Abstract
In the last ten years the genes responsible for Huntington's disease and several other neurological disorders have been localized to specific chromosomes. Huntington's disease is now known to be caused by a gene on the short arm of chromosome 4. While these advances are welcome, they also pose a problem. All studies to date have been passive observational studies; none has been experimental. Arguably the most successful and significant attributions of cause in neurological history have been achieved by non experimental methods. This realization is somewhat disconcerting, considering that virtually all textbooks in epidemiology insist that true experiments (randomized trials) are the prime means of establishing cause. One cannot help but ask how these unauthorized successes were achieved. What are the principles of investigation that made them possible? In this paper we argue that the recent advances in neurological genetics, along with much else in human genetics, depend heavily on “specificity of effect.” Just what this principle is and how it allows successful attribution of cause is explained. Various applications elsewhere in the medical and behavioural sciences and in psychiatry are discussed.