No CrossRef data available.
Published online by Cambridge University Press: 28 February 2013
After the decipherment of Linear B and the attribution of phonetic values to most signs, several others still remain undeciphered. Any attempt to reconstruct their phonetic value is beset with difficulties. In this paper we will attempt to set out some basic principles for making reconstructions of undeciphered characters in well-studied scripts. We shall then look specifically at the range of evidence available and the various theories built around the decipherment of Linear B sign *56 trying to construct a strong argument in favour of one possible phonetic attribution. In doing so we shall comment on the strengths and weaknesses of the various arguments, thus illuminating the usefulness of such principles in making a secure attribution.
Principles for secure reconstructions
Most phonetic attributions at this advanced level of script decipherment, where most if not all the basic phonemes or groups of phonemes can be represented through the characters already deciphered, are supported by words that are spelt in two alternative ways, with the undeciphered character and with the deciphered one respectively, enabling the equation of the phonemic value of the two characters. The general principle that governs this decipherment method is that both ‘alternative spellings’ should clearly refer to the same thing. Another principle in this should be that the more complicated the situation, the safer the interpretation as there will be less potential for coincidence.