In his 1900 excavation at Knossos on the Greek island of Crete, Sir Arthur Evans, an English archaeologist, discovered and identified clay tablets inscribed with unknown scripts. From these artifacts, scholars were able to identify three major branches of scripts: Cretan hieroglyphs, a pictographic writing system which was invented around 2000 BCE; Linear A, which was used as early as 1850 BCE, far before the disappearance of the hieroglyphs; and Linear B, developed around 1450 BCE. While there are resemblances between Linear A and Cretan hieroglyphs as well as between Linear A and B, it is nevertheless difficult to trace the development of one to the other. (Packard 20) Of these scripts, only Linear B has been deciphered, a process that was finished in 1953 CE by Michael Ventris, an English philologist and classicist, about a decade after Sir Arthur Evans’ death. Linear A and B share many symbols, and based on the dating of the inscriptions, we know that Linear B developed from Linear A. The scripts also both have a similar, linear appearance, unlike the earlier hieroglyphs’ pictorial nature. Still, about 80 percent of Linear A’s logograms are unique, and the script was primarily written left-to-right or occasionally boustrophedon, while Linear B was written right-to-left. There has yet to be a consensus about Linear A’s decipherment, despite people’s peaked interest in the script after the decipherment of Linear B. Scholars generally accept that while Linear B was used by the Mycenaean civilization, who lived in the last phase of the Bronze Age of Ancient Greece, Cretan hieroglyphs and Linear A were used by another group of people who lived in Crete before the Mycenaeans, and they are called the Minoan civilization, due to the fact that Sir Arther Evans believed that he had excavated the palace of King Minos, who had a labyrinth on Crete in Greek mythology. Linear A’s language, although no studies have identified it, is also called Minoan. It appears to be unrelated to any known language. (Violatti)
In attempts to decipher Linear A, the natural first step was for scholars to try to transliterate Linear A inscriptions with reasonable counterparts from Linear B. In 1989, Yves Duhoux, building off the work of Jean-Pierre Oliver and Luis Godard, compiled a list of 30 homomorphic signs that are found in similar sign groups in both scripts. (Owens, Structure 21) According to Gareth Owens, these 30 values are critically important to showing that the syllabary of Linear A is almost identical to that of Linear B. (22) However, Owens’ argument is weakened by the fact that this is less than half the syllabary in Linear A. While there is not even a general consensus among scholars about the number of signs in Linear A, the number is thought to be between 77 and 85, which indicates that Linear A is likely a syllabic writing system. (Violatti) This number suggests that the language is a syllabary as there are too many signs for an alphabet, but too few for a logographic system. Yet, scholars are currently unable to pinpoint how many of these signs are ideographic, if any at all, as this assumes Linear A is similar to Linear B in that the script has phonetic symbols and ideograms.
Also, Owens argues that the similarities between the signs of the two scripts highlight the important connection between them. According to him, there are more similarities between signs in Linear A and B inscriptions on Crete than between Linear B inscriptions on Crete and on the Greek mainland (371). This is not completely ill-founded, as Linear A has been shown to be connected to Linear B, however, this requires an agreement on the homomorphisms, which does not always happen among scholars. In addition, people must take care to show that there is a connection between the signs that is not only graphical, as signs may actually be entirely different and still look similar to each other.
David Packard’s 1974 statistical analysis of Linear A and B presents a more convincing argument of the similarity between the two scripts. In his work, he assumes that the phonetic values in the signs of Linear B can be carried over to those in Linear A. He analyzed the frequency of each sign with regard to its position within a sign group in both Linear A and B. (Packard 70) He drew simple connections between the Minoan and Greek language in his study, and he also used computational linguistics in order to account for random chance in addition to accounting for personal names and place names, which may show a false connection because place names do not tend to change and personal names may have retained a degree of similarity overtime. From this study’s conclusions, Owens points to evidence that suggests Packard showed that there is a 20 percent difference between the languages expressed by both scripts by looking at consonant and vowel sounds and pure vowel sounds. (Owens, Structure 23) While it may seem that Owens is not taking all the probability errors into account at first, it is essential and interesting to note that this number has grown increasingly smaller when Packard’s analytical study has been repeated over the years with larger databases, as more Linear A inscriptions are collected. It has now been demonstrated that the difference between the phonetic patterns of the languages expressed by the two scripts is now believed to be between 9 and 13 percent. (Owens, Epilogue 297) This statistic is accepted by many scholars.
Considering that Linear A is not currently connected to any known language, this information about the phonetic patterns of the languages indicates that Minoan was very similar to Mycenaean Greek. This similarity may be part of the reason why Linear A was developed into Linear B by Mycenaeans on the island of Crete in order to write Greek, and it helps to explain why the scripts had some degrees of overlap in their timelines. This similarity also shows that it is very possible and likely that Minoan is a part of the Indo-European language family. Of course, this weakens other arguments presented by some scholars, who suggest it expresses a pre-Hellenic language that is unrelated to Greek or that Minoan is not in the Indo-European language family.
In addition, many people, including Packard, have also brought up the fact that Linear A may have been used to write more than one language. This is very probable, especially considering that the script has been found over a large geographical area and over many years in antiquity. Also, Linear A is documented on a variety of different materials, and it is possible that the texts inscribed on clay tablets (which may be literary) are written in a different language than the inscriptions found on over 70 clay balls. Short Linear A inscriptions have been found on vessels and pottery as well as on stone, metal, and stucco walls. Unfortunately, many of these inscriptions are short, fragmentary, or incomplete, greatly hindering the process of decipherment. (Packard 22) Linear B inscriptions, on the other hand, have primarily been found on clay tablets. (Oliver 386) Analyzing the materials along with the scripts allows scholars to hypothesize about what information they may contain, which may help decipher Linear A. According to Packard, administrative and transaction information was written on clay tablets, while other materials, such as stone vases, often have votive inscriptions. (22) It is possible that the clay tablets with Linear A inscriptions may express a different language than the inscriptions on other materials that were more likely used for religious dedications and consecrations. (Oliver 387) Yet, the number of languages inscribed by Linear A is still inconclusive with only our current information. Because perishable materials may have also been used to write Linear A and B, we may be lacking needed information to draw a conclusion.
As more Linear A inscriptions are found each year, the chances of decipherment increase. Many scholars seem hopeful that at least most of Linear A will eventually be deciphered, especially considering that its descendants, Linear B and the classical Cypriot syllabary, can already be read. Further analysis of Linear B in addition to the context in which Linear A inscriptions are found seem likely to help with the decipherment, especially as more material is uncovered and studied in the future.