It is not often that the cost of political leaders being wrong gets quantified. It's doubly important to bring science to bear on such a question when people have paid with their lives, as happened with the AIDS epidemic in South Africa under the leadership of then-president Thabo Mbeki.

An article from the Harvard School of Public Health AIDS Initiative published this week (P. Chigwedere et al. J. Acq. Immun. Def. Synd. 49, 410–415; 2008) provides such a service by estimating the benefits lost through underuse of antiretroviral (ARV) drugs in South Africa. Using modelling and clinical data of ARV-drug efficacy, it compares the number of people who received ARV therapy and of pregnant women who received treatment to prevent mother-to-child transmission of HIV between 2000 and 2005 with the number who could feasibly have been treated during the same period. It concludes that the lack of an ARV drug programme caused the loss of more than 330,000 lives — consistent with an estimate along different lines by the South African economist Nicoli Nattrass (N. Nattrass Afr. Affairs 107, 157–176; 2008) — and that 35,000 babies were needlessly born with HIV. The authors, by their own account, have been conservative in deriving these estimates.

The South African electorate has no more appetite for AIDS denialism.

Mbeki was deposed in September, and his successor, Kgalema Motlanthe, moved swiftly to replace Mbeki's chief accomplice, health minister Manto Tshabalala-Msimang, with Barbara Hogan. Hogan has rejected the dissident line, subscribed to by both her predecessor and Mbeki, that AIDS is not caused by HIV. Moreover, describing herself as “ashamed” about the estimates, she has declared that “the era of denialism is over completely in South Africa”.

This is almost certainly true. Motlanthe, who supported Mbeki's stand on HIV–AIDS at the time, has his support base in the country's trade-union movement, which was never happy about Mbeki's line on this issue. Jacob Zuma, leader of the African National Congress and a presidential contender in next year's election, also supported Mbeki's standpoint before he was fired by him. He now espouses only populist causes, and has moved on from denialism about AIDS: the South African electorate has no more appetite for it.

The needless deaths that occurred in South Africa prompt reflection on Mbeki's now infamous presidential AIDS advisory panel on the link between HIV and AIDS, the fate of which was chronicled in this journal in 2000–01. Its inconclusive report enabled Mbeki and his cabinet, who must bear collective responsibility, to portray this link as “deeply contested, and contestable”, to quote Nattrass. Certainly, the AIDS dissidents (much criticized by Nature in the past) couldn't wait to participate in the panel. But should orthodox scientists have signed up?

Even in retrospect, this is a difficult question to answer. Once leading South African scientists, such as Malegapuru Makgoba, then president of the South African Medical Research Council and an outspoken critic of Mbeki, had agreed to do so, others were bound to follow suit in support. In turn, members from outside of the country in good faith believed that their colleagues deserved similar support, and so agreed to participate. Ultimately it became clear that these efforts were a waste of time, as there was no possibility of consensus being reached among the panel's two diametrically opposed camps.

Mbeki used this lack of consensus to justify a national policy that refrained from rolling out ARV drugs until late 2003, although some of the country's nine provinces, which enjoy a level of autonomy on matters relating to health, defied this with varying degrees of success. Even as late as 2006 the South African health department's stand at the International AIDS Conference was extolling the virtues of garlic, beetroot and lemon juice as a solution to the epidemic.

In retrospect, the panel, constituted as it was, should never have been supported. Yet several of the country's key scientific institutions explicitly endorsed its establishment, and also desisted from criticizing Mbeki. Along with his cabinet, they bear some culpability for the consequences that have now been documented. There is a moral to this tragic tale that may prove relevant in other contexts. In a young democracy with a historically hierarchical culture, and with attitudes often hardened by a colonial past, scientific institutions need not only to guard their independence fiercely but also to make their reasoned voices heard above the fray of political sycophancy.