Introduction

The immediate impetus for this contribution is an article published as Lass (1977), but which I had the privilege of hearing presented in 1975 at a meeting of the Philological Society – albeit as viewed from the perspective of a quarter century later. In this article, Lass takes a devastating look at the reliability of internal reconstruction as a method for seriously reconstructing the linguistic past. An added stimulus is the article Lass (1993), which applies similar argumentation to that used in the 1977 article to the comparative method. Perhaps somewhat surprisingly for a contribution that I am dedicating to Roger Lass, my own contribution will take issue with some of the conclusions that Lass draws in these articles, in particular the former. But I believe that the contribution is nonetheless appropriate; indeed I hope it may even be welcome, because I have learned much from trying to formulate more explicitly the points of agreement and disagreement and because I hope that my contributions, like Lass's, will push the debate forward. Moreover, as I will try to show both in the body of my contribution and to summarise in the conclusions, my approach stems from an acceptance of much of what Lass says in these two articles, while asking a different set of questions from those posed by Lass, and thus almost inevitably coming to somewhat different conclusions.