ABSTRACT
This chapter discusses a minor 'rational reconstruction' of some of the most important underlying tenets of comparative method and related techniques. Reconstruction, however, is a theoretically loaded and complex operation. It might be useful to ask some fairly basic questions about the status of reconstructed entities, given their role as foundation-stones of the whole edifice; in particular, the results of comparative method, which is perhaps the most important component of our underlying body of knowledge. Linguistic reconstruction is a 'palaeontological' operation, if not quite the same as in palaeontology proper. Truth in reconstruction is both relative and constrained, though there are still satisfactory, if at times rather hermetic and self-referential, ways of telling whether the output of a reconstructive operation is a 'true' picture of a linguistic past. Reconstruction is creating state descriptions and accounts of transitions between states, where either the states or transitions are missing from the record.
