In this book we have discussed and analysed a range of language contact situations. Most of these dealt with circumstances where closely related varieties have come into contact with each other. Our focus has been primarily, but not solely, upon the English language and its history. In this final chapter a number of the issues which have been raised earlier in this study will be considered. From this it is hoped that consensus can be reached over some of the processes and outcomes involved in close-relative contact, whether the original inputs are considered dialects of the same language or as discrete, but related, languages.

In the first part of the book we concentrated on contact between dialects of what would normally be analysed as the same language. Various theoretical models were introduced, most of which were found to have a genuine bearing upon many of the issues involved. No one model could explain all of the features involved in the process of new dialect formation, however. The founder principle proposed by Mufwene, for instance, is an intriguing concept which helps explain the nature of the development of many new dialects. The problem comes, however, when we find that multiple founder populations appear to have made distinctive contributions to the development of a variety; moreover, initial founder impulses may be replaced by later forces. By the same token, swamping as an explanation is not entirely convincing (although nor is it anything near worthless). Indeed, at first glance the opposite of the founder principle, since swamping assumes that the influence of earlier linguistic inputs on a new variety can be erased by a (metaphorical) flood of speakers whose language is either nearly or wholly homogenous, the two theories are actually fairly compatible. As Mufwene has constructed his theory, it is not the first population which necessarily provides the blueprint for the new dialect; rather, it can be the first significant population which acts as the founder. A swamping event is perfectly in tune with such an analysis, yet it cannot be said to be applicable to all situations.