Birnbaum Henrik. Linguistic Reconstruction: Its Potential and Limitations in New Perspective

Journal of Indo-European Studies, 1978. — 84 p. — (Journal of Indo-European studies: Monograph 2).The paper sets out to redefine and explore the methods of linguistic reconstruction, its functions and constraints. Specifically, the issue of a more precise definition of internal reconstruction, implying a procedure of “projecting backward” a set of synchronically ascertained data and processes of one language, is addressed, and a more stringent delimitation of internal reconstruction vis-a-vis the use of the comparative method for the purpose of reconstruction is proposed. In addi- tion to the two chief traditional approaches for recovering lost stages of linguistic evolution, some further methods, aimed at either combining these two techniques (the ‘integrated comparative method’) or supplementing them (‘external reconstruction’ based on extra-systemic language data provided by loanword and onomastic material as well as extra-linguistic evidence), are discussed and their merits and shortcomings assessed. The drawbacks of lexicostatistics/glottochronology, which has been proposed as a further tool in linguistic reconstruction, are noted. The structural affinity of reconstruction and prediction, and the once claimed parallelism between linguistic diachrony and the order of synchronic rules in trans- formational-generative language description are briefly com- mented upon. The potentials and limitations of a modeling approach to linguistic reconstruction, superimposing, as it were, related-language structures for the purpose of arriving at new insights concerning their relationship (esp. chronology), are discussed. The specifics of reconstruction at various levels of linguistic structure — phonological, morphological, syntactic, and semantic — as well as the very nature of linguistic change are reexamined. The data and references used to illustrate the above theoretical considerations are primarily drawn from Indo-European, notably Slavic, but some suggestions are also made regarding the possibility of ascertaining, by ‘external comparison’ (genetic and/or typological), distant linguistic relationships (or ‘macrofamilies’) and their underlying parent languages (‘preprotolanguages’), or an abstract model thereof, such as those of the redefined ‘Nostratic’ languages of the Old World.