointed out in one of last week’s workshops, the numbers just don’t work: “There are a thousand-odd researchers in the CSIR”, he said, “and we are each required to publish two articles per year in accredited journals. Where are we going to find accredited journals to publish more than two thousand articles from the CSIR alone? And who is going to read them?”

The scholarly publishing system is not working even in the powerful knowledge economies that call the shots. As the leading radical IP lawyer, James Boyle, said at the iCommons Summit in Rio, “We have a scientific publishing system that is massively dysfunctional and really, really broken.” Or, as Lindsay Waters, the Humanities Editor at Harvard University Press put it in one of his papers; “The patient is dying! Call the ambulance.”

So where does that leave us, here in South Africa? And what can we do about it?