An Encyclopaedia of Indian Archaeology v1.c10.s3: 5.3 Carbon-14, ¹⁴C, Radiocarbon
An Encyclopaedia of Indian Archaeology v1.c10.s3
- Language
- English
5.3 Carbon-14, ¹⁴C, Radiocarbon
v1.c10.s3.1
Radiocarbon dating method assumes that the ¹⁴C/¹²C ratio has remained constant as the ¹⁴C production rate has not changed in the last million years. But tree-rings of sequoia and bristlecone pine, which live for thousands of years, when ¹⁴C-dated, show a discrepancy between the ¹⁴C dates and the dendrochronological dates: the ¹⁴C dates are progressively younger than the dendrochronological dates as one moves back in time. This fact goes against the assumption of the constancy of ¹⁴C/¹²C ratio. Or are there other explanations? The problem has been hotly debated several times, yet there is unanimity neither on the causes nor on the magnitude of variations so far. Below are discussed the various suggested corrections and mechanisms and their effect on Indian chronology.
v1.c10.s3.2
Suggestions for corrections to ¹⁴C dates are: (1) C = 1.26R—700 (Stuiver’s formula, 1970); (2) C = 112 + 0·690 R + 0·152 × 10 ³ R²—0·38 × 10⁻⁷ R³ (Wendland and Donley’s formula, 1971); and (3) MASCA factor (Ralph *et al*’s formula, 1973) (where C = corrected age and R = uncorrected radiocarbon age). Even these formulators have changed their own formulae several times. Ralph *et al*.’s formula is quite well-known as it has been very widely circulated in the *MASCA Newsletter*, 11, 1, of the Univ. Pennsylvania. They have given detailed tables where ¹⁴C dates from A.D. 1800 to 4140 B.C. can be converted into MASCA-corrected dates. Already several archaeologists have started using these corrections, perhaps a little too indiscreetly. Before the details of how these corrections affect Indian chronology are gone into, the basic problems involved may be reviewed. If there is a discrepancy between the ¹⁴C dates and dendrochronological dates there can be several alternative reasons for it: (1) variation in cosmic ray flux may have affected the ¹⁴C production; (2) changes in the magnetic field may have affected the ¹⁴C production; (3) the sequoi and bristlecone pine trees grow in California at heights of more than 3000 m and therefore there may have been *in situ* production of ¹⁴C; (4) there may have been radial diffusion of younger plant material into the older rings; and (5) geographical factors may be responsible for such observed variations. These technical controversies apart, a few facts may be pointed out. In New Zealand scientists have worked out a sequence of ¹⁴C dates for Kauri pine which contrasts strongly when compared to the Californian tree curve. ‘But the New Zealand dates recalibrated against the Kauri pine curves seem to make more archaeological sense than calibration with the bristlecone pine. Thus a local factor seems to be at play.’ Diffusion inwards of resin in the tree-rings has been reported by Rafter. Olsson pointed out 100% atomic bomb effect in pre-bomb (1954-50) tree-rings. Besides these uncertainties of radial diffusion of plant material there are significant differences in measurements between Arizona and La Jolla laboratories which range from 6% to 70%. It is to be noted that these measurements were made on the same tree-rings. We thus see that the calibration of the radiocarbon calendar is still beset with many uncertainties and problems, though the workers in the USA seem to be accepting tree-ring data for corrections.
v1.c10.s3.3
In the adjoined table ¹⁴C dates, archaeological ages and MASCA-corrected ¹⁴C dates are compared. The table shows that up to a few centuries B.C. there are no significant differences in any of the dates and they compare well with each other. For the older periods, while the uncorrected ¹⁴C dates and archaeological dates compare well, MASCA corrections make them too old. Thus, for Mohenjo-daro we have now seven ¹⁴C dates from the top level giving a compounded date of *c*. 2000 B.C. If we take even a minimum of 300 years of duration of the occupation of the site, we got the following ¹⁴C time-spread: *c*. 2300-2000 B.C. (uncorrected ¹⁴C dates and extrapolated beginning); and *c*. 2870-2180 B.C. (MASCA-corrected). Even this early beginning of *c*. 2870 B.C. is difficult to accept. But if we take 500 years as the duration of the occupation at the site, the beginning would go back beyond the 3rd millennium B.C., which is archaeologically unacceptable.
v1.c10.s3.4
The only solution out of this confusion seems to be to ¹⁴C-date long tree-trunk sequences from bog-buried oaks. Already German and Irish workers have established some floating dendrochronologies. A long sequence of ¹⁴C dates on such dendro-dated tree-rings will alone tell us the real magnitude of ¹⁴C/¹²C variations, as these tree-trunks would not suffer from the disadvantage of bristlecone and sequoia where a single tree lived for *c*. 5000 years. The second solution is a comparison of well-dated archaeological and ¹⁴C-dated samples of short-lived material from Egyptian and Mesopotamian sites. Till these uncertainties are resolved one can continue to use uncorrected ¹⁴C dates based on 5730-years half-life value. Uncorrected dates will not create any confusion, but corrected ¹⁴C dates based on various correction formulae can create a lot of confusion if archaeological inferences are based on them. It is better to wait till a consensus emerges and the uncertainties are removed.
v1.c10.s3.5
HISTORICAL AND ¹⁴C DATES BETWEEN *c.* A D. 500 TO 200 B C
v1.c10.s3.6
| *Sample* | *¹⁴C dates in B C / A D based on half-life 5730 years* | *Range of historical dates in years B C./ A.D* | *MASCA-corrected dates in years B.C./A D* |
| TF-38 | A.D 145 ± 110 | A.D 40-120 | A.D 205 |
| TF-93 | A.D. 245 ± 110 | A.D. 60-100 | A.D 320-290 |
| TF-98 | A.D. 435 ± 95 | A D. 505-15 | A.D. 540 |
| TF-364 | 135 ± 105 B.C. | 0-100 B C | 10-100 B.C |
| TF-681 | 270 ± 105 B.C | A.D. 100-100 B C | 230-380 B C |
| TF-185 | 295 ± 95 B C. | A.D 25-25 B.C. | 270-390 B.C. |
| TF-171 | 185 ± 100 B.C. | A.D. 25-25 B.C. | 150 B C |
| TF-248 (a) | 205 ± 100 B.C. | 100-200 B C. | 180 B.C |
| (b) | 235 ± 110 B.C | 100-200 B.C. | 210-350 B.C |
v1.c10.s3.7
Radiocarbon dating has now come to age in India. It started at the Tata Institute of Fundamental Research, Bombay, and now continues at the Physical Research Laboratory, Ahmedabad. In the course of the last 15 years the laboratory has measured about a 1000 ¹⁴C dates. This dating method and the problems regarding calibration of ¹⁴C dates have been discussed above. Below we shall only deal with the ¹⁴C chronology of the Middle Palaeolithic [4.3], Mesolithic [4.5], Neolithic [4.6], pre-Harappa [4.8], Harappa [4.9], Chalcolithic [4.14] and Iron Age [4.16] cultures. All ¹⁴C dates are based on 5730-year half-life value and are uncalibrated for any ¹⁴C/¹²C variations.
v1.c10.s3.8
Rajaguru has worked hard to collect ¹⁴C samples from Middle Palaeolithic sites; most of the dates are based on his samples. The Bardia hill in the Bhadar valley, Dhom Dam in the Krishna valley, INAMGAON on the Ghod, Mula Dam on the Mula and Ratikarar on the NARMADA have been ¹⁴C-dated. All these sites have Middle-Palaeolithic-associated sediments which yielded shells for dating. Except for Mula samples which are of wood, all other dates are based on shell material. Shell is not very reliable for dating and we therefore think that the dates obtained indicate only the upper limits. Except for one Mula sample (TF-217) which was older than 40000 years, all other Middle Palaeolithic samples have been found to be younger than 37000 B.P. We should bear in mind that even a little younger contamination can make an indefinite age look like 30000 B.P. The upper limit is about 12000 B.P. for the Middle Palaeolithic.
v1.c10.s3.9
Only one shell sample has been dated from the Upper Palaeolithic [4.4] levels of the BELAN deposits in Dt Mirzapur and the date is 17765 ± 340 B.C (TF-1245).
v1.c10.s3.10
Most of the ¹⁴C-dated Mesolithic sites are located in Uttar Pradesh, Madhya Pradesh and Rajasthan. Below are given the approximate time-spreads, all B.C., of the major sites: ADAMGARH, *c.* 5500-1000; BAGOR, *c.* 5000-3000; BHIMBETKA *c.* 5500-1000; KOLDIHWA, *c.* 5500-4000; LANGHNAJ, *c.* 2000; LEKHAHIA *c.* 2500-1500; and SARAI NAHAR RAI, *c.* 1000.
v1.c10.s3.11
The main ¹⁴C-dated Neolithic sites are BARUDIH, BURZAHOM, CHIRAND, HALLUR, SANGANAKALLU, T. NARSIPUR, TEKKALAKOTTA, TERDAL, KODEKAL, PALAVOY, UTNUR and PAIYAMPALLI. On the basis of the dates the Neolithic of s. India as a whole can be bracketed between *c.* 2500 and 1000, that of Bihar between *c.* 1800 and 1000 and that of Kashmir between *c.* 2400 and 1500.
v1.c10.s3.12
Except for KALIBANGAN Pd I, there are no proper pre-Harappa sites in India that have been ¹⁴C-dated. There are other sites like BARA, MITATHAL, etc., which have some pre-Harappa traits but they obviously continued much later. KALIBANGAN Pd I is to be bracketed between *c.* 2400 and 2000 on the basis of ¹⁴C dates. In the upper limit there is an obvious overlap with the Harappa. For the Harappa culture several sites in Rajasthan (KALIBANGAN Pd II), Gujarat (LOTHAL, ROJDI, SURKOTADA) and Haryana (BANAWALI, MITATHAL) have also been dated. The time-spread for the Indian Harappa culture is *c.* 2200-1700. In the metropolitan areas of Mohenjo-daro and Harappa it may go back to *c.* 2400-2000.
v1.c10.s3.13
Some Chalcolithic culture sites show the following time-spreads: PRABHAS, *c.* 2400-1500; BANAS [4.14.1], *c.* 2000-1400; KAYATHA, *c.* 2000-1800; MALWA [4.14.5], *c.* 1700-1400; Jorwe [4.14.16], *c.* 1400-900; and JODHPURA, *c.* 2500-2000.
v1.c10.s3.14
The n. Iron Age, associated with the black-and-red ware, PGW [11.14] and NBPW [11.16], has the following dates: PGW, 800-300; and NBPW, 600-50. The s.
v1.c10.s3.15
Iron Age, typified by the Black-and-Red Ware [11.15] and megaliths [4.17], has the following dates: Neolithic-Megalithic overlap, c. 1000; and Megalithic, 600 B.C.-0.
v1.c10.s3.16
Agrawal, D.P., & Kusumgar, S., 1974a, in *Purātattva*, 7, pp. 70-3 (for calibration of ¹⁴C dates); 1974b, *Prehistoric Chronology & Radiocarbon Dating in India*, New Delhi. Agrawal, D.P., Krishnamurty, R.V., Kusumgar, S., & Pant, R.K., 1977, in *National History of Human Populations in India* (for prehistoric chronology). Berger, R., 1970, in *Philosophical Trans. of the Royal Soc.*, London, 269, pp. 23-6 (for Egyptian ¹⁴C chronology). Kusumgar, S., & Agrawal, D.P., 1975, in *International Conf. on Low Radioactivity Measurements & Applications*, Sept. 1975, High Tatras (for magnitude of ¹⁴C/¹²C variations). Olsson, I., ed, 1970, *Radiocarbon Variations & Absolute Chronology*, Uppsala. Rafter, T.A., & Grant-Taylor, T., ed., 1972, *Proc. VIII International Radiocarbon Dating Conference*, Lower Hunt. Ralph, E.K., Michael, H.N., & Han, M.C., 1973, in *MASCA Newsletter*, 9, 1. Wendland, W.M., & Donley, D.L., 1971, in *Earth & Planetary Science Letter*, 11, pp. 135-9 (for ¹⁴C calendar age relationship).—D.P. AGRAWAL