Your editorial (see Can Africa pioneer a new way of doing science?) argues that 'mode 2' science (science that promotes social inclusion and alleviation of poverty) is more appropriate to Africa's needs than 'mode 1' science (science driven by purely academic interests).

On the face of it, this line of thinking does make a lot of sense. However, the devil is in the details. Who decides (and how do they decide) that a particular research project is mode 2 science? Even if we all agree that social inclusion and poverty alleviation are the measures by which research in Africa should be assessed, how does this translate into practice?

For example, when a research proposal comes before a government-funded research body — such as the National Research Foundation in South Africa — how does this organisation decide that the proposal satisfies the criteria for mode 2 science? Sure, they could use a checklist of 'objective' criteria to score each proposal but how reliable would this be? And how do we know that the benefits would outweigh the costs?

The danger is that some research projects — that are legitimately part of the big picture of poverty eradication — might be excluded from funding because they fail to satisfy the review committee. Let me explain my argument with the following scenario: when a scientist embarks on a research career he often has a 'big picture' in mind, for example, the inefficiency of reproductive cloning. If this technology were made more efficient, it could be of enormous value to the agricultural and conservation sectors of developing countries.

The first step for this scientist is to develop a hypothesis. They might plausibly propose that the inefficiency of cloning is linked to inefficient epigenetic programming (affecting a cell, organ or individual without directly affecting its DNA) before the embryo is implanted in the uterus. The researcher might then submit a proposal to test this hypothesis and submit a proposal to a funding agency that supports mode 2 science.

On the face of it, what relevance does this research have to poverty alleviation? Clearly a link can be made because revenue of the agricultural sector could rise if cloning were more efficient. How will the panel know this? The answer depends on how explicit the researcher makes the link. Thus, those proposals that are approved will be those written by people who are well-informed about the government's mode 2 science policy and by those who are sufficiently creative in making the necessary connections.

Another pitfall of mode 2 science policy is that it makes the presumption that mode 1 research does not contribute to a country's immediate social needs. However, academic institutions are centres for training, as well as research, and the skills imparted to their students — however arbitrary the nature of their research — can be reapplied to mode 2 science. Furthermore, mode 1 research can lead to serendipitous discoveries that could not be anticipated at the outset of the research.

Thus, by excluding mode 1 science we are impeding our future potential for unexpected innovations. Mode 2 science policy is good on paper; but we need to be aware of the potential pitfalls of putting research into 'worthy' and 'unworthy' categories. This approach seems overly simplistic because worthiness and unworthiness exist in every academic institution and research endeavour. Those who are best able to hide the latter will be those who benefit the most from mode 2 science policy.