Gordon Stewart's claim to have suffered from censorship because of his heterodoxy ("What happened to scientists' academic freedom?", THES, May 21) will have irritated epidemiologists, although perhaps few will consider it worth taking up.

I feel quite well qualified to comment as a mere ex-academic medic who has not sought any grant to work on Aids/HIV, but who was at the Royal Society symposium in 1989.

The models proposed there would lead to different numerical predictions under different initial conditions and parameters, so the call at the Royal Society was for data. But indeed Stewart's predictions based on minimal heterosexual transmission in the general UK population have been as close to the actual outcome as would be the result of any model on the same assumption.

Meanwhile, there has been the global pandemic, in which most transmission is through heterosexual sex, and the predictions of Sir Robert May, Roy Anderson and others for African countries have come to pass.

ÖйúAƬ

Stewart's disingenuousness or self-deception shows up in the way the phrase "developed countries" suddenly appears in his piece as an essential qualifier of his heterodox, unmentionable theory. This follows paragraphs in which this social factor has been conflated with the genuine scientific heterodoxy of Peter Duesberg or Kary Mullis (Letters, THES, May 21), whose belief that HIV does not cause Aids has faced what should be lethal falsification by an enormous weight of data.

If the United States is a "developed country", then much heterosexual and mother-child transmission of Aids has occurred in the socio-economic disaster zones of its inner cities, corresponding to the appearance in the recipient of viral nucleic acid (not just antibodies), in a challenge quite comparable to the flat-earther's when faced with pictures of our planet from space.

ÖйúAƬ

Stewart's thesis is the weakest possible evidence for censorship in science: I suspect and fear that something stronger could be found.

E. S. Cooper

London N4