Africa lags behind other developing regions in local collaboration — and in building its own scientific capacity, says Linda Nordling.

Collaboration was a time-consuming affair for the pioneers of scientific discovery. It took nineteenth century naturalists months to receive replies to their correspondence.

Today, such missives are sent with the click of a button. Months of waiting have shrunk to seconds, and telephones and video linkups have made real-time interaction a cinch.

It is not surprising, then, that the 'average collaboration distance' between scientists working together on projects has skyrocketed in the past thirty years — from 334 kilometres in 1980 to more than 1,500 kilometres in 2009, according to data published by scientists at Leiden University in the Netherlands and Stellenbosch University in South Africa last year. [1]

But the global trends mask a more intriguing regional picture. For developed countries, the increase in this rather unorthodox measure probably reflects the increasingly globalised world in which their scientists operate. But for Africa it's not quite that simple.

Forging new ties

Traditionally, scientific activity in many parts of the developing world has had strong links with — and in extreme cases even been steered from — faraway science hubs in Europe and North America.

For such countries, many of which have seen research output rise sharply in recent years, a decrease in average collaboration distance could reflect their growing scientific capacity. It is a sign that internal networking is catching up with, if not outpacing, long-distance networking.

The data published by the scientists from Leiden and Stellenbosch support this explanation. The mean collaboration distance of scientists in France, the United Kingdom and the United States had an annual growth rate of 4.9 per cent, 6.7 per cent and 4.7 per cent respectively between the years 2000 and 2009 — reflecting a tendency to collaborate with more faraway colleagues.  

For emerging economies whose scientific output has boomed, however, the increase in collaboration distances is modest, or even negative. China clocks up 1.1 per cent annual growth between 2000 and 2009, and South Korea 1.7 per cent. But in Brazil and Malaysia, average collaboration distance has been decreasing by 2.7 per cent and 2.2 per cent respectively every year.

The modest or shrinking collaboration distances in the world's emerging science powerhouses means that these countries, and their surrounding regions, are able to grow their local science base without roping in experts from the West.

But not in Africa

Unfortunately, such signs are difficult to see in Africa.

Data shared with SciDev.Net show that, of the eight African countries with rapidly growing scientific output surveyed by the scientists, only Ghana has the low growth in collaboration distances (1.8 per cent each year) seen in emerging economies in Asia and Latin America.

In the remaining countries surveyed — a heterogeneous bunch including Cameroon, Egypt, Ethiopia, South Africa and Uganda — annual growth rates range between 3.2 per cent (Egypt) and 5.5 per cent (Uganda).

The researchers did not include many African countries in the study because their scientific output was considered too small to be statistically meaningful.

The data chime with other reports showing that African scientists work with colleagues abroad more than with each other. In 2010, Thomson Reuters [2] identified the United Kingdom and the United States as the two top collaborating countries for six African countries including Kenya, Nigeria and South Africa.

What is perhaps more telling is that none of the six African countries in the report made each other's 'top five'.

The full picture

But these data do not show the full picture. While they reflect the fact that non-African partners still loom large in the continent's science landscape, African scientists are increasingly working together.

In fact, collaboration on the continent has blossomed. But often, it still involves one or more non-African partners, which keeps overall collaboration distances long.

For example, Stellenbosch University's collaborations within the African continent have increased by a factor of at least three in the past five years, says Christoff Pauw, manager of international academic networks at the university.

But these links are still overshadowed by opportunities for in-country collaboration — perhaps better here than other Sub-Saharan African countries — and by well-established long-distance connections.

Curing an African disease

For a South African university, such links are not a sign of too much steer from abroad. But for universities in under-resourced countries, the long collaboration distances are a symptom of one of the chief ailments hampering long-term scientific endeavour: reliance on foreign funding.

Outside funding has driven the continent's recent networking, such as the 'Africa Call' of the European Union's Framework Programme, which funds collaborations between African and European scientists. Similarly, where nationally funded science grants do exist, they do not encourage collaboration.

"As long as this goes on, the opportunity to narrow the collaborative distance with co-publishing, deepening research networking and establishing a tradition for closer relations in research will continue to be difficult," says Mammo Muchie, an Ethiopian international development scientist based in South Africa.

There are moves in the right direction. Both Senegal and Tanzania are busy setting up national funding bodies that will distribute grants more efficiently and — if they are clever — promote collaborative efforts both within and between African countries.

Long-distance partnerships have their place, but a drop in average collaboration distance would be a welcome sign of African science coming into its own.

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, Nature and others.