Trade-off logic helps explain why some sexually transmitted pathogens are more harmful than others. Competitors that exploit the hosts soon after infection will be able to generate more progeny; they may therefore win out in competition with any other competitors within the host and may be better able to infect any sexually contacted host than they would be if they exploited at a lower level.

They and we pay a price when they exploit intensively, but this price takes on a character that differs from that of the typical acute infectious disease. The tricks that sexually transmitted pathogens employ in their long-term persistence may eventually cause some essential part of the host machinery to fail. The particular part that fails usually depends on the parts of the body that the pathogen tends to inhabit. Because HIV lives inside immune cells, the more aggressive variants of HIV tend to cripple the immune system. Because herpes viruses tend to live in neurons, their aggressive variants tend to cause neurological problems. Because the syphilis bacterium can persist in the face of the immune system, its aggressive variants can invade most organs of the body and cause tissue destruction wherever they go. In the brain they cause mental illness; in the neurons along the spinal cord they cause lack of coordination; in the circulatory system they cause aneurysms.

Once again, this trade-off perspective raises the critical question, What factors favor the more aggressive variants over the benign variants? Evolutionary theory implicates the potential for sexual transmission. In some populations the potential is low because people tend to have fewer sexual partners or use condoms as a means of birth control and for protection from disease. Evolutionary theory suggests that when sexually transmitted pathogens find themselves in such populations, the particularly benign competitors—those that have genetic instructions for curtailed exploitation—are favored.

For the sake of argument, imagine that everyone in a population stayed with the same sexual partner for five years and then changed partners. If a pathogen lost its opportunity for transmission during the fourth year of the infection (because it was controlled by the immune system or the person died), it would be a loser. To break even, a sexually transmitted pathogen must be transmissible for a period extending into the duration of the next sexual partnership. To prosper, the pathogen must be transmissible for periods spanning more than one change in sexual partners. A low potential for sexual transmission should favor pathogens with such restricted activity inside a person that the infection would be transmissible over decades. Their restricted activity would undoubtedly lead to a trickle rather than a flood of pathogens out of the body, but a trickle would be sufficient. Although the trickle would tend to reduce the probability of infection per contact to low levels, under these conditions of high fidelity there would be years of sexual contacts during which transmission could occur. For a sexually transmitted pathogen, success is measured one sexual partner at a time. Until the sexual partnership breaks up, the pathogen cannot do any better than to infect that one partner. A low probability of infection per sexual contact is compensated for by a high number of sexual contacts per partner, and natural selection should favor a benign relationship.

Now imagine the other side of the spectrum of sexual activity. If people were changing sexual partners every week, the benign trickier would lose the competition. The competitors that are programmed to exploit more and reproduce more would be able to capitalize on the few contacts per sexual partner. Each sexual partner might have an increased chance of death or severe damage years down the line, but the loss of a single host is weighed by natural selection against the additional new hosts that could be infected as a result of the more aggressive strategy.

This evolutionary logic leads to an important prediction: sexually transmitted pathogens will evolve increased virulence in populations that have high potentials for sexual transmission. This prediction can be tested by assessing whether it accords with the information on variations in virulence that has been recorded in the medical literature. It does.