In 1998 ANISTORITON (Archaeology News, A987. 07 N 1998) first announced to the world community that a Greek amateur linguist, Mr. M. Tsikritsis was about to complete a new study on the decipherment of Linear A. In the Aegean Archaeology list AEGEANET [Greek???, e-mail: 981113.03, threads and following messages some were surprised and interested in the outcome of Tsikritsis's study, others rather bored that still another "nationalist" lunatic thinks that he can read the Minoan symbols. The Editorial Board of Anistoriton even received few e-mail messages from Archaeologists who were embarrassed by our announcement and they criticized the Journal staff for their decision to publish it. However, we must not forget that M. Ventris (deciphered Linear B in the early 1950s) was also an amateur and that H. Schliemman (a German merchant with a passion for Homer) was widely ridiculed in the late 19th c. by German University professors when he announced that he had located and excavated Troy.

In the last a hundred years the study of Linear A has shown that it is the written form of a language that cannot be read by linguists. A. Evans (excavator of Minoan Crete) and others have brought to light some tablets and objects with Linear A words. However, it has been impossible for Archaeologists to decide on whether the language is hittite or luwian or semitic or greek as the number of words (composed of syllabic symbols) are few. Experts in Mediterranean languages, therefore, have labeled Linear A as a non-deciphered language. Well, we may not be able to read the symbols of Linear A but comparing them to symbols of Linear B (some are almost the same, others look alike) and using statistical methods we may come to a conclusion on whether the script and/or the language is related to Linear B or not. In other words, if we use the sounds of symbols in Linear B (we know from Ventris decipherment) that are the same or similar in Linear A and we read Greek words (we know from classical Greek) or we read place names (we know on Crete), then we may come to the conclusion that most probably the language of Linear A script is related to the language of Linear B (Greek). There may be some words in Linear A that are not found in Linear B in the same way that some words in Linear B are not found in classical Greek. The explanation is, that not all words of a language survive from generation to generation.

M. Tsikritsis, a computer scientist and a text analysis specialist has had the expertise needed to launch the task of a statistical and machine comparison study of the Linear A and Linear B symbols. A few months ago, his book entitled Linear A. A Contribution to the Understanding of an Aegaean Script (259 p.) was published by the Vikelaia Library of Herakleion, Crete, Greece. Unfortunately, the book was published in Modern Greek by an important but small publishing house operated by Vikelaia Library. Moreover, the Greek Archaeological community has no expert on the field of Aegean scripts and cannot evaluate Tsikritsis's work. Most likely, Tsikritsis study will be forgotten in the years to come as, written in Modern Greek, it will have a very limited audience.

Nevertheless, Tsikritsis has used an original method in order to test his working hypothesis that same or similar symbols in Linear A and B represent the same syllabic sounds. All Linear A words discovered until 1995 were recorded in a computer file. Special code (similar to the one used in Thesaurus Linguae Graecae) was employed and the 679 words were divided into syllables and the position of each syllable in the word was given a specific number. Also, the same file included a code for the location where the word (the tablet or object) was found as well as the number of times this word was repeated in Linear A texts. A second file was constructed with Linear B words. This second file contained a random number of Linear B words but the frequency of the number of syllables in both files was the same, i.e. 6.48% of 5-syllable Linear A words was used in the first file and 7% of 5-syllable Linear B words was used in the second file. The Linear B words were also recorded in special code and were given special codes and numbers referring to the location of the tablets they were written on, the position of syllables in the words and the number of syllables. Finally, the syllables contained in the words of the two files were compared with the use of a computer. Thus, phonetic values of Linear B syllables were assigned to Linear A syllables according to the similarity of symbols and MAINLY according to the frequency of the position of the syllables in the words.

Although one might think that the above technique solved the problem, Linear A could still be an unknown language written in an unknown script because phonetic values of Linear B could mean nothing in Linear A. The only way to solve this problem was to test whether there were words that could be used as keys to the decipherment. According to Tsikritsis the words su-ki-ri-ta and a-ka-ru were found to represent the Cretan place names of Sygrita and Arkalo. Actually 12 words representing place names in both Linear A and B were identified. This led to the conclusion that most probably the phonetic values assigned to Linear A were correct. Also, other Linear A words that were identified were found to be closely related to the object they were written on (e.g. a vase) as well as to the usage of the object. This discovery made Tsikritsis more confident that his decipherment is correct. It must be mentioned, though, that not all words in Linear A can be understood with the use of the above method as it is also the case in some Linear B words and in most decipherment methods used for ancient and/or unknown languages. However, as it has already been mentioned above, in the life span of a language some words may become obsolete from generation to generation and disappear from the vocabulary in use.

In conclusion, according to Tsikritsis, Linear A script represents an archaic form of Greek. Specifically, it must be an early aeolic dialect the speakers of which used linguistic archaisms (words and sounds that later became obsolete) that are not found in Linear B and therefore can not be read.

Back to Cover