Accounts of the evolution of human language must, by their very nature, express claims of a historical sort: claims about why, when, where or how human language emerged and/or developed in some distant past. What is more, it is of the essence of those claims that they are put forward in the absence of direct evidence about the events and factors that may or may not have been involved in the evolution of language. In modern work on language evolution, however, scholars have come up with various strategies aimed at lessening or even solving this problem of evidential paucity. One of these approaches starts from the assumption that language evolution can be studied by examining other phenomena about which there is direct evidence. This article sets out, in informal terms, what it is to use such phenomena as windows on language evolution.