In order to describe the way in which a language has developed over a given stretch of time we require a theoretical framework (or model) within which the facts may be stated and explained. Ideally such a model should be capable of accounting for all the changes which have taken place in the language by reducing them to a systematically integrated set of rules. Any particular phenomenon will then be considered as ‘explained’ if we can state it in terms of these rules. All three chapters which follow will deal with the same subject-matter, namely the observable phenomena of language change; they will, however, differ in the way in which they describe these phenomena and in how they integrate them into an explanatory system. This diversity of treatment raises the obvious question of whether the mere restatement of some particular regularity in different theoretical terms is in itself any advance on previous knowledge. However, different models are bound to result in different questions being asked of the data and this in turn can lead not only to better explanations of already known facts but also to the discovery of new ones.

In this first, systematic, section of the book the emphasis will be on the linear development of language through time. The fact that contact between societies results in mutual influence will for the time being be disregarded, so that loan-words and areal features once adopted will be treated as forming an integral part of the language.