27 May 2010 | EN | FR
African scientists are less isolated from each other now thanks to regional networks, but they need broader scope, says Linda Nordling.
It is often said that African research suffers because scientists are isolated from their colleagues elsewhere on the continent. For decades, scientists in Abuja or Nairobi have been more likely to collaborate with researchers in the United States than with each other.
But this fragmentation could be coming to an end. Intra-continental links do now exist, albeit in certain groups, as an overview of African research networks published last month by Thomson Reuters shows.
The report, Global Research Report: Africa, looks at scientific papers published between 2004 and 2008, and identifies four subcontinental groupings with strong research connections.
The first lies in North Africa and includes Algeria, Egypt, Libya, Morocco and Tunisia.
A Francophone, largely West African, group centres on Cameroon and includes Benin, Burkina Faso, Congo, Côte d'Ivoire, Mali, Niger, Senegal and Togo.
Language — English this time — also plays a big part in a third group that links Nigeria, the Gambia and Ghana in West Africa with Botswana, Ethiopia, Kenya, Malawi, Tanzania, Uganda, Zambia and Zimbabwe in the east.
And South Africa provides the gateway to group four, which includes Gabon, Lesotho, Madagascar, Mozambique, Namibia, Sudan and Swaziland.
Each group centres on one or two scientific powerhouses that provide the hub for regional collaborations: Egypt, Cameroon, Nigeria and Kenya, and South Africa. This is not surprising as they are the strongest science nations in their geographical areas.
But how far has Africa come in breaking down this major research barrier?
History weighs heavily…
The picture that emerges from the Thomson Reuters study is that African research collaborations remain heavily dependent on — and limited by — the colonial past. This is particularly true for the North Africa group, which maintains strong links with Europe. Algeria and Tunisia, for example, have unique connections with France, co-authoring an exceptional share of publications.
The barriers to creating networks are not only rooted in language, culture and history but also in geography: most intra-continental networks exist among countries that lie close to each other — with two notable exceptions.
One is Nigeria, which, while having important connections with its West African neighbours, occupies a more significant position in the Anglophone network and also connects strongly to South Africa.
And South Africa, which is identified by Thomson Reuters as the "outstanding research leader" in Africa, is the only country to participate in all four groups.
But overall there is an absence of connections between groups, which has important implications.
For example, Egypt's scant research links with Sub-Saharan Africa will do little to help its ambition to push the pan-African science agenda (see Egypt says it will propel African science agenda).
Networks built on historical relationships are artificial, adds Mammo Muchie, professor of innovation studies at the Tshwane University of Technology in South Africa.
"We don't want these silos. It not only short-changes Africans, who should be able to collaborate freely, but also foreign collaborators, who only get access to a limited part of Africa's expertise," he says.
…but globalism is the future
The authors of the Thomson Reuters report do not see the current networks and regional hubs as a problem. They present the networks as "development opportunities" with the strongest countries having "a potential transformational role" to provide the glue for more links.
"The future of the African research enterprise must depend to some significant extent on the ability of these countries to help facilitate further growth through leadership, strong local investment and the creation and support of key facilities and centres to draw in and assist currently less well resourced partners," they conclude.
The authors add that the strongest nodes will be critical in linking African networks to global ones. For example, Nigeria is geographically, if not linguistically, well-positioned to extend its links westwards towards South America, and Egypt is a link to the Islamic research world.
These networks have implications for policymakers, says Caroline Wagner. She is the author of The New Invisible College, in which she argues that country-based science is giving way to international networks. She urges governments to move beyond national science policymaking towards networked models of science that focus on common problems.
But research links are not an end in themselves, she says. "To seek interconnectivity between African countries simply to have links isn't necessarily going to promote development. If knowledge creation in Africa is about solving problems, then the important links are the ones that will help countries access or produce solutions to concrete problems."
Regional research hubs, built to address common problems, could be the best strategy for achieving this.
Building on momentum
Of course, the Thomson Reuters study does not provide a complete picture. It is limited to an analysis of internationally indexed journal articles and so excludes much — perhaps the majority — of African science, which more frequently appears in local journals (see also Raise the profile of research in local journals).
But whether the links are purely regional or not, it is clear from last month's report that African researchers are beginning to work together more widely.
Efforts to boost science on the continent must capitalise on this momentum. Capacity building should not only stimulate the potential of scientists and their institutions to carry out research, but also promote the networks and linkages that are essential if Africa is to build its own knowledge societies.
Journalist Linda Nordling, based in Cape Town, South Africa, specialises in African science policy, education and development. She was the founding editor of Research Africa and writes for SciDev.Net, The Guardian, Nature and others.
All SciDev.Net material is free to reproduce providing that the source and author are appropriately credited. For further details see Creative Commons.