Observations on the phonetic structure of the Minoan Linear A script

Abstract

Ventris's decipherment of Linear B in 1952 raised hopes among both archaeologists and linguists alike that the earlier Linear A script would itself also soon be deciphered. Fundamental here, of course, was the observation that both scripts shared almost identical sign repertoires and that is was therefore a simple matter of "transferring' the established Linear B phonetic values to similarly shaped Linear A signs. The result has been a plethora of "decipherment theories", ranging from Louvo-Hittite to North-West Semitic, each of which necessarily contradicts the other. Not surprisingly, there has also been marked disagreement among "would-be decipherers" not only as to the total number of phonetic signs actually employed by the Linear A syllabary, but also regarding the identification of many individual Linear A signs actually employed by the Linear A syllabary, but also regarding the identification of many individual Linear A signs with those of Linear B' Significantly, each of these "decipherment theories" deals only with a relatively small portion of the extant Linear A material. None is "holistic" in lcope. This "shortcoming" is itself partly due to the fact the Linear A inscriptions are comparatively scarce, often poorly preserved, and seldomly "neat'" in appearance. The many inherent epigaphical and textual ambiguities, as well as the somewhat terse nature of many of the Linear A inscriptions, have made the Linear A script a tough not to crack. Not only does Linear A remain "undeciphered", it appears that it is, in fact, 'undecipherable". At the same time, it is widely maintained by scholars that the ascription of any phonetic values to Linear A (whether or not these are derived from Linear B) can only be substantiated by a "'cogent" decipherment of the Linea¡ A material. This means, of course, that there must be clear grammatical and syntactical structure according to the rules governing a known language. Since Linear A has not been deciphered" it follows also that the transference of the B values to Linear A has not been demonstrated. Indeed, the likelihood that Linear A is, for all intents and purposes, "undecipherable", further implies that the application of the B values to Linear A is, in effect" "unprovable". The present study is divided into two broad sections. The first introduces the many arguments for and against the application of the Linear B phonetic values to Linear A as determined by stylistic similarities between individual A and B signs. The second section is more ambitious in scope, and involves a detailed statistical analysis of the frequency distribution patterns generated by corresponding Linear A and B signs. The aim, of course, is to test, in lieu of a "cogent" decipherment, the premise that similarly shaped Linear A and B signs also share the same phonetic values. The results speak for themselves.