Mon séjour en Afrique du Sud (Cape Town)

This was a major victory for people with HIV in SA and internationally. It established in the military what was already accepted in all other areas of employment: that people with HIV should not be discriminated against solely on their HIV status.

The victory has been widely acclaimed but there is one insufficiently publicised issue. In response to hundreds of pages of expert medical evidence put forward by the applicants, the SANDF's main "scientific" defence of its HIV policy was an undated, unpublished study conducted during a Zimbabwean military training course.

This study, in its own words, "(sought) to prove the hypothesis that the type, intensity, frequency and duration of exercise" in a military environment will lead to "opportunistic infection(s) and an early onset of (full-blown AIDS) among HIV-seropositives". In other words, the study's goal was to subject people with HIV to conditions directly harmful to their health.

This is contrary to even the most basic grasp of medical ethics, which stand and fall on the principle of "do no harm" and is underscored by the fact that six HIV-positive participants died as a result.

It is alarming that Surgeon-General Vejay Ramlakan had little understanding of the ethical problems inherent in this study, else it would not have held such prominence as the main point of defence. Leslie London, a member of the National Health Research Ethics Council and a professor at the University of Cape Town, argued that the fact that the study made no statement as to whether antiretrovirals were made available to participants invalidated its relevance and raised echoes of the Tuskegee study perpetrated on African-American men in the US. The Tuskegee study is infamous for infecting men with syphilis and stopping them from getting treatment, to map the progression of the disease. More than 100 participants died as a result.

Here, if this study was done without access to antiretrovirals, it is inapplicable to the SANDF, where they are available. More worrying is the possibility that antiretrovirals were available, but were withheld to track the declining health of the HIV-positive participants. If so, it is a gross violation of medical ethics.

While the ethical failings of this study alone call into question any reliance on its results as being biased and untrustworthy, it is also worth noting how London's affidavit dissected in minute detail the fundamental errors in its methodology, data collection, data analysis, presentation, and interpretation of its findings.

Based on this, London concluded "the study may have been entirely fabricated -- in other words, that there was no such study conducted or that there was a study that was so poorly done or returned undesired results that the researchers simply filled in results suiting their a priori hypothesis".

The constitution sets a high standard for respect of human rights by all parts of the government, including the SANDF. It is unacceptable for high-level SANDF officials not to critically evaluate their "evidence" to ensure it is in line with these principles. An unethical study stays unethical, regardless of who commissioned it, who conducted it or who brought it to their attention.

This was lost on the surgeon-general, whose response to the critiques set out above was that "the SANDF did not commission or participate in the Zimbabwean study ". This is a wholly inadequate response, as it still wil fully ignores the study's basic scientific failures.

As London said: "Critical appraisal of research is a standard competency our medical and other health professional students are expected to achieve (to) distinguish good-quality evidence from poor research (not fit) for clinical decision-making or policy development." This study should never have made it to the courtroom. It rather belongs in a medical ethics classroom as an example for future researchers of how research can fail so disastrously to respect the dignity and human rights of research participants.

For the government to rely on the results of flagrant human rights abuses is to advocate for such abuses to occur in the future. Real medical science requires careful analysis and respect for the people involved, neither of which is shown here.

There is already a long history of denying the integrity of medical science by this government, especially as it relates to HIV. This is yet another unacceptable example.

Here, though, Ramlakan has gone the further step of trying to mislead and confuse the judiciary on the state of medical science, and ultimately to establish in law that unfair discrimination is acceptable under the constitution. It is not; nor is this conduct. SA has been down that path before.

Honermann is Tolan Fellow in International Human Rights at the AIDS Law Project.