My earlier piece on Keeladi, which said that the excavations have raised more questions than providing answers, has stirred up a hornet’s nest. There are many hornets buzzing around trying to sting me and those who agree with me. Not one of them, even the archaeologist who carried out the excavations earlier, Mr Amarnath Ramakrishna, has chosen to answer me. Instead, they keep attributing statements to me which I haven’t made and, what is more, these statements are more or less exactly opposite to what I have consistently been saying.
One of the statements attributed to me is that the lack of skeletons makes Keeladi an uninhabited site! In my earlier article, I had said that the present excavations have found remains of animals but not humans. The human burial sites may have been kept away from dwelling units, but the site doesn’t have any evidence of a significant number of people living there. No regular dwelling units or streets have been excavated. What has been excavated could be a site for activities and there is no evidence that they were thriving as far back as the 6th century BC.
The second statement attributed to me is that carbon dating is not reliable. In reality, I have been crying hoarse that carbon-dating (the AMS dating) is the most accurate way to determine the age of an ancient object containing organic material.
However, I have also been saying that even carbon-dating may not be of much use in determining the age of the artefacts that had remained buried along with it. No book on archaeology will tell us that if a piece of charcoal is dated 2,600 years old by carbon dating, the artefacts that were found very next to it will also be dated as old as 2,600 years.
Yes, the charcoal’s age will give us some idea about the likely age of the artefacts, if there are no other contradictory factors. In the fourth phase of excavation, the archaeologists who excavated the site have stated the following: The chronology of this site Keeladi with three periods could be scientifically fixed between 6th century BCE and 11, 12th century CE.
In this context, period-I represented the cultural settings of the site from 6 century BCE to 3rd century BCE. While period-II existed between 3rd century BCE and 4, 5th century CE, the last phase, period-III, constituted the finds belonging to date from 4, 5th century CE to 11, 12th century CE.
Hence, period-I is the earliest phase found at the lower level of the deposit and consequently the other two over it. Thus, the maximum that could be said about the artefacts found at the lowest level is that they could be dated between 3rd Century BC and 6th Century BC.
Now, there is no indication in the report about the level at which the artefacts, especially the potsherds with Tamil Brahmi inscriptions, were found. It is highly probable that they were found at a level higher than the lowest.
If this is the case, even the above assumption cannot be made. The reason for this confusion is that the archaeologists’ report has been kept vague, either accidentally or deliberately.
Amarnath Ramakrishna’s report
Let us compare this with the report (rather paper) on the first two phases of excavation submitted by Mr Amarnath Ramakrishna and others in an archaeological journal in October 2018. The paper is available at this website: Heritage University of Kerala.
In this paper, Mr Amarnath classifies the period of the excavated site as follows:
Period I – Iron Age (pre 300 BCE)
Period II – Early Historic Period (300 BCE – 300 CE)
Period III – Post-Early Historic Period (Post 300 CE)
He also says that the proposed dates are purely tentative and subject to further scientific analysis of other material evidence. The site has, according to the paper, yielded cultural deposits ranging from 2.5 metres to 4.5 metres. Here too, AMS dating was done at the level of 2.5 metres and 1.95 metres and it has yielded calibrated dates ranging between 200 BCE and 195 BC.
Tamil Brahmi inscribed potsherds and how they are dated
Tamil Brahmi inscribed potsherds were also found at various depths. This is what the paper says:
About 70 Brahmi inscribed shreds were collected from two field seasons. All the inscribed sherds came from Locality II coeval with the structural activities of Layer 2. The nature of these writing on the pottery was primarily aimed to register names of individuals probably its owner’s. The inscribed sherds revealed an array of pure Tamil, pure Prakrit and Prakrit converted to Tamil language having names such as ‘athan’, ‘tisan, ‘uthiran’…
The paper also tells us at what depths these sherds were found. For instance, ‘At a depth of about 1.00 metre the structure yielded another remarkable in situ find in the form of another intact black and redware dish inscribed in Tamil Brahmi script… Centan Avathi.
Similarly at the depth of 1.1 metre another inscribed (madaicime) potsherd was found. At yet another pit, such sherds were found at the depth of 1.6 and 2.16 metres. Mr Amarnath doesn’t superimpose the AMS dates on these potsherds. Instead, he says very clearly that “on the basis of stratigraphy and palaeographical grounds, these inscribe shreds may be dated to 2nd Century BCE – 1st Century CE.
Now, we have three sets of dating- carbon dating, stratigraphic dating and palaeographic dating. Let me try to explain them in simple terms.
Carbon dating and strati-graphic dating
Carbon dating places artefacts in time by measuring their radiocarbon content. Stratigraphic dating, on the other hand, relies on the layers of earth surrounding those artefacts. In simple terms, carbon dating is about a particular artefact.
Stratigraphic dating is about the layer in which the artefact is found. Thus, stratigraphic dating will tell you that the artefacts found in a defined layer could be from a period within a well-defined range.
That range will be determined by artefacts reliably dated – either by carbon dating or by other methods. In this paper too, Mr Amarnath sets the upper range of the historic period by dating a coin to the 10th century CE.
What is palaeographic dating?
Palaeographic dating is dating an inscription by studying the type and variety of script used and the style of writing. In the case of Tamil Brahmi inscribed in pottery and stone, experts date them by carefully studying its evolution.
Where does logic lead us?
Any person who believes in logic will immediately realize that carbon-dating is for individual artefacts, and for artefacts for which dates cannot be correctly fixed, stratigraphic dating could give a solution.
Now, we must ask a set of questions and seek the answers.
What has been dated in the latest excavation? A piece of charcoal.
Are the depths at which the inscribed sherds were found given in the available material? No. For all we know, it is probable that they too were found at depths between 1 metre and 2 metres, as was the case in the earlier excavation done by Mr Amarnath.
Even if we assume that the shreds were found in the same layer as that of the charcoal, can we ascribe the charcoal’s date to them? No. If we do that it will be against the basic tenets of archaeology and logic. The shreds, at best, could be dated within a broad range of period, exactly as Amarnath has correctly done, on stratigraphic grounds.
The dates could be between 6th century BC and 3rd Century BC if they were found at the same level as the dated charcoal or to a later period if they were found at higher levels. Such dating will still be purely tentative and subject to further scientific analysis of material evidences, as Mr Amarnath has succinctly said in his paper.
Can we follow one set of logic for the first two phases of excavation and quite a different one for the fourth phase? Unfortunately, Mr Amarnath seems to think so. Otherwise, by following the methods adopted by Mr Amarnath himself in the first two phases there is no way can we say that the potsherds with Tamil Brahmi inscriptions could be dated to 6th century BCE.