The very ancient history of writing begins in the Paleolithic

We all learned in school that the first documented forms of writing were developed by the Sumerians, in the fourth millennium BC. A new discovery, however, calls into question the origin of the symbolic representation of information through signs on a medium – the origin, therefore, if not of writing proper, at least of a form of proto-writing. Indeed, the results obtained by a group of international researchers seem to indicate that certain sequences of non-figurative symbols have a precise meaning, in conjunction with animal representations made tens of thousands of years before the Sumerians began inscribing their tablets.

In fact, this research group has proposed in a new publication an interesting interpretation for frequently repeated sequences in association with a representations of prey of the ancient inhabitants of European caves, sequences consisting of three precise symbols, namely vertical lines and bifurcated lines. In about 400 European caves, such as those at Lascaux, Chauvet and Altamira, Homo sapiens left non-figurative marks from at least 42,000 years ago and animal drawings from at least 37,000 years ago. The meaning of non-figurative signs, however, has so far escaped the formulation of coherent hypotheses, capable of being tested on a sufficiently large sample of representations.

The group that published the work discussed here compiled a database of animal images and associated graphic symbols found on cave walls or artifacts dating from 20,000 to 10,000 years ago, the time when Most of these motifs were created. Then using statistical tools, the team looked for repeating patterns for the sequences of non-figurative symbols, starting with the observation that some symbols were particularly common. On 606 depictions of animals, in particular, sequences of dots or lines were needed depending on the particular animal represented. Horses, for example, typically had three markings, while mammoths had five. 256 instances of these sequences also had a symbol that was typically in the second position of the sequence.

According to the researchers, when in close association with animal images the line “|” and the dot “•” make up numbers denoting the months and form constituent parts of a local phenological/meteorological calendar starting in spring and recording time from early spring using the lunar month as the unit. The sign “Y” would have the meaning of “to give birth”. The position of the “Y” within a sequence of signs would therefore denote the month of delivery.

If the researchers are correct, then, the purpose of this system of associating animals with calendar information was to record and transmit seasonal behavioral information about specific types of prey in the geographic regions concerned – a mixed proto-writing system, which associated a sort of symbolic “caption” to a figurative image using three types of symbols.

I believe that the results stated by the researchers should be divided into two parts, both of extreme interest, but not of equal solidity. The first result is the demonstration that in the Upper Paleolithic sequences of symbols with well-defined repetitive patterns were used, recurring in association with a well-defined subject – the specific type of animal represented graphically. This result, examining the statistics presented, appears solid and very significant, because it indicates that, with good probability, the symbolic representation of information had already developed in Upper Paleolithic Europe.

The second result is the meaning attributed to the sequences of recurring symbols, linked by the authors to a calendar and to the time of reproduction of the preys. Although this hypothesis is interesting and is quite robust when tested in the data set used, it has the defect of not being compared with any other alternative hypothesis of meaning, which could prove to be even more robust, and furthermore it is based on assumptions unsubstantiated about the breeding time of various animal species in different geographical locations. This part of the results presented needs a more solid basis: that is an interesting hypothesis compatible with the data, which however cannot be considered proven.

In any case, whatever its meaning, the use of recurring patterns of symbols to store and represent complex information appears robustly supported in a much earlier era than hitherto known; it is therefore really probable that at least the precursors of the successive complex forms of writing have a very ancient history, showing a remarkable stability for tens of thousands of years and a wide geographical diffusion. In a very remote era, perhaps, people began to write on the walls of a cave, and thanks to moderate scientific investigation today we can resume those very ancient writings on a newspaper page.