Introduction

Although comparative and internal reconstruction have a long and honorable history in linguistics, application of these techniques to syntax has frequently been criticized as unworkable and as fundamentally different from phonological and morphological reconstruction. For example, Jeffers (1976b: 5) contends that: “A straightforward transfer of the principles of the comparative method to the reconstruction of syntax seems totally inappropriate.” Similar opinions concerning the assumed non-feasibility of syntactic reconstruction abound. While no one would suggest that the techniques of comparative reconstruction can be applied in syntax exactly as in phonology, we argue here that it is nevertheless both possible and appropriate to use the methods of comparative and internal reconstruction to reconstruct syntax. Although there are clear limitations on the effectiveness of these techniques, they are applicable to a wide variety of problems in diachronic syntax. In section 12.2 we discuss how correspondences (or equations) can be established in syntax. Although most of the discussion in the literature has focused on reconstruction, the comparative method relies upon correspondences which establish that change has taken place. We show in section 12.3 how the established techniques of the comparative method can be applied to reconstruct syntax. Section 12.4 deals with obstacles – real and imagined – to the establishment of correspondences or to actual syntactic reconstruction.