However, the article reads like an uncritical and unreflective account of the experience of one white woman being invited to a funeral and attracting the interest of a drunken man while there, and does not seem to shed any light on the possible contribution of 'disco funerals' to HIV transmission, which is probably very small indeed.
The author appears to have believed everything she was told, and even found other published articles to support some of her claims. However, any kind of direct connection with HIV transmission seems tenuous for several reasons:
First, it is claimed that the Luhya of Western province and the Luo of Nyanza province engage in 'disco funerals'. Yet HIV prevalence is several times higher among the Luo than it is among the Luhya.
Second, these practices, as the author goes to some length to explain, take place in remote areas. Yet HIV prevalence is generally much lower in remote areas than it is in towns and cities.
Third, the entire account is anecdotal, it tells us nothing whatsoever about HIV.
Agreed, HIV is sometimes transmitted through unprotected sex, but not always. Evidence of 'unsafe' sexual behavior is not evidence of HIV transmission. Also, evidence of HIV transmission is not evidence of unsafe sexual behavior. The author of the article seems to have accepted both fallacies.
Clearly, there are social problems in these provinces, such as alcoholism, drug abuse, sexual abuse and the like, just as there are everywhere. But what is described in the articles and labelled as a 'cultural practice' sound very much like a funeral (albeit different from what the author may have experienced in Harvard, or anywhere else in the US).
The fact that some people drink too much and engage in various forms of behavior that can carry all kinds of risk, including sexual risk, does not make the events much different from parties, weddings and other get-togethers, that take place in many countries aside from Kenya, perhaps even in the US.
The term 'disco funeral' sounds very much like something made up by a journalist, perhaps similar to the one who wrote an article about this subject in IRIN, a publication that prefers a more sensationalist angle when addressing these 'issues'. But it seems unlikely that identifying social problems associated to a greater or lesser extent with funerals is the key to high HIV prevalence among some tribes in some places and low prevalence among other tribes in other places.
Various sources, apparently including academic journals, seem to publish just about anything about African countries, as long as it contains magic words like 'culture' and 'tradition', and florid descriptions of commonplace practices. But even identifying sexual practices that could be referred to as an aspect of 'culture' or 'tradition' does not necessarily tell us anything about how HIV is being transmitted.