A Comparison of a Manuscript with the Indus Script A Thesis by Lucy Zuberbühler (2009)

[ Pobierz całość w formacie PDF ]
.In order to investigate this possibility, the top 60 most common Indus signs weresearched for in the Kabul data.It was discovered that nine of the ten most commonIndus graphemes and 17 of the top 20 Indus graphemes occur in the manuscript.Whenmore signs are included in the comparison it becomes clear that the total continues toincrease but less steeply.23 of the top 30, 27 of the top 40, 30 of the top 50 and 35 ofthe top 60 Indus symbols are present on the Kabul birch bark (Parpola 2000: 70-78 andWells 1998: 60-70).Figure 17 shows the top 60 Indus graphemes.Signs, which werenot found in the manuscript, have been marked with red print.In comparison only two of the top ten, four of the top 20, ten of the top 30, 11 of thetop 40, 14 of the top 50 and 16 of the top 60 Indus signs appear in the Easter Islandscript (Possehl 1996: 94-95).This means that the Easter Island script has a considerablylower total of frequent Indus signs than the Kabul text and it also shows that the total26increases far less steeply in the collection of Easter Island symbols.These statistics havebeen graphically summarized in Figure 18.These results would seem to imply that the Kabul signs are indeed related to theIndus script while the Easter Island script is not.The Easter Island total is too low whilethe total number of Indus symbols on the Kabul manuscript is still in the acceptablerange.The Kabul total would not be expected to reach a considerably higher amountthan 35 because only 172 graphs are visible on the bark fragment.Not all the 60 mostcommon Indus signs would therefore be expected to appear in such a relatively shorttext.This frequency argument would gain even more strength if the most common Kabulgraphemes also occurred in the 60 most often repeated Indus signs.An exploration ofthis matter reveals that the most frequent Indus sign is also the most common Kabulgrapheme and shows that ten of the top 11 most common Kabul graphemes appear inthe list of the 60 most frequent Indus signs, as can be gleaned from Figure 20.Whenless common graphemes are added to the frequency comparison, the amount increasesless steeply once again.14 of the top 16, 18 of the top 25 and 23 of the top 37 Kabulsigns are amongst the 60 most common Indus signs, as can be seen in Figure 19.Thiswould also support the claim that the sign shapes of the two grapheme lists were relatedto each other.4.ConclusionWhether this last statistic also indicates that the signs encode similar phonetic andlogographic values in the two systems, remains unclear however.It just seems tooproblematic to judge with any certainty, whether the Kabul text could exhibit thispattern of sign frequency overlap, if the symbols represented entirely different linguisticunits in the Indus corpus.One would need to decide, whether sign frequencies couldcoincide in this manner, if a non-Indus people had stumbled upon a collection of Indusinscriptions and merely adopted the shapes of some of the most common symbols fortheir script without adopting their encoding value.If this scenario could be discountedon the basis of signs frequencies, then it could be assumed that a number of Kabulgraphemes share the same value as their matching Indus symbols.This, in turn, would27imply that the Indus symbols belong to a speech-encoding writing system and thehypothesis of the non-linguistic Indus symbol system would be refuted.If this problem were solved, then researchers would need to establish whether theKabul text employed the exact same writing system as the Indus Civilization or whetherthe system was modified in any way.The manuscript might represent a later simplifiedstage of the Indus script for instance.This hypothetical stage could contain a reducednumber of logographs and rely predominantly on syllables instead.Evolutionarypatterns of other scripts show that this development is a widespread phenomenon(Parpola 2000: 38).A large portion or maybe even all of the Kabul signs might therefore encodesyllables.The amount of proposed Kabul graphemes would not rule out either of thesepossibilities because the total of Kabul graphemes might increase drastically, or merelya little, if a larger corpus than 172 graphs were available for examination.Thisrelatively limited corpus simply does not allow predictions on the matter.Therefore, itwould certainly be worthwhile uncovering and analysing the other layers of bark to findout how many more graphemes would join the 62 already visible graphemes.It would also be extremely interesting and potentially revealing to compare thefrequencies of particular signs sequences.If it could be demonstrated that the mostcommon Indus grapheme sequences also occur relatively frequently in the manuscript,then it would seem likely that the Kabul penman not only used the Indus script but thatthe text encoded the language, or one of the languages, of the Indus Civilization.As itstands the manuscript might encode a non-Indus language and merely utilise the Induswriting system.Many issues therefore still remain unresolved and numerous aspects beg furtherinvestigation.So far, a graphemic analysis of the Kabul manuscript has only enabled thepositing of a Kabul grapheme list and its subsequent comparison to the symbols of theIndus corpus has merely lead to the definite conclusion that the majority of the signs onthe analysed birch bark were borrowed from the collection of Indus symbols [ Pobierz całość w formacie PDF ]

Linki