19th Ave New York, NY 95822, USA

Cultural Dynamism


Cultural Dynamism Prehistoric Hunter-Gatherers in Mid/Late Holocene Sri Lanka


Anneli Ekblom, Christian Isendahl and Karl-Johan Lindholm (eds.) 2018. The Resilience of Heritage: Cultivating a Future of the Past. Essays in Honour of Professor Paul J.J. Sinclair. Uppsala: Uppsala University. ISBN 91-506-2675-9 Throughout his career, Paul Sinclair has encouraged students to pursue a concerned archaeology that goes beyond establishing cultural chronologies to formulating critical inquiries fundamental to our world and for our future. This book honours his achievements by exploring urbanism, resilience and livelihoods, contacts and trade, and heritage and landscape. In the tradition of Paul Sinclair’s eclectic multi-, inter- and transdisciplinary approach to archaeology and historical ecology, this book expands the scope of archaeology by combining the examination of the material record with climatology, paleoecology, ethnography, sociology and archival sources to address both past and present interactions between people and environment. In doing so, the contributions to this volume highlight the value of knowledge about the past in contemporary society.


INTRODUCTION ………………………………………………………………………………. 9
Anneli Ekblom, Christian Isendahl, Karl-Johan Lindholm and Carole Crumley
THE CATALYST ……………………………………………………………………………….. 31
Graham Connah
FROM BIRKA TO BANYON ………………………………………………………….. 35
Neil Price
THE SUPERVISOR ………………………………………………………………………….. 43
Anna Källén
THE BOLIVIAN EXPERIENCE ……………………………………………………… 49
Vicky Saenz
THE CHILD FROM NYANGA ……………………………………………………….. 55
Svante Fischer
THE CROCODILE ON SAMOS ……………………………………………………… 61
Gunnel Ekroth
ANCIENT ENVIRONMENT TODAY? …………………………………………. 69
Gullög Nordqvist
INVESTIGATING COMPLEXITY ………………………………………………….. 83
Great Zimbabwe from a Multidisciplinary Perspective Innocent Pikirayi
MAKING INVENTORY ………………………………………………………………… 101
Mapping and Exploring Zimbabwe Sites Anders Lindahl and Edward
A HIERARCHICAL SYSTEM ………………………………………………………… 119
Central Places in the Zambezi Valley, Mozambique, c. 1400–1900 CE
Solange Macamo, Yolanda T. Duarte and Ricardo T. Duarte
Origin and Development of Urbanism in Madagascar: The Case
of Antananarivo Chantal Radimilahy and Jean-Aimé Rakotoarisoa
Revisiting ‘Low-Density, Agrarian-Based Urbanism’
Elisabeth Graham and Christian Isendahl

The Shashi-Limpopo Lowveld Munyaradzi Manyanga
Lower Limpopo Valley Michel Notelid and Anneli Ekblom
CULTURAL DYNAMISM ………………………………………………………………. 225
Prehistoric Hunter-Gatherers in Mid/Late Holocene Sri Lanka
Raj Somadeva, Anusha Wanninayaka, Dinesh Devage and
Chandima Abeysiriwardhana
ENVIRONING ……………………………………………………………………………….. 253
The Archaeology of ‘Real Life’ Remains Karl-Johan Lindholm
Facing up to the Challenge of Chronological Control
Martin Finné and Erika Weiberg
Reconstructing Past Networks between the Coast and Interior
Felix Chami and Christowaja Ntandu
HOW SMALL THE WORLD ………………………………………………………….. 307
Intercontinental Trade in the 8th to 9th Century CE Illuminated by
a Handful of Glass Beads Marilee Wood
CUISINE AND CULTURES ………………………………………………………….. 323
Middle Asian and African Culture Defined Randi Haaland
THE FINAL FRONTIER ……………………………………………………………….. 343
Colonisation and Colonialism in the 19th Century Maloti
Mountains, Southern Africa Peter Mitchell
WORDS APART ………………………………………………………………………………. 363
Archaeology, Indigenous Communities and the Power of Definition
Charlotta Hillerdal
LIVING HERITAGE ………………………………………………………………………. 393
Cultural Heritage Management and Archaeological practice in
Mozambique Solange Macamo and Anneli Ekblom
Mapping Sitescapes with UAS and Photogrammetry
Daniel Löwenborg

The resilience of Mesolithic hunter-gatherers in Sri Lanka is archaeologically under-explored. For long, the lack of evidence has delayed any attempt to address issues of change and continuity of these societies. However, recent archaeological fieldwork has revealed data to support the argument (Perera et al. 2011; Somadeva 2014) that hunter-gatherers exploited large-scale floral resources during the mid-Holocene (cal. 4300–3355 BCE). Such a change indicates an inclination towards a new economic behavior in response to a mid- to late Holocene environment marked by wet and dry phases. This paper builds on results from the research project Hunters in Transition (HTP) initiated by the Postgraduate Institute of Archaeology and the National Science Foundation in Sri Lanka, and aims to explore possible correlations between environmental and cultural changes during the mid-Holocene in the mountainous land[1]scape of mountainous hinterland in Sri Lanka (Figure 1). This pa[1]per is based on the results from analyses of artifacts obtained from a series of regional-scale reconnaissance surveys and several stratified excavations at selected cave dwellings. A systematic investigation may open avenues for understanding the cultural dynamics of this time period, and for the development of the subsequent iron-using proto-historic (900–600 BCE) and early historic periods (600 BCE –300 CE) (Deraniyagala 199)

Raj Somadeva is Professor in Archaeology in the Postgraduate Institute of Archaeology, University of Kelaniya, Sri Lanka and a founder Fellow of the Sri Lanka Council of Archaeologists. Anusha Wanninayaka (MSc) is a Research Assistant of the Hunters in Transition Project of the Postgraduate Institute of Archaeology and the National Science Foundation in Sri Lanka. Dinesh Devage (MSc) is a Research Assistant of the Hunters in Transition Project of the Postgraduate Institute of Archaeology and the National Science Foundation in Sri Lanka. Chandima Abesiriwardhana (PhD) is the Principle Research Scientist in the Industrial Technology Institute (ITI) Colombo, in Sri Lanka.

709–713) in a perspective of a long-term cultural continuity. Most of the stratigraphies excavated in caves and alluvial plains suggest an ‘abrupt leap’ from the Mesolithic to the proto-historic Iron Age without leaving any evidence of a techno-cultural shift. The terminal phase of the Mesolithic is archaeologically discernible by the presence of the prehistoric lithic implements and food debris in stratigraphic levels dated to the c. 1800 BCE (Deraniyagala 1972). This leaves a chronological hiatus highlighting of a period of nearly a millennium that remains culturally unexplained.

 Within the HTP project, it was hypothesized firstly that the inferred techno-cultural shift would have been preceded by trans[1]formations discernible in technology, subsistence strategies and cognition within the Mesolithic hunter-gatherer communities, and secondly that these transformations took place as way of ensuring resilience in face of climate variability (DeMenocal 1995, 2004; Petraglia et al. 2010). Thirdly, we hypothesized that intermediate climate zones (situated between wet and dry climatic areas where the annual rainfall vary between 1650 and 2300 mm) that experience regular short-span seasonal variations (rainfall, wind pattern and temperature) demanded a high degree of resilience for survival, and that in these zones we would find a higher degree of trans[1]formations/adaptations than in zones with more stable climates. Building resilience to climatic oscillations would have provided an impetus to generate a comparatively complex behavior among the prehistoric hunter-gatherers making intermediate climatic zone ecologically more sensitive than other zones with more stable cli[1]mates. As has been discussed by a number of scholars (e.g. Akazawa 1982; Gamble 1986; Rowley-Conwy 1986; Shinde et al. 2004), sensitive climatic zones have a greater ecological productivity and an adequate bio-mass to sustain livelihood for a prolonged period. With the objective of addressing the problem of resilience, transformations and adaptations of Mesolithic hunter[1]gatherers, a series of field investigations has been carried out since 2006. The project activities focused on a transect geographically extending from the Horton Plains in the central highlands in the north, stretching along the southern slope passing the Kaltota escarpment down to the UdaWalave plain in the south. This area covers three different climatic zones: a wet zone, an intermediate zone and a dry zone, showing abundance of resource distribution and a complex landscape setting (Figure 1). Here we will present The results of surveys and excavations in the context of our original research questions discussed below in this chapter.

A map showing the distribution of three different climate zones in the study area


Envirronment aand Holoceene climaticc change

The survey area of the project is geographically confined too the southeastern dry-arid climatic zone (<1000 mm mean annual rain-southeastern dry-arid climatic zone (<1000 mm mean annual rain-fall) in the east and the wet lowlands (3500–4000 mmm mean annual rainfall) in the west. The norther periphery is demarcated by the mountain ranges of the watt uplands (22500–3000 mmm annual rain-fall) and the coastal belt in the Sothern dry zone. The middle part of the area experiences intermediate climatic characteristics where the annual rainfall varies between 1000 and 2000 mm (Chandrapala 2007). The terrain diverges from plateaus to rolling slopes and escarpments to valleys. This divergence in topography corresponds to the distribution of various soil types within the area. With the exception of palaeocological studies by Premathilake (2003; see also Premathilake and Risberg 2003; Premathilake and Gunatilleke 2013), our knowledge of the Holocene environment in Sri Lanka is limited. A certain degree of understanding is only possible after comparison with paleo-environmental data from widely scattered geographical areas in South Asia. Overall, the climate regime of Southern Asia has experienced relatively humid condi[1]tions during the Holocene (Singh 1971; Premathilake 2003: 1) mainly due to the monsoonal climatic conditions that affected the greater part of South Asia (Shinde et al. 2001; Krishnamurti et al. 1981). However, the South Asian region have also experienced several changing cycles of dry-wet climates during the Holocene. Sedimentalogical data from saline lakes in Western and Central Rajasthan and Central Gujarat, indicate an inter-pluvial climatic regime had dominated immediately before 10 000 BP (Rajaguru 1973: 69–70; Agrawal and Kusumgar 1974: 64; Gupta 1974: 644– 647; Fairbridge 1976: 542). Comparable data also exist from the Horton Plains in the Central Highlands in Sri Lanka indicating similar climatic conditions. The lower boundary (9900 BP) of local pollen zone 2 (LPAZ 2) indicates an increase of upper montane rainforest arboreal taxa showing a transition of the climate from semi-arid to relatively high humid conditions (Premathilake and Risberg 2003, 7). The pollen records from Rajasthan show the emergence of fresh water lakes suggesting an increase of precipitation (250 mm higher than the modern annual rainfall) (Goudie et al. 1973, 254; Deraniyagala 2004, 157). This wet phase corresponds to the climatic regime signaled by the pollen records from LPAZ 3 (9900–5400 BP) where the increase of upper montane rain forest taxa (e.g. Calaphyllumwalkeri) suggest the amplification of annual precipitation (Premathilake and Risberg 2003, 8). A third pollen phase (LPAZ 4 [5400–3600 BP]) suggests semi-arid environment. Summarizing most of the important works that have been done on the Holocene climatic history in South Asia, Deraniyagala (2004: 158) suggests four climatic trends during the Holocene for Sri Lanka: > 10,000 BP very dry; 6200 BP dramatically wet; 5000 BP dry; 3600 BP dry. These shifts are also likely to have affected seasonality and geographical shifts, particularly in the are[1]as that are now intermediate between the dry and wet regions in Sri Lanka and which has been the focus of our present study. Owing to the insufficiency of recent studies on paleo-climate in South Asia in particular, however, we will base our interpretation on the existing knowledge of climate variability of the region.

The transition between the Mesolithic and the Iron Age The terminal phase of the Mesolithic is archaeologically discernible in stratigraphic levels ascribed to the early second millennium BCE (Deraniyagala 1992). 14C dates from charcoal samples obtained 10.20 m below the surface of a mound excavated at Matota in the northwestern arid zone has revealed three dates: 3830, 3850 and 4170 cal BP (Deraniyagala 1992: 701) . The terminus ante quem of the Mesolithic culture at this site dates to c. 3800 cal BP and correlates geologically with the event of the Younger Peron high sea[1]level (Fairbridge 1976). These dates suggest the most recent relia]ble date assigned to the Mesolithic culture in Sri Lanka goes back to c. BCE 1800. The period from 900 BCE is characterized by the emergence of iron technology, a local variety of Black and Red ware pottery, cultivation of rice and the use of domesticated cattle (Deraniyagala 1970, 1992: 709), based on excavations conducted at the ancient citadel area in Anuradhapura. Another attribute of newly emerged protohistoric Iron Age is the megalithic cemeteries scattered pre[1]dominantly in the dry zone areas. Only four cemeteries (i.e. Ibbankatuva, Yatigalpotta and Kalavallaulpota in the Matale District, Galsohonkanatta in the Kurunegala District) have been investigated archaeologically. Dates range over the period BCE 768–383 (Deraniyagala 1992: 734; Bandaranayke 1992). The material culture (use of Iron and the emergence of agriculture) manifested by the stratigraphic levels in the Anuradhapura citadel excavation has no indication of mixing or connection with Mesolithic material, and therefore an abrupt leap from the Mesolithic to the iron-using culture has been suggested.

The uncalibrated dates are 3520±45 bp (BM-2340); 3550± 70 bp (BM-2341); 3790±70 bp (BM-2342). Dates have been calibrated to BP dates using Pearson and Stuiver (1986) (Deraniyagala 1992:701).

Hunters in Transition Project

Hunters in Transition Project

Surveys and excavations

As discussed above, the ecological focus of the HTP was a wider transects stretching from the Handagiriya plateau crossing the Kaltota escarpment to the southern slope of the central highlands (Figure 2). Fieldwork in the years from 2006 to 2009 focused on the middle basin of the river Walave, where the elevation of the terrain was confined to 152 masl. Subsequently surveys were shifted towards the headwaters of the same river situated up to 914 masl. The 2010–2011 fieldwork included an extensive ground recon[1]naissance survey program and two excavations were conducted in the higher elevations (800–3000 feet masl) of the southern slope of the central mountains. The main focus of the ground reconnaissance survey was the geographical area extending from Opanayake to Haputale, which is approximately 50km apart. Several archaeological sites were located, scattered on the mountain slopes (3000–5000 feet amsl) from the Horton Plains down to the Haputale Plateau on the southern gradient of the central mountains. The scatter of sites suggested a dispersal of settlements across different biomes (for further reference on transhumance, see Butzer 1971, 1985; Lieth 1973). The surface survey has revealed a wide scatter of sites (identified from quartz and chert implements on the surface) on the mountain slopes. Ninety individual locations were identified within about 20 square km of the transect. The number of identified sites, at first glance, does not suggest a thick density of sites. However, on the basis of the regional land registry information maintained by the State Department of Land Reclamation we estimate that only 40% of the original number of sites remains after the destructive landscape modifications taking place in connection with expansion of tea plantations since the early 19th century.

Figure 2. A map showwing the intensivelyly surveyed transeect of the study aarea.

The surveys identified a number of potential sites for further execrations focusing on the middy/late Holocene and geographical rep-recantation in the three separate climatic zones outlined above. recantation in the three separate climatic zones outlined posits that are confined to less than five separate soil layers. Four excavations were carried out in four selected caves (Rassagala, Lunugallge, Udupiyyangalge and Alugalge), each representing different altitudinal and geographic zones.


Rassagala Cave is situated in the wet uplands at an elevated height (730 masl). The cave is located in the village of the same name, belonging to the Polvatugoda GS division of Balangoda. There is a huge rock that could be seen on the summit of this height. Be[1]neath the eastern edge of the rock there is a cave, its mouth facing southeast. Dimensions of the interior of the cave are 10.5 x 17.4 m. At the time of the first visit to this location there was a construction of a room covering a greater part of the interior space. The cave has rubble walls erected up to the roof level of the cave. Subsequently, the construction was removed. The sub-surface of the floor area seemed somewhat disturbed owing to the digging of foundation trenches. According to village informants, this place had been occupied twice by Buddhist monks for short periods sometime back, the latest being in the early 1960s. According to a date inscribed on the rock surface, the cave was also reoccupied by monks in 1983. Except for the disturbances caused during the recent constructions, no other serious transformations seemed to have occurred at the location. The surrounding landscape of the Rassagala Cave has been modified for tea cultivation since the mid-20th century. Only a tiny part of undisturbed vegetation re[1]mains in the adjoining land strip near the cave. On the right side of the cave, there is a heap of soil, probably washed down through slope processes from the hilly terrain above the cave. There is infiltration of soil and other colluvial materials on the right-side edge of the cave that can be attributed to soil erosion. Debris amassed inside the cave might have been removed during the construction of the dwellings for the monks.

On the left side of the cave, a huge rock rises vertically about 30 m forming a fairly protective shelter. The floor area of this rock shelter has also been covered by a masonry foundation constructed in 1983.

Excavation at Rassagala Cave has revealed a collection of prehistoric lithic implements comparable morphologically as well as typologically with the other cave sites excavated in the vicinity. However, the dating of the site was constrained by the contamination of sub-surface layers by the modern occupation in the cave (see below).


Lunugalge (550 masl), is situated about 30 km south of Rassagala in the border between the wet and intermediate climatic zones. The cave is situated in the village Illukkumbura of the Valigepola Divisional Secretariat region. Lunugalge Cave is situated on a narrow plateau in a steep mountainous slope with its mouth facing south. Dimensions of the cave are 15.5 x 5.9 m and the maximum height of the cave roof is 5.4 m. A series of cavities visible on the cave wall suggest a prolonged physical erosion. Despite detailed surveys, no painting or etching of a prehistoric origin could be seen on the rock wall. The left-side portion of the cave floor, down to 2 m from the surface (22 m2 , which is 24% of the total area of the floor), had been destroyed by treasure hunters (Figure 3). Diggers had dumped the soil of their illegal trench on the cave floor at the right side. A large heap of excavated soil was still remaining when we first visited the cave, as a result thousands of prehistoric stone implements were scattered in and around the cave. Three pitted-hammer stones were collected from the surface of the cave interior during the excavation. Lunugalge Cave revealed three consecutive soil layers containing prehistoric lithic implements (similar to that found in Udupiyangalge Cave presented below). The most important finding from the cave was an assemblage of charred seeds of 25 individual varieties recovered from the soil layers of prehistoric origin (see below). The top[1]most soil layer of the interior would have been occupied in the early historic period and it suggests a prolonged occupation of the cave perhaps intermittently for at least 3500 years onwards since midHolocene.


Udupiyangalge (250 masl) falls in to the dry zone and was excavated in early 2016 (Figure 4). The cave is situated on a hillslope below the village Molamure of the Balangoda Divisional Secretariat division. Udupiyangalge is a fairly large sheltered space facing southern direction. Total habitable area of the interior is approximately 330 m2 . The interior floor area appears divided into two levels; the remains of the lowest floor, level 1, can be found in the right corner. The floor of the right side had been elevated by the construction of a low retaining wall to create required space for ritual and residential requirements. According to the local informants, the upper level (0.58 m above the lower floor, 250 masl) was filled with soil brought from outside of the cave premises. An abandoned gem-pit dug out by the villagers is found on the right side edge of the cave. Udupiyangalge Cave was archaeologically investigated by P.E.P. Deraniyagala in 1936. He did a brief investigation in the interior cave floor and stated that:

… the presence of the semi-lunate microliths and the pebble hammer stones suggests that this was a Mesolithic culture that had evolved into the Neolithic (Deraniyagala 1958:246).

In 2016, excavations were carried out at a location in the elevated terrace in the left-side edge of the cave floor of an area 3 x 2.40 m. Excavations reveled three consecutive soil layers down to 1.55 m. The first two layers consisted of prehistoric lithic implements (quartz and chert) of several new forms which are not previously seen in lithic assemblages in Sri Lanka. The density of the artefacts gradually decreased down to the third layer. A rich collection of bone tools were also found. The presence of a bulky assemblage of prehistoric food residues including charred animal bones and land snail shells (Acavus sp., Paludomous sp. and Oligospeira sp.) may suggest either a prolonged occupation or an existence of a considerably large group on a seasonal base in the cave. Two pendants made out of chert flakes are the most fabulous artefacts that have been recov[1]ered from the excavation, and which we interpret as symbolic (see below).


Alugalge Cave (360 masl) is situated in the threshold region between wet and intermediate climatic zones (similar to Rassagala Cave, presented above). Located in the village Illukkumbura, the cave sits on a steep slope in the woods of the mountainous landscape of Maddekanda in the village Illukkumbura. The mouth of the cave is facing a southeastern direction and it prevents the flow of wind and water into the cave during the monsoon rains. The width of the entrance of the cave is 6 m and the maximum roofheight is 2.5 m. Both sides of the entrance are enclosed by rock boulders of different sizes forming a cladding effect to the interior. A considerable accumulation of rock fall from the cave roof was observed in the interior before the excavation. The approximate extent of the cave floor is 30m2 . The cave interior has a prehistoric deposit of 1 m depth. The first season of the excavation in Alugalge (July 2016) revealed an assemblage of artefacts which is similar to the other caves excavated in the area but shows some striking characteristics pertaining to the prehistoric economy during the mid-Holocene. Shark teeth and several pieces of corals indicate exchange with the coast, 40 km south to the cave. As we saw in Lunugalge, the pre[1]historic occupation in the cave was continued for a considerable period as suggested by the AMS dates.

Results Surveys in 2013 yielded an assemblage of material evidence including grind-stones, pestles and finely made microlithic implements showing a probable new leap towards the intensive utilization of wild grasses still available in the surrounding landscape. 14C dates obtained from the recent excavations indicated that such a deviation could be securely placed somewhere in the 5th millennium BCE. Therefore the cave site excavations focused on this time period, and fortunately we also obtained more detailed data from botanical remains to address the resource shifts. Below, we will discuss the results from excavations when it comes to resource exploitation, technology and symbolism.

Resource exploitation

As discussed below, basic experiments on exploiting wild floral resources (seeds, grains and nuts) are suggested to have been initiated around BCE 5000 (Table 1) in the marginal area between wet and intermediate climatic zones in the uplands. This is archaeologically indicated by an assemblage of charred seeds excavated from three caves (Lunugalge, Udupiyangalge and Alugalge) situated in wet, dry and intermediate zone respectively. Preliminary investigations show that there are 25 different varieties of seeds/nuts (see Table 2) in the excavated assemblage. One identification could be made visually of the nut of Carneriumzeylanicum (local name dikkekuna), the species of which is endemic to the wet zone in Sri Lanka. The presence of this seed in Lunugalge and Udupiyangalge is a continuation of a prolonged experience of the Mesolithic hunter-gatherers occupying the wet lowlands, as suggested by finding the same in other locations in Sri Lanka (Deraniyagala 1992: 452).

Most of the cell structures of the optically analyzed seeds were destructively changed owing to the high temperatures during food preparation; therefore laboratory testing (pharmocognistical analysis) of the plant remains was conducted in the Industrial Technology Institute (ITI) in Colombo. Microscopic observations of the thin-sections obtained from the individual seeds show the presence of different characteristics, such as sclerenchyma tissues, annulerxylem vessels and trichomes identical to modern plant residues (Figure 5a, b). The preliminary analyses show that the pharmocognistical method is a productive technique to identify the floral origin of morphologically transformed plant materials.

An ethnobotanical plant survey (Figure 6) carried out by the local medical practitioners in thee area around Lunugala Cave has also guided the tentative identification of some of the seeds in the as-semblage (Figure 8). All seeds represented in the archaeological excavations are locally available in thee surrounding landscape to-day. Some varieties are currently used by the local medical practitioners. The presence of Kodo-millett (Paspalum sp., local name:amu) in the excavated seeds collection could be taken interesting case. The oldest evidence of millet been reported from as an South Asia is from a site off Harappa inn northern Pakistan dated to a period between BCE 33000 to 1900 (Kenoyer 19998; Weber 19999).

However, the presence of wild millet in this assemblage suggests ex-peri mentation with this species before domestication. The grind stones and pestles (Figure 7) found from a location in Walmeetalava of Haaldummulla during the site survey inn 2013 could be taken ass one of thee technological inventions prompted by the intensification of the exploitations of wild floral resources (Somadeva 2014). In addition, several pieces of crude hands-made pottery have been discovered on the surface at a location called Mantena in Haldummulla during the reconnaissance survey conducted in 20010 (Somadeva 2014). The origin and dating of these ceramic shreds remains to be confirmed, but we find it plausible that they belong too a pre-Iron Age context. When it comes to the boon assemblages, bones were recovered and analyzed from the Lunugalge Cave, Uduppiyangalage and Alugaalge. The identified bones represent a variety of animal species, notably small animals, rodents, cats, civets or reptilians. Monkeys were also represented,, primarily small monkey’s. The choose of relatively small game might have been a decision taken by the Holocene hunter-gatherer groups to reduce the energy invested in hunting (see Table 3). Especially in comparison with the other Mesolithic sites in the country’s, the decrease of haunting ceroids such as spotted deer (Ax-is axisis) and Sambhur (Cervus unicolor) is noteworthy. This choice might have been influenced by the relative abundance off plant food re-sources that would harness the energy input to hunt big game animals.


The stone implements recovered from the surface of more than 80 individual locations scattered in the western slope of the Central Highlands show a considerably higher degree of technological di-varsity in comparison with the similar artifacts recovered from other Mesolithic hunter-gatherer sites in the country (Deraniyyagala19922: 185; Perera 2010: 1107). Manufacturing of form-trimmed implements are considerably more common in excavated context than elsewhere in Sri Lanka: 24.92% (n=1226) of the total assm- blage sampled ((n=4911) are form-trimmed implements and this is an increase of 12.46% (Figure 9). The thigh representation of form-trimmed implements is very different from other collections of Mesolithic implements analyzed by Perera (2010), where represent- station of form-trimmed implements is just 0.2 %. In the contexts presented here, a greater preference is also discernible for Manu- fracturing fine quartz blades. New plane forms including arrowheads, fish-hooks, chisel-type implements (burins??) and stone e nee-dales emerged (Devage 20144). The archaeological layers in the excavation at the Lunugalge Cave has sown an increase of manufacturing stone projectile points (quartz)) over time, suggesting the hutting of small animals like specially birds and rodents. This core-responds to thee faunal assemblage from Lunugalge where small animals are in majority (Table 3).

Symbolic artifacts

Symbolic artifacts found in the material repertoire of Mesolithic hunter-gatherer sites in Sri Lanka are not well represented in comparison with other material objects. However, the four caves excavated have yielded artifacts that suggest that considerable energy has been invested in producing what seems to be non-utilitarian objects (hence their interpretation as symbolical). Some of the most elaborated artifacts present in the collection are described below. Pitted stone (Figure 10a): The pitted sandstone that has 19 shallow cup-marks was reported from context 4 in the Illukkumbura Cave. The length of this object is 32 cm and the maximum width is 17 cm with a thickness of 0.9 cm. The depth of the cup[1]marks remains within a range of 0.3 cm and 0.8 cm. The perimeter of the hewn area was clearly delineated by a line along the edge of the rock and confirms that the entire piece has been decorated purposefully. Stone pendants (Figure 10b): A stone flake (red chert) intentionally shaped into a human heart (length 2.55 cm and width 1.8 cm) has been discovered (context 3) from the interior of the Udupiyangalge Cave. The upper part of this object has a tiny perforation that would have been utilized to hang it by a thread. Substantial proof is available in prehistoric caves in Sri Lanka to show as such perforated objects were used as beads or pendants (Deraniyagala 1992; Perera 2010). Two similar artifacts were recovered from this cave. Beads (Figure 10c-f): The excavation at Alugalge Cave has revealed 4 beads made out of animal bones and teeth. The bead of animal tooth (level III) shows a comparatively high skill of craftsmanship: its cylindrical shape has been enhanced by two edges to obtain a collar-like appearance. The shaping of the tooth required hard labor and curation. The most sophisticated bead recovered is an altered shark tooth. Its nutritive groove was enhanced though a perforation. The natural color variation between the crown and the roots in the tooth together with the natural symmetry most probably created an aesthetic appeal to the bead. A diagram (Figure 11): A gneiss stone slab (0.46 m x 0.30 m x 0.10 m) recovered from the Alugalge Cave has a depiction of a grid demarcated by a series of geometric lines on one of its surfaces. The lines had been defined using a certain kind of sticky glue. The pale whitish color of the patches of glue lines contrast with the dark surface of the stone. Nine vertical lines and four horizontal lines were found remaining in this diagrammatic representation. Three of the lines had been drawn as slightly angular projections. The artist responsible for creating this graphic representation had the intention to follow a certain tempo on his work as suggested by the distance variations between the vertical lines and the sporadic discontinuations maintained in the horizontal lines.

Phallic objects (Figure 10g-i): An assemblage of ancient phallic representations made out of stone (gneiss and sandstone) have been unearthed from the interior of the limestone cave excavated in Walmeetalava of Waeliya in Haldummulla. The collection included sculptural icons of 2 female pubic triangles, a single portrayal of a female lower body part with enhanced sexual organs and 2 male phalluses. It is a well known phenomenon that the emergence of making human body parts in an iconic form was a widespread practice among the Neolithic societies in different geographic areas in the world. Neolithic phallic icons have been interpreted along diverse intellectual avenues (Bailey 1996; Haaland and Haaland 1996; Hamilton 1996; Marcus 1996; Ucko 1996; Lesure 2002). The existence of phallic objects in the present cave suggests an emergence of a system of new beliefs centered on somatic representation among the advanced hunter-gatherers of the midHolocene.

Animal motives (Figure 12): A terracotta motif of a dog (Canis sp.) has been reported from level III of the excavation in the interior of the Alugalge Cave. It is a terracotta work that has a thin coating of lime on the surface. The evidence suggesting the presence of dog in the prehistory of Sri Lanka is scanty. Skeletal remains of a dog were reported from Bellanbandipalassa dated to c. 6500 BP and also from Nilgala Cave in the Ampara District (Deraniyagala 1992: 454). A global scale study of mitocondrial and Ychromosome DNA of the modern dogs suggests their single East Asian origin (Savolainen et al. 2002; Ding et al. 2012). Domestication of dog in that region had occurred in a period between 14000 and 9000 BCE (Freedman et al. 2014).


From the site surveys we can conclude that there is a wide distribution of individual sites, most of them probably only temporary camps, suggesting that the contemporary hunter-gatherers had invested much labour in the quest for food, searching a wider area along the mountainous tracts. Unpredictability of edible resources in a fluctuating climatic condition of the Holocene would have have influence on decision making, and the indubitable response on this nature was to turn to reliance on gathering more floral resources. The major characteristics visible in the archaeological matrix of this period are the presence of charred botanical remains and several new form of technologically refined lithic implements. This shows a receptiveness by the traditional hunter-gatherers towards intensive plant exploitation.

In the introduction and as a premise to the HTP project we proposed that pre-adaptations in Mesolithic hunter-gatherer com[1]munities made way for later Iron Age cultural and technological transformations. As shown here our surveys and excavations suggest an intensification of the use of wild plant resources around 5000 BCE, with findings of grindstones and use of wild millets as well as other plant material. We have also found evidence of technological changes in lithic technology and what can perhaps be interpreted as intensified use of symbolic expression in the same contexts.

Although more data is needed together with complimentary analyses of both archeological and paleoecological data, the results of the HTP project surveys and excavations suggest a culturaltechnological shift in the mid-Holocene that does correlate with the shift from wet to dry conditions at 5000 BCE. Most of the evidence of transformations are also from the intermediate climate area, i.e. the area that today lies in-between the wet and dry areas and which we expect experienced high variability in the period of a regional climate regime shift. Thus, tentatively we also have some preliminary evidence that in areas of higher environmental variability communities were more adaptable, in terms of transforming livelihoods from a broad spectrum to an intensified use of plant resource, to enhance resilience. By BCE 5000, the climatic conditions of the mid/late Holocene in the mountainous hinterland in Sri Lanka had critically influenced the Mesolithic hunter-gatherer population. The effects we further augmented by the regional climatic variations (wet/dry/intermediate) that made a distinction in the local stock of flora and fauna. The stress triggered by a fluctuating climatic regime demanded a high degree of resilience from the prehistoric groups, resulting in local transformations manifested by material culture. The symbolic objects presented above indicate that inhabitants of the investigated caves had attempted to see the external world in a radically new and different way than their predecessors. This move does not mean that they had fully transformed to a novel life style, but we argue that there had been a fundamental change both in livelihood and mentalities that made way for later transformations in lifestyles with the introduction of the Iron Age.


We are greatly indebted to the National Science Foundation (NSF) of Sri Lanka to raise funds (grant number IK/2014/1) to conduct this research. It was also partly funded by the Postgraduate Institute of Archaeology of the University of Kelaniya and the Yuga Vimasuma Society in Colombo.


Agrawal, D.P. and Kusumgar, S. 1974. Prehistoric Chronology and
Radiocarbon Dating in India. New Delhi: Munshiram Manoharlal.
Akazawa, T. 1982. Cultural change in prehistoric Japan: Receptivity to rice agriculture in the Japanese archipelago. In: Wendorf,
F. and Close, A.E. (eds), Advances in World Archaeology 1, 151–

  1. New York: Academic Press.
    Bailey, J.F., Richards, M.B., Macaulay, V.A., Colson, I.B., James,
    I.T. et al. 1996. Ancient DNA suggests a recent expansion of
    European cattle from a diverse wild progenitor species. Philosophical Transactions of the Royal Society Series B 263: 1467–1473.
    Bandaranayake, S., 1992. The settlement patterns of the protohistoric-early historic interface in Sri Lanka. In: Jarrige, C. (ed.),
    South Asian Archaeology, Papers from the Tenth International Conference
    of South Asian Archaeologists in Western Europe, 15–24. Madison:
    Prehistory Press.

Butzer, C. 1971. Environment and Archaeology: An Ecological Approach
to Prehistory. Chicago: Aldine.
Butzer, C. 1985. Archaeology as Human Ecology. Cambridge: Cambridge University Press.
Chandrapala, D. 2007. Rainfall: Physical and Biological Environment.
The National Atlas of Sri Lanka, 2nd edition, 58–59. Colombo:
Survey Department of Sri Lanka: Colombo.
DeMenocal, P.B. 1995. Plio-pleistocene African climate. Science 270
(5223): 53–59.
Deraniyagala, P.E.P. 1958. An open-air habitation site of Homo
sapienbalangodensis. Spolia Zeylanica 28: 223–261.
Deraniyagala, S. 1970. The citadel of Anuradhapura in 1969 excavation in the Gedige area. Ancient Ceylon 2: 148–169.
Deraniyagala, S. 1992. The Prehistory of Sri Lanka. Colombo: Department of Archaeological Survey.
Devage, D.D. 2014. Technological Characteristics of the Stone Implements
in the Pre- and Protohistoric Transition in Sri Lanka (unpublished
MSc thesis). Postgraduate Institute of Archaeology. Colombo:
Postgraduate Institute of Archaeology.
Ding, Z., Oskarsson, M., Ardalan, A., Angleby, H., Dahlgren, L. et
al. 2012. Origins of domestic dog in southern East Asia is supported by analysis by Y-chromosome DNA. Heredity 108(5):
Fairbridge, R.W. 1976. Effects of Holocene climate change on
some tropical geomorphic processes. Quaternary Research 6: 529–

  1. Freedman, A, Gronau, I., Schweizer, R., Ortega-Del Vecchyo, D.,
    Han, E. et al. 2014. Genome sequencing highlights the dynamic
    early history of dogs. PLoS Genetics 10(1): e1004016.
    Gamble, C. 1986. The Mesolithic sandwich: ecological approaches
    and archaeological records of the early postglacial. In: Zvelebil,
    M. (eds), Hunters in Transition Mesolithic Societies of Temperate Eurasia and their Transition to Farming, 33–42. Cambridge: Cambridge
    University Press.
    Goudie, A.S., Allchin, B. and Hegde, K.T.M. 1973. The former
    extensions of the Great Indian Sand Desert. Geographical Journal
    139(2): 243–258.

Gupta, H.P. 1974. Late Quaternary vegetational history in western
region. In: Suranga, K.R., Lakhanpal, R.N. and Bharadwaj,
D.C. (eds), Aspect and Appraisal of Indian Palaeobotany, 644–650.
Luckow: Birbal Sahni Institute.
Hamilton, N. 1996. The Personal is Political. Cambridge Archaeological Journal 6(2): 282–285.
Haaland, G. and Haaland, R. 1996. Levels of meaning in symbolic
objects. Cambridge Archaeological Journal 6(2): 295–300.
Krishnamurty, R.V., Agrawal, D.P., Misra, V.N. and Rajaguru, S.N.

  1. Palaeoclimatic influences from the behaviour of radiocarbon dates of carbonates from sand dunes of Rajasthan. Proceedings of the Indian Academy of Sciences (Earth Planet Science) 90: 155–

  2. Lesure, R.G. 2002. The Goddess diffracted: thinking about the
    figurines of early villages. Current Anthropology 43(4): 587–610.
    Lieth, H. 1973. Primary productivity: Terrestrial ecosystems. Human Ecology 1: 303–332.
    Marcus, J. 1996. The importance of context in interpreting figurines. Cambridge Archaeological Journal 6(2): 285–291.
    Pearson, G.W. and Stuiver, M. 1986. High-precision calibration of
    the radiocarbon time scale, 500–2500 BC. Radiocarbon 28 (2B):
    Perera, N. 2010. Prehistoric Sri Lanka Late Pleistocene Rock-shelters and
    an Open-air Site. BAR International Series 2142. Oxford: Archaeopress.
    Perera, N., Kourampas, N., Simpson, I.A., Deraniayagala, S.U,
    Bulbeck, D. et al. 2011. People of the ancient rainforest: Late
    pleistocene foragers at the Batadomba-lena rock shelter, Sri
    Lanka. Journal of Human Evolution 61(3): 254–269.
    Petraglia, M.D., M. Haslam, D.Q. Fuller, Boivin, N. and Clarkson,
    C. 2010. Out of Africa: New hypotheses and evidence for the
    dispersal of Homo sapiens along the Indian Ocean rim. Annals
    of Human Biology 37(3): 288–311.
    Premathilake, R. 2003. Late Quaternary Palaeoecological Event Stratigraphy in the Horton Plains, Central Sri Lanka. Stockholm: Stockholm
    Premathilake, R., and Risberg, J. 2003. Late Quaternary climate
    history of the Horton Plains, central Sri Lanka. Quaternary Science
    Review 22: 1525–154.

Premathilake, R., and Gunatilaka, A. 2013. Chronological framework of Asian Southwest Monsoon events and variations over
the past 24,000 years in Sri Lanka and regional correlations.
Journal of National Science Foundation in Sri Lanka 41(3): 219–228.
Rajaguru, S.N. 1973. Late Pleistocene climatic change in western
India. In: Agrawal, D.P. and Ghosh A. (eds), Radiocarbon and
Indian Archaeology. Bombay: Tata Institute.
Rowley-Conwy, P. 1986. Between cave painters and crop planters:
aspects of the temperate European Mesolithic. In: Zvelebil, M.,
(ed.), Hunters in Transition Mesolithic Societies of Temperate Eurasia
and their Transition to Farming, 17–32. Cambridge: Cambridge
University Press.
Savolainen, P., Zang, Y., Lunderberg, J. and Leitner, T. 2002. Genetic evidence for an East Asian origin of domestic dogs. Science
298 (5598): 12387–12390.
Shinde, V.S., Yasuda, Y. and Possehl, G. 2001. Climatic conditions
and the rise and fall of Harappan Civilization of South Asia.
Monsoon 3: 92–94.
Shinde, V.S., Sinha-Deshpande, S. and Yasuda, Y. 2004. Human
response to Holocene climate changes in western India between
5th and 3rd millennium BC. In: Yasuda, Y. and Shinde V.S.
(eds), Monsoon and Civilization, 383–406. New Delhi: Roli Books.
Singh, G. 1971. The Indus Valley Culture. Archaeology and Physical
Anthropology in Oceania 6(2): 177–189.
Somadeva, R. 2014. Archaeology of Mountains. Colombo: Postgraduate Institute of Archaeology.
Ucko, P.J. 1996. Mother, are you there? Cambridge Archaeological
Journal 6 (2): 300–304.
Weber, S. 1999. Seeds of urbanism: Paleoethnobotny and the Indus civilization. Antiquity 73(282): 812–826

Related Posts

error: Content is protected !!