Indo-European and the Indo-Europeans: A Reconstruction and Historical Analysis of a Proto-Language and a Proto-Culture.

• Author: Melchert, H. Craig

This long-awaited English version of an already famous work appears in a two-volume format, differing from that of the Russian original. The first volume contains the languages and their written sources, a methodological introduction, phonology and morphophonology, morphology and syntax, dialectology, lexicon, homeland and migration patterns, and a brief epilogue. The only addition to the original that I noted is the very last footnote on page 864. The second volume contains bibliography and indices and, oddly, repeats the information on languages and sources.

Translation of this massive and complex text was an enormous task, and we owe a great debt to Johanna Nichols for taking on the burden of making this important work accessible to those who do not read Russian and carrying out the assignment with distinction. I cannot say the same for her tendentious and hyperbolic preface. Gamkrelidze's and Ivanov's Indo-European grammar is without doubt the most original comprehensive treatment since that of Hermann Hirt, and it was not necessary to misrepresent the history of Indo-European studies to make the present work appear even more revolutionary than it is. Neither the use of typology nor the "deductive canon" is a novelty in the field (both began with Franz Bopp!). What has changed is our conception of language universals and language typology. One of the great merits of Gamkrelidze and Ivanov is that they have brought to their analysis not only the results of recent Western scholarship in these areas, but also the work of scholars in the former Soviet Union either unknown or under-appreciated in the West.

Predictably, the authors' own admirably succinct and lucid presentation of their methodology contains nothing at odds with standard practice in the field. The issue is not their method, but their application of it: does their model of PIE meet their own stated conditions (pp. xciv-xcv) of squaring both with the specific historical facts and established typological data (synchronic and diachronic)? I regret to say that the answer must be a resounding "No!"

The authors' approach suffers from two major defects. Their use of the "deductive canon" leads to rampant aprioricity, resulting in: (1) circular reasoning (evidence pointing to regressive voicing assimilation in stops in PIE must be dismissed [p. 133, n. 4], since it is incompatible with the "glottalic" reinterpretation espoused earlier); (2) glaring inconsistencies (on p. 142, note 13...