Return to Table of Contents

Science and Circumcision

Basic tenets of science: These can be used as a litmus test for claimed benefits.

  1. Conclusions based solely on known flawed data cannot be considered valid.
  2. All studies have known flaws - flaws of commission and/or flaws of Omission.

The worth of a theory is determined by its ability to make accurate predictions.

The rates of these are HIGHER in many circumcising countries than in many intact countries, hence the alleged predictions are contradicted by empirical evidence. Unlike "medical science," SCIENCE demands that prediction be fulfilled EVERYTIME to be credible.

Elaborating on both the flaws and failure of predication, here is a critical analysis of recent "studies" purporting that circumcision reduces the incidence of HIV.

Randomized Controlled Trials

In SCIENCE, a (R)andomized (C)ontrolled (T)rial is a trial where all of the factors are CONTROLLED with one, and only one factor (chosen at random) altered, and the results are examined.

For the Circumcision/HIV studies, we have the following:

None of the factors are CONTROLLED. These factors are examined, only based on self-reporting (notoriously unreliable) and then examined STATISTICALLY.

The only things that might have been random were those chosen to be circumcised.


  1. The time needed for healing for those circumcised BEFORE the trial was initiated - less exposure time.
  2. Rates of exposure for each group
  3. Dry sex:
  4. Anal sex
  5. Homosexual sex
  6. Genital warts reoccurrence: how they were treated, and the final efficacy of that particular treatment. (Excision or chemical)
  7. The accuracy of the tests to determine seroconversion--rates of false or negative determinations
  8. The time needed to manifest all seroconversions
  9. Any chance of a follow-up as the test was stopped early and ALL subjects were circumcised.
  10. Equal amount of "safe-sex counseling" for both groups:
    1. Postoperative follow-up visits were scheduled at 24–48 hours, 5–9 days, and 4–6 weeks.
    2. All participants in both groups were followed up at 4–6 weeks, and at 6, 12, and 24 months post-enrolment

Control is control, "playing with the numbers" is merely an accounting scheme.

And all of this involved nothing but statistical analysis.

Author bias is a common problem with many studies (and all authors have long been circumcision advocates):

Another shortcoming of the studies is the small sample size.

Small sample size: With few subjects, the law of small numbers applies with a vengeance.

(Roughly, when you're dealing with small numbers, random variations assume disproportionate importance. IE, there might be 3 murders in a small community one year and 14 the next, but it's stupid to say "The murder rate has more than quadrupled!" and blame the difference on policing, penalties, or anything else. Next year there might be 7 or 1.)

Finally, enough men dropped out of the studies before completion to completely nullify any claimed result.


And in an attempt to lend credence to these studies, there are SPECULATIVE mechanisms given for this claimed reduction. None of these mechanisms have any scientific and logical support--most have already been refuted by the facts and evidence.

  1. "The protective effect of circumcision against HIV infection is thought to derive in part from postsurgical development of a layer of keratinised squamous epithelial cells that limit viral entry to underlying HIV target cells. How long it takes the residual tissue to fully heal and become keratinised has not been studied.
    1. Keratinization is a long term phenomena and could not have any effect on conversion is the short time period of the studies.
    2. Some circumcisers claim that keratinization (and the loss of subsequent sensation) does not occur: How does male circumcision protect against HIV infection?
      "There is controversy about whether the epithelium of the glans in uncircumcised men is keratinised; some authors claim that it is not,15 but we have examined the glans of seven circumcised and six uncircumcised men, and found the epithelia to be equally keratinised."
      In a correspondence to Short, Szabo claims to have done a study, but this study was never published. He claims the epithelia of the exposed glans is not keratinized, but infers (without any evidence) that the remaining foreskin of the circumcised penis is and that of the intact foreskin is not.
  2. The inner preputial mucosa is unkeratinised, making it vulnerable to HIV infection."
    1. The CDC has shown that undamaged epithelia do not transmit the HIV virus: Mechanisms of HIV Transmission through Epithelial Cell Barriers
    2. Ironically, since keratinization logically causes a further loss of sensation and sensitivity, an ardent circumcises claims to have done a study (never published , and not seen by anyone) proving that keratinization (and the loss of subsequent sensation) does not occur: How does male circumcision protect against HIV infection?
  3. That circumcision reduces the risk of male HIV infection is biologically plausible. The foreskin is rich in HIV target cells (Langerhans' and dendritic cells, CD4+ T cells, and macrophages).
    1. Erroneous speculation on many levels.
      1. The langerhan cells produce langerhin, which kills the HIV virus
      2. Macrophages digest the HIV virus
      3. The foreskin produces lysozyme which also kill the HIV virus
  4. The foreskin is retracted over the shaft during intercourse, which exposes the inner mucosa to vaginal and cervical fluids.
    Mechanisms of HIV Transmission through Epithelial Cell Barriers
  5. Also, breaches in the mucosa can occur due to microtears during intercourse, especially at the frenulum,

    False, in fact the opposite is the truth, a tight circumcision results in more abrasive and friction laden sex which would lead to more breaches in the tissue.

    The case against circumcision

    [also see p.61, May issue, Men's Health] The work of Laumann et. al JAMA 1997, 277: 1052-57, Taylor & Lockwood, BJU, 1996, 77:291- 5, and Halata & Munger, Brain Research, 1986, 371: 205-30 explains:

    The mutilated man with his keratinized, desensitized glans, and absent the fine-touch receptors and erogenous mobility of the foreskin, ultimately requires inordinate stimulation of his residual penile nerve endings to achieve pleasure and orgasm. When this becomes plunger sex resulting in dryness, abrasion, pain and bleeding, female orgasmic potential shrivels. This requires we take a closer look at the notion of vaginal dryness as a ‘female’ problem.
  6. Uncircumcised men are more susceptible to genital ulcer disease, which could increase HIV entry.
    1. Actually the opposite is true
    2. STD's
    3. Failure of prediction-In science, it only takes ONE exception to invalidate a hypothesis or theory.

Circumcision rates:

Japan <1%  
USA 80.00%  
Ethiopia 100.00% Circumcision and HIV: A lie will be halfway around the world before the truth has got its pants on.

HIV rates:

Japan 0.10%
USA 0.60%
Ethiopia 4.40%
Protective factor: 1/0.5 =2X

US vs. Japan

USA (80% X 2) + (20 X 1) = (160) + 20) = 180
Japan (1 X 2) + (99 X 1) = (2) + (99) = 101
Protected ratio: 180/101 =1.8 LOWER in the USA.
Reality rates: 6X HIGHER in the USA
Discrepancy: 6.0 X 1.8 = 11X = 1100% error

Ethiopia vs. Japan

Circumcision rate:

Ethiopia 100.00%
Japan 0.40%

HIV rate:

Ethiopia 4.40%
Japan 0.10%
Ethiopia (100 X 2) = 200
Japan (1 X 2) + (99 X 1) = (2) + (99) = 101
Protected ratio: 200/101 = 2X LOWER in Ethiopia
Reality rates: 44X HIGHER in the Ethiopia
Discrepancy: 2 X 44 = 88 X = 8800% error

The largest acceptable error in science is 1 sigma = 5%

Those advocating circumcision to reduce the incidence of HIV either ignore this empirical and contradicting evidence, or try to dismiss it with various excuses, like: Intravenous drug usage or homosexual activities.

However, they never provide any hard numbers of these factors for various countries or a model that should include them to explain this discrepancy.

For these excuses to be credible, one would need to assume:

That the US has 11X as many homosexuals than Japan; and Ethiopia has 88X as many as Japan—when it is widely accepted that the rate of homosexuality is 5 –>10 % in all cultures...


That even though people in Ethiopia can barely afford food and shelter, they can afford 88X the recreational drugs than Japan and 8X that of the USA.

For a critique of epidemiological studies in general, here is one:
Epidemiology critique

Information quoted on this site is with the written permission of the authors.

Return to Table of Contents