Background

Genetic studies suggest that the earliest modern South Asians migrated from Africa perhaps as early as 70 000 years BP (Macaulay et al. 2005). This migration is often assumed to have occurred along the coastline of southern Asia. Direct fossil and archaeological evidence for this scenario is, however, currently lacking in South Asia and recent studies have not targeted sites of the appropriate age or location. The state of Gujarat in north-western India has the potential to fill this gap with a coastline of c. 1600km and Palaeolithic sites that have been dated to the Late Pleistocene (Reference MaratheMarathe 1981; Baskaran et al. 1986). By re-examining evidence from previous studies and accumulating new data from the peri-coastal areas of Gujarat State, the Gujarat Palaeoanthropology Project (GPP) aims to provide new insight into Late Pleistocene early human coastal adaptations and shed light on the origins and dispersal of modern Homo sapiens in South Asia.

Objectives

The project's objectives can be summarised as follows:

1. 1. Re-assess peri-coastal Palaeolithic sites located within the Saurashtra peninsula (Gujarat State) that have been dated in previous studies to Oxygen Isotope Stage 4 (OIS 4, c. 70 000–50 000 years BP).

2. 2. Survey Saurashtra for new palaeoanthropological localities.

3. 3. Refine the Palaeolithic chronology of the region via modern chronometric techniques such as Optically Stimulated Luminescence (OSL) dating.

4. 4. Assess archaeological materials for evidence potentially indicative of modern human behaviour.

5. 5. Reconstruct the Late Pleistocene palaeoenvironments of Gujarat through studies of fossil faunal remains and stable carbon isotopes of fossil soils.

Figure 1.

Digital elevation model of Gujarat (interval 25m). The white contour indicates the approximate OIS 4 coastline. Hatching indicates the survey area.

Preliminary results

Initial investigations focused on the Bhadar River valley in south-central part of the Saurashtra peninsula (Figure 1). In general the density of lithic finds within the study area was found to be quite low. A total of 336 lithic artefacts were collected from more than 20 sites, primarily from buried in situ alluvial deposits. These artefacts are typologically indistinct with a few exceptions (e.g. disc and blade cores, handaxes). Prevalent upland lithic raw materials sources were surveyed but no traces of prehistoric activities were observed at these sites. It appears that most lithic raw materials were obtained from channels.

Sections on both banks of the Bhadar River were scrutinised in the vicinity of Jetpur. A composite stratigraphy of the Bhadar was developed (Figure 2) which is essentially consistent with earlier observations (Reference SankaliaSankalia 1965). The lower coarse fluvial unit observed at Jetpur and throughout the Bhadar River valley is Acheulean. However, lithics from the second fluvial unit cannot be positively classified as Middle or Late Palaeolithic. Two disc cores found within the Bhadar channel suggest the presence of 'Middle Palaeolithic-like' technology in the region; the majority of the collections were, however, simple whole flakes and cores (Figure 3). Previous workers classified tools from the second fluvial unit as Middle Palaeolithic (Reference MaratheMarathe 1981). A more conservative view adopted here is to refer to them as "later" Palaeolithic.

Figure 2.

Composite stratigraphic section at Jetpur.

Figure 3.

Lithic artefacts from the Saurashtra peninsula, Gujarat, India. Scales are 1cm.

Fieldwork near the present-day villages of Badalpara (Badalpur) and Kajli east of Veraval was carried out following reports of Palaeolithic tools associated with marine deposits (Reference MaratheMarathe, 1981). Only a few dubious stone tools were found during this fieldwork. Most sections here were found to be poorly exposed and limited in thickness. A large open carbonate quarry between Badalpara and Kajli was targeted due to the presence of thick, laterally extensive and unobstructed outcrops. One heavily rolled Acheulean handaxe was recovered from this quarry. A dramatic revision of the reported finds from this site is necessary to accommodate these new observations.

Figure 4.

Partial skull of a fossil equid recovered from beach rock in coastal Gujarat. Scale is 5cm.

Geological samples were obtained for geochronological and geochemical analyses. Six OSL samples were collected from four sites and another dozen samples were collected from alluvial palaeosols for palaeoenvironmental analysis of soil organic matter δ¹³C and pedogenic carbonate δ¹³C. A site yielding numerous Pleistocene vertebrate fossils — the first such assemblage discovered in the state of Gujarat — was discovered in beach rock on the Saurashtra coast. Analyses of all these samples are ongoing but a preliminary study of the fossil fauna reveals a diverse assortment of taxa, including equids (Figure 4) and on stratigraphic grounds, it probably belongs to the time interval of interest (i.e. OIS 4).

Prospects

Our work to date has not uncovered clear evidence for early modern human migrations in Gujarat. However the research has made positive progress in providing a clearer picture of the Palaeolithic and palaeoenvironmental record of the region. The chronological and environmental data produced by this work will provide some possible explanations for why evidence for the southern dispersal of Early Modern Humans is elusive in this study area.