Letter to Scientific American, 19 February 1996.

John Rennie, Editor-in-Chief
Scientific American Inc
415 Madison Avenue
New York, New York

February 19, 1996

RE: Caldwell JC. Caldwell P. The African AIDS Epidemic. Scientific American (March 1996): 62-68.

To the Editor:

The Caldwell's survey of the literature on the African AIDS Epidemic ignores several key studies. De Vincenzi and Mertens of the World Health Organization reviewed 23 studies that suggest that male circumcision may be an associative factor for HIV infection in Africa. They conclude that the methodological errors in these studies preclude any conclusions about the alleged efficacy of male circumcision.

A recent study by Grosskurth, et al. of the London School of Hygiene and Tropical Medicine found that circumcision had no effect on HIV infection. It is instructive to note that the idea that male circumcision could prevent HIV infection was invented in 1987 by an American circumcisionist who hoped to influence the American debate over circumcision. The studies which later attempted to prove this hypothesis using African villagers and prostitutes as subjects are also by North Americans who have strong cultural and religious biases in favor of male circumcision. The very notion of considering circumcision as a factor in the first place reflects this bias. Australia too has had a history of infant male circumcision. Could the Caldwell's bias evinced by their exclusion of contradictory data be a reflection of this unfortunate history?

The Caldwells seriously misrepresent the 1952 study by Asin. Asin did not conclude that chancroid was less prevalent in circumcised soldiers because they were circumcised, nor did he conclude that the intact penis is harder to clean. That 4% of the 1,402 soldiers studied (71% of whom were Black) were circumcised merely reflects the circumcision rate among adult Americans at that time. The statement that 33% of U.S. soldiers in Korea were intact is false and found nowhere in the medical literature.

Researcher bias is a serious problem in any study or survey. In Africa, male circumcision and female circumcision occur in the same communities. An African or Moslem researcher with a bias in favor of female circumcision could just as easily use the maps presented by the Caldwell's to suggest that it is female circumcision that protects against HIV. Amputating parts of the reproductive organs of either sex will not prevent venereal infections. Only education will accomplish this goal.

Paul M. Fleiss, MD, MPH

Frederick Hodges


  1. de Vincenzi I. Mertens T. Male circumcision: a role in HIV prevention? AIDS 1994;8:153-160.
  2. Grosskurth H. Mosha F. Todd J. et al. A community traial of the impact of improfed sexually transmitted disease treatment on the HIV epidemic in rural Tanzania: 2. Baseline survey results. AIDS 1995;9:927-934.
  3. Asin J. Chancroid: A report of 1,402 cases. American Journal of Syphilis, Gonorrhea, and Verereal Diseases 1952;36:482-487.

[CIRP note: Fleiss and Hodges's complete letter, as submitted for publication, is presented here. Serious deletions were made by Scientific American before this letter was published. The effect of these deletions, intentional or not, was to portray Fleiss and Hodges's conclusions as unsupported and ridiculous.

A response from Caldwell and Caldwell was also published, which repeated the egregious errors of the original article. Of course the response failed to address Fleiss and Hodges's key arguments—they had been deleted!]

Cite as:
(HTML code added 22 October 1998)

Return to CIRP library