Project supported by:



Within the literature of the last decades there is abundantly confusion of terms denominating types of script. In fact there is not one single "Libyco-Berber" script. It is not at all surprising that a script used in such a huge territory reaching from Libya to the Canary Islands and during a time of more than 2000 years shows a confusing multitude of varieties.

The basis of our knowledge are the monumental inscriptions of Dougga/Thugga (Tunisia). Thanks to a group of biscripts the phonetic value of nearly all characters of the alphabet is well attested. Therefore, this alphabet is the starting point for all studies comparing sets of signs.


With a rapidly increasing number of documented inscriptions all over North Africa, it became clear that many of them were written in differing alphabets. Upon this fact some scholars came to suppose a "western" alphabet, connected with the culture of the Masaesyles, and an "eastern" alphabet, connected with the culture of the Massyles. Nowadays we know that this separation is only a very rough temporary measure which does not do justice to the existing variety. In face of these conditions it is recommended to use a very cautious term for the whole family of scripts (following Galand 1993): "scripts of the Libyco-Berber type".

If we examine the corpus of a certain region, we have to compare the whole set of signs with the well attested "classical" alphabet of Dougga/Thugga and look for accordance and differences. If the "classical" sign does not appear within the corpus, we have to examine possible substitutes by detailed stylistic and statistical analysis. In the extraordinary lucky case of biscripts (e.g. Latin-Canary and Libyco-Berber) we are able to prove accordance.

The study of Canary inscriptions led to the conclusion that there are even slight differences between the individual islands. Thus it is necessary to present special alphabets for each island:





The rock inscriptions of Morocco do not present a uniform character as well. A detailed analysis would be far beyond the scope of this short introduction and is, therefore, left to special publications. For the present purpose it is sufficient to distinguish between an "ancient" alphabet and a "transitional" one.


For the practical purpose of this database we have to do without special characters in the commentaries. For the semi-consonants we use:

H = semi-consonant in final position
W = "U"
Y = "I"

The most instable group of characters among the alphabets of the Libyco-Berber type is that of the sibilants: the sings often changed within the group. Therefore, as it is not possible at the present state of research to distinguish the phonetic value of the Canary and Moroccan signs for sibilants, they are noted as neutral as possible by S1, S2, S3, S4 and Z1, Z2, Z3.