A matter of fitness

The primary aim of this book is stated in Section 1.4 as that of illuminating the Windows Approach and assessing it as a means of investigating the evolution of language. The preceding chapters have accordingly been concerned with clarifying the conceptual foundations of this approach in a concrete way. This still leaves the question of how good a conceptual tool it is for acquiring a better understanding of what the evolution of language might have involved. In this final chapter, I discuss the following overall features of the Windows Approach that need to be taken into account in addressing this question:

Overall features of the Windows Approach

1. F1: It has heuristic potential in two broad domains.

2. F2: It is not in an a priori way limited in regard to scope.

3. F3: Its conceptual foundations include appropriate constraints on empirical work on language evolution.

The discussion of specifically feature F2 will be seen to have a positive spinoff: providing further illustration of the varied nature of the putative windows that have been used in work on language evolution. Though the windows examined in the preceding chapters of the book are representative of the three main kinds of windows – analogue, correlate and abduction windows – they by no means exhaust the stock of such windows. But let us turn to features F1–F3.

Spread potential

As may be expected, the evolution of language is the primary domain in which the heuristic potential of the Windows Approach is meant to lie. Thus, as illustrated in some detail in preceding sections, this approach is geared to drawing sound inferences about what the evolution of language might or might not have involved. The heuristic potential of the approach is not restricted, however, to what it may reveal about the evolution of language. There is a second domain of phenomena in which its use may lead, albeit indirectly, to the acquisition of new knowledge: the domain of window phenomena. How is that possible?

Recall that, in terms of the Grounding Condition, inferential steps leading to conclusions about language evolution need to be firmly grounded in (analysed) data about window phenomena that are well understood. Some window phenomena have been studied in depth, with the result that there is a sufficient body of firm data about their properties.