After recognizing fundamental errors in my first approach to find a key to read the Rongorongo tablets by means of quantitative analytics, this is the second approach which uses the mathematical and statistical properties of the language...
moreAfter recognizing fundamental errors in my first approach to find a key to read the Rongorongo tablets by means of quantitative analytics, this is the second approach which uses the mathematical and statistical properties of the language and the writing system for this purpose. If the first, naive attempt to assign every sign of Rongorongo to syllables of Rapanui yielded only few results, the approach presented here is much more promising. Instead of assuming a purely syllabic script, the ratio of syllable signs to logograms was estimated using the Modified Power Law (MPL). Contrary to the claim that Rongorongo is not a real writing system and therefore undecipherable, subsequent examination of the quantitative properties of the texts compared to the quantitative properties of the isolating Polynesian language Rapanui suggests that Rongorongo is a complete syllabic-logographic writing system.
While not much has been changed to the algorithm for the quantitative assignment of statistical properties of signs to sets of syllables, the separate examination of each tablet has produced new insights. So we find different sets of signs used in the texts of each tablet, but a core set of 80 signs used with similar frequency in each examined tablet. This suggests that different authors may have been involved in the writing of the texts, maybe at different times, but the complete inventory of Rongorongo signs was not used by every author. However, everyone was familiar with the basic subset of probably predominantly syllable signs, that is, they had a common essential writing system.
After correcting the errors of the first approach while maintaining the basic principles, there was a good agreement between the mathematical properties of this unique Polynesian writing system and the Rapanui language on which it is based. The statistics of language and writing could be correlated quantitatively, but ambiguities in the assignment of syllables to signs could not completely be resolved. The attempt at deciphering has been made from various points of attack, which are reflected in subsets with defined properties of writing and language. The quantitative method included counting, selecting, sorting, assigning, etc. This resulted in different readings with different probabilities for the signs in Easter Island script. Only the two most probable readings of a sign were used to transcribe the texts of the tablets. The number of resulting different readings is still very large. As an example, all variants of the most probable readings of the beginnings of the texts were generated recursively.
These readings, which are still incomprehensible, were subjected to an experiment: online translation programs were used to test which languages actually deliver readable, if not necessarily meaningful translations. They were mostly languages of the Polynesian language family. This is in line with the expectations. However, this could also be an artifact, since ultimately the statistical properties of Rapanui were mapped onto the statistics of the Rongorongo signs. In the worst case, the results would only be a kind of Markov chain.
On the other hand, if it is a kind of transcript of the tablet texts, which is pointing in the right direction, then from now on the help of native speakers of Rapanui and insiders of the cultural tradition of Rapa Nui as well as linguists is essential/indispensable! Collaboration with experts on the history and mythology of Rapa Nui and native speakers of the Easter Island language Rapanui is imperative and necessary before any further significant progress can be made in correctly reading the Kohau Rongorongo.
That it is at all possible to map statistical and structural properties of Rapanui to the Rongorongo texts is an encouraging indication. However, the following critical questions must be answered first:
• Is the Rapanui database used here useful?
• Are there passages in the transcripts that make any sense?
• Does the text or the rhythm of the prose remind one of old Polynesian or even Rapa Nui stories, songs, poems or prayers?
If there are no further approaches for the application of the formal, quantitative method used here, then an extended examination of the signs and a comparison with their descriptions and interpretations is necessary.
In particular, the core signs should be given a special attention. In a further investigation, the question should therefore be answered as to whether the core set of signs is also used in other tablets.
Another hypothesis is, whether parts of the signs already carry meaning. However, this requires a new way of describing the signs. So far, the systematic list of Barthel's numbers has been used. However, Metoro's readings of the tablets come into consideration (Barthel [3]), as well as his statements on the "names" or meanings of the signs. The work by Werner Wolff [11] provides a well-founded supplement, even if Wolff initially assumes a purely logographic script, but later also suspects syllabic signs as well. In a first step, the different decipherments should be compared with Metoro's descriptions of the signs. The next step might be a detailled examination of the signs itself. The linguistic work of Albert Davletshin [12] could be used in a further step.
This research is currently in progress.
Note: The main text of this document is written in German.