Owen has a provocative link regarding two seperate approaches to combating AIDS meeting two different results:

Now some actual facts:

AIDS victims in 1987: Philippines 135 / Thailand 112

In 1991 the WHO predicted the Philippines would have 80,000 to 90,000 cases and Thailand 60,000 to 80,000 AIDS victims.

Thailand promoted the use of condoms in massive campaigns where Catholic Philippines promoted 'Abstinence' and 'Be faithful'.

The prognosis of the WHO was wrong for both countries:

1999: Philippines 1,005 / Thailand 755,000 AIDS victims

Source: British Medical Journal, volume 328, April 10th 2004

The question here is whether or not the success rate of condoms is outstripped by the increased number of people having sex due to promiscuity. To take a seemingly unrelated example, let's look at rugby and football:

Many people think of Rugby as an exceedingly rough and dangerous sport. However, the instance of injuries as a result of playing Rugby are, in fact, far lower than those sustained in American Football, and are generally similar to that of Football (soccer). The reason for this, first and foremost, is that any of the padding worn by American Football players is not for defense, but for offense. A secondary consideration is that the other side is also without pads, which moderates a large amount of dangerous behaviour. Finally, the rules, unlike in American Football and many other contact sports, as they do not limit the number of times a player/team may be tackled before surrendering the ball, do not encourage situations that exacerbate injuries, such as fighting for the last yard.

A simple example, from personal experience. American Football players are taught to tackle straight into the target's chest, often including significant pushing with the helmet. This is simply asking for neck injuries. Additionally, no serious rules are in place regulating the technique, leading to many dangerous tackling forms, as well as the exceedingly dangerous (due to its unexpected nature) blocking tackle. In contrast, Rugby players' tackles are done at the shoulder, usually against the target's legs, allowing that person's momentum to complete the tackle, rather than attempting to stop them and force them backwards (again, a product of the lack of "downs" in Rugby). Furthermore, a Rugby player may only tackle the ball carrier, and their form is strictly regulated- you can be ejected from a game for dangerous tackling (using the neckline of the jersey or otherwise going above the chest and/or failure to wrap your arms around the opponent's legs).

In comparing rugby to American football, Rory Miller makes a strong case that the increased protections of the latter have, in fact, lead to football being a more dangerous sport. With the protections, athletes are more reckless and rely on their equipment so much that the equipment's success rate at protecting athletes is undone by higher repitition of more dangerous physical hits.

Let's say that condoms have a 10% failure rate when it comes to stopping the transmission of HIV from an HIV-negative individual to an HIV-positive individual and 100 such sexual acts occur. That means that 10 people that did not have HIV acquire it. For the sake of argument, let's say that sex without condoms between an HIV-positive and HIV-negative person have 100% transmission rate. If 100 such sexual acts occur, you have 100 people that did not have HIV acquiring it. Now, if condoms enable 10x the number of people to have sex with people with the HIV virus, then you have 1000 people having sex and 100 people getting the virus. If the muliplier is less than 10, then condoms have proven successful at slowing the spread of the disease. However, if the muliplier is greater than 10, then condoms have actually assisted in the transmission of the disease by making people feel invulnerable and therefore being less cautious about who they are having sex with.

If the British Medical Journal is correct, and if there are not other factors involved to explain why the Philippines had more success than did the Thai, then it seems pretty clear that a condom-based approach is not successful. That does not necessarily say that the Philippine method would translate the same amount of success to the Thai people if the latter are more inclined towards sexual promiscuity. That'ss not out of the question, given that Thailand is known for some unsavory things in the sex industry. That would still suggest that cultural pressures against promiscuous sex are actually effective against extramarital sex, which is an argument that conservatives have often made and liberals deny.

It should be noted that I do not have a link to the British Medical Journal. Shea got them from an email, which means that the numbers could be a hoax. The only source I have found thus far is from a biased organization. I encourage anyone with disputing statistics to please make them available to me. There is more information regarding the low HIV rate in the Philippines here and the HIV problems in Thailand here. They don't contradict the BMJ numbers. It's also worth noting that with all the turmoil with AIDS in Africa, Uganda has had enormous success with an abstinence plus policy.

Regardless of the specifics, the multiplier statistic is useful to remember. I don't believe 10x is a particularly unreasonable number when you're comparing one set of people that believe they are not vulnerable to people with HIV and another set that don't have easy access to protection, but I would not be surprised if the multiplier were significantly higher or lower than 10.