ARCHAEOLOGY OF DISORDER: A SITE-SPECIFIC SURFACE STUDY AT A NEWLY DISCOVERED ANCIENT MOUND IN THE SABARAGAMUVA PROVINCE, SRI LANKA
The objective of this article is to present the results of an archaeological surface survey conducted at a natural mound that had a dense scattering ‘of artefacts on its surface, in a village called Galpaya in the Ratnapura District of the Sabaragamuva Province This article discusses the theoretical issues pertaining to fieldwork, particularly the methods of handling surface artefacts such as sampling, recording, quantifying and analyzing in an intra-site scale survey, with special reference to the site referred to. To begin with, an overview of the ideas expressed by different authorities on the subject is presented as a prelude to the discussion that follows in the rest of the article so as to render it easier for the reader to grasp the background of the subject focused in the article.
One of the inherent characteristics of most of the archaeological finds is their secondary nature. Artefacts transform from their systemic context to the archaeological context (Schiffer 1996) through the artefact itself as a physical entity, but its primary utility matrix – I use this term here instead of the word ‘function’ -also changes, disappears or transforms through time. Perhaps these taphanomic processes pose two major challenges that archaeologists should invariably confront in order to interpret the ancient material culture. They are: (1) conversion of the existing patterns of artefact arrangements into an anthropological meaning that would generate ideas pertaining to their associated behavioral norms; (2) provision
of a temporal scale to organize the entire spectrum of such arrangements in order to show the development aspect of the events which is an integral part of the long-term processes that are reflected by the artefacts.
Study of past human behavior and culture is a task, which is common to both archaeologists and anthropologists (Schiffer 1999). Reconstruction of past human behavior and its driven ideas are mostly inferred from material remains. Different aspects of the artefacts are taken into consideration to reach the desired interpretative goals. The most straight forward and what seems the more appropriate among them to sort the behavioral affiliations of such is the identification of artefact distribution patterns that are found on the surface. Sometimes such attempts may end up with a scope of subjectivity but as Kent (1987) points out, archaeologists are pessimistic in addressing at least three specific questions pertaining to their material assemblages especially in the site-specific intra-site level studies, viz:
1.How do people organize and use space?
2.What are the variables potentially influencing this behavior?
3.What is its interrelationship with culture materials and aspects of culture?
On a qualitative or a quantitative basis, it is taken for granted that the artefact distribution patterns somehow delineate the behavioral patterns in which the relevant material entities are involved (e.g. Spaulding 1960). But there are different factors which
constrain the archaeologist’s inference of that which appears even at the data collection stage. Different site formation processes are responsible in rearranging the surface artefact scatters in various scales and intensities.
Particularly the archaeological sites distributed over modern agricultural areas are highly vulnerable in preserving their surface artefact scatter patterns. The majority of the ancient settlement sites in the dry zone areas in Sri Lanka are severely affected by agricultural ploughing and therefore plough-induced artefact patterns are common occurrences on the surface of most of the sites identified. The question of distinguishing between ancient behavioral patterns and the plough-induced patterns is sometimes subjected to a heated debate among the field archaeologists. Lewarch and O’Brien (1981) have pointed out five important factors that should be considered in the recognition of the surface artefact pattern, viz: (1) horizontal displacement (2) vertical displacement (3) changes in class frequencies that take place after ploughing (4) changes in the condition and preservation of artefact assemblages and (5) destruction or alteration of features and layers. The observations made during fieldwork carried out in the last decade in the dry zone areas of Sri Lanka suggest that, out of the five factors highlighted above, the commonest and the most frequent is the horizontal displacement of artefacts. The degree of horizontal displacement of surface artefacts in a given site may vary for several reasons. As pointed out by Roper (1976) and Trubowitz (1978), the size of the artefacts and the type of equipment are the most critical factors in this regard. Lambrick (1977) shows that the degree of horizontal displacement caused by the mouldboard plough is lower than that caused by the disc plough. Closely maintained observations show that ploughing in alternative directions results in a minimal longitudinal displacement of about 5 meters over time (Clark and Schofield 1991). Our observations made
at several sites in the southern and southeastern dry zone areas in Sri Lanka show that the plough-induced horizontal displacement of artefacts made a lesser impact and was yet the most influential in the intentional clustering. Intentional clustering is an action that takes place when farmers tend to purposefully gather the artefacts, (particularly potsherds) to clear up their agricultural fields. In the process of such intentional gathering, farmers dump such accumulations in nearby locations or otherwise push them away up to the border of the cleared area.
Individual choice of selecting the dumping locations in such cases is based on spontaneous decisions of individuals and as such no prediction can be made on the probable distance of the horizontal displacement.
THE CASE STUDY
The present site of Valamkatuyaya presents another unusual experience in the site formation process, which is not adequately discussed in the literature. Some important studies however, have discussed the theoretical aspect of the intentional effects caused upon surface artefact assemblages. But the present study deals with an extensive surface assemblage that had been intentionally moved and displaced. As told by village informants, the mound had been looted by treasure hunters for a period of nearly five years. Hundreds of robber pits still visible on the mound suggest the high degree of vandalism that had occurred at the site. Perhaps this is the first occasion in Sri Lanka, in which a large-scale archaeological site, which has been subject to continuous and intentional destruction over a considerably long period, has been found.
Compared to the issues pertaining to site formation resulting from agriculture related activities as mentioned in section 1, the agents of the site formation in the present mound are completely different. The
activities that have occurred at Valamkatuyaya were intentional and have directly focused upon the artefacts and buried structures. Three major modes involved in site formation at the location during recent times have been identified.
- Mixing of different soil layers in the sub-surface during illegal diggings (which has resulted in a higher degree of decontextualizingof artefacts)
- Removing of some selected artefacts from the site (which has affected the quantitative and qualitative domain of the content of material evidence of the site)
- Displacing of artifacts while being taken away to the nearby river for cleaning and washing. This is reflected in the artifact scattered on the riverbank and the riverbed (resulting in an off-site artifact assemblage that has less interpretative significance.
Beside the international distribution made upon the cultural content at the site, the other site formation processes involved are discussed in section 3.2 given below.
The Galpaya Project
In June 2006, a four week fieldwork session was organized by the Postgraduate Institute of Archaeology in order to initiate a comprehensive surface documentation at the mound called Valamkatuyaya (Map 1) in the village of Galpaya in the Ratnapura district of the Sabaragamuva province (Somadeva et al 2007). This site was identified as one of the major locations of the ancient settlement developed on the northwestern frontier area of the south and southeastern territory of Sri Lanka, which was historically known as Ruhuna (Mv.ix:10).
The purpose of then first field session was to carry out and initial controlled survey and documentation leading to a series of excavations scheduled to be conducted in the second field session in the year 2008. There was three objectives in the documentation viz. (1) to identify the size of the site (2) to identify the horizontal distribution variation of the artifacts in order to map the specific behavioral location on the mound and (3) to establish a relative chronological sequence in the development of the settlement at the location using earthenware shreds.
The Valamkatuyaya (lit. the yard of the potsherds) is a natural mound which is a common features of the undulating terrain of the dry zone of Sri Lanka (Cooray 1984). It is situated about 16 km east of RathnapuraEmbilipitiya main road. There are two access roads to the site, one which lies from Pallebadda, the town via the village of the Telambugahaatura and other from a place close to the Kolambageara junction via Panahaduva. The site is bordered by the Uda Walava wildlife sanctuary to the north and the east.
The mound covers about 75 hectares most part of which are now occupied by farms and home gardens. It was reported that the site had been intermittently dug out by treasure hunters for at least for a period of five years and that the information about such activities had not reached the archaeological authorities due to the remoteness of the location and perhaps the political strength of the organized looters who were capable of the forcefully influencing regional government authorities.
As already mentioned above, the activities of the first field session were confined to complete documentation of the mound surface at a micro level and the immediate landscape of the mound at a micro level. The area covered by the surface scattering of the artifacts on the mound is approximately 460 m x 165 m. and a part of that extent was recorded as a representative sample. The uneven distribution of the artifacts on the mound surface suggest that it either had resulted from the organization pattern of the space of the people who had inhabited the mound before it was abandoned or from the forces of the post-abandonment period (for further details, Somadeva et al. 2008). It is difficult to ascertain the exact hight of the mound because of the distribution caused by treasure hunters. According to the information furnished by village elders of Galpaya, the mound was nearly 9.1 to 10.7 meters above the surrounding ground before the looting began. At present the minimum height of what has remained does not exceed 1.5 to 1.8 m above the surrounding ground level.
Other than the northern sector (95×165 meters), the remaining area of the mound is currently occupied by villagers. There houses are made of wattle and daub with thatched roofs and hardly ever present any example of having been covered steel sheets. Some parts of the northern sector are being cultivated by local farmers. On the western side of the mound, there is a man-made reservoir called Galpayavava, probably a later construction of the historical period. The earthen wall of the reservoir runs in a northsouth direction and at the northern end of the mound it turns westwards to cross at the confluence of the adjacent rivulet called Kuda Oya, with the river, Walaveganga. This earthen embankment stands as the western boundary wall of the mound. A large depression on the eastern side of the mound is probably a remnant of an ancient moat running in a north-south direction parallel to the mound’s edge on the eastern side. This moat had been connected to the river, Kuda-Oya, as suggested by remains of the earthen embankments, which are still visible in the neighbourhood of the mound. Further south of the mound there are two places with ruined stone pillars. The largest among them is the one that is situated at the base of the hillock called Ranminikota. It had been a Buddhist monastery, which was perhaps, maintained by past inhabitants of the mound. A fragmented pillar inscription of the ninth century CE at the site suggest that this monastery functioned at that time (Map 2).
The present site lies between two major mountain ranges in the country i.e. the Rakvana-Bulutota massif and the Kaltota-Balangoda mountain ranges. The Kaltota-Balangoda escarpment comprises several other low-height ridges namely Bambaragala, Heendodangahatenna, Ulgala and Hataramune.
These low-height ranges stand as a rolling platform from the highpeak ridges called the ‘southern wall’ (Cooray 1984: 52), which divides the southern dry zone from the wet uplands. The Valamkattuya mound lies in the transitional area between two ecological zones. The two mountain ranges on either side make a valley type terrain forming favourable environmental conditions for agriculture, especially rice and other drought resistant varieties of cereals such as millet in and around the present site. Two varieties of millet i.e. finger millet (Elucene coracana) and fox-tail millet (Siteria italica) still grow in the area but farmers have been discouraged from cultivating these cereals as these have been eclipsed by various market oriented cash-crops. The valley type terrain gradually becomes a vast hinterland further south of one of the fertile agricultural plains in the country of today. It was the major settlement area in the south and southeastern Sri Lanka, which is historically known as Rohana or Ruhuna (Mv.ix:10). The impact of this topographic setting of the regional environment is manifold. One notable indicator is the annual rainfall budget that (1250-1875 mm) contrasts with the higher annual rainfall regime of the immediate wet zone (1875-2500 mm) and the lower figures of the immediate dry zone (1250-1000 mm). Especially the rainfall variation makes the area an intermediate zone in an ecological sense, between the wet uplands and the southern dry zone. We do not intend here to make a descriptive survey of the environment in relation to the cultural development of the area because the environmental analysis is still in progress.
Immediately to the east of the present site are ruins of an ancient runoff irrigation system, popularly known as rain-fed which suggest that the early farmers occupied the eastern sector of the area at least during the early centuries of the first millennium BCE. This phase of cultural development is akin to phase 1 (900- 650 BCE) of the settlement development of the Lower Kirindi Oya basin that has been secured by a series of 14C dates (Sinclair & Somadeva et a12003; Somadeva 2006). Perhaps this is one piece of strong evidence to show the presence of a sedentary farming group in and around the area at least in the early century of the first millennium BCE. It also contrasts with the absence of evidence of such a run-off irrigation system in the valley area where the present settlement mound and several other similar settlement mounds are situated. Any explanation of this peculiarity needs more data on the environment and a well-stratified cultural sequence of the site.
The Mesolithic habitation of Bellanbandipalassa situated at a distance of about 5 km as the crow flies to the northeast of the present site shows one of the major sites of earliest human occupation in the area (Deraniyagala 1958). The quartz implements found in the natural rock shelters at Bambaragala (Fig. 1) suggest that prehistoric communities had been occupying some places around the present site, but the presence of any open-air site like that of Bellanbandipalassa is still unknown in the vicinity.
The greater part of the Valamkatuyaya mound had been dug-out by treasure seekers. This is the main factor for the disturbance of the cultural deposit at the site and the recent influence that has made a decisive impact to modify the site as well. The dense scattering of artefacts visible on the ground throughout an area of 7 hectares is a result of such transformation and modification activities (Figs. 2 and 3). The arrangement of the artefact clusters and scatters, which are visible at present on the surface, may result in misleading observations unless attention is focused upon site formation processes through time. Apart from the conscious interferences, several other events have also been identified as agents of site formation including cultural and natural transformations, which are presented below.
The soil mixing caused by erosion and the surface runoff could be described under this category of disturbance. The principle aquaturbation agent of the Valamkatuyaya mound is a water channel that flows through the southwestern sector (zone ‘C of the mound. This water channel has been an outlet of the nearby Galpaya tank. It has been formed gradually or accidentally after the period when the tank fell into disuse. The width of the channel differs moderately from point to point. However at the widest point, it is about 5 meters. The calculations suggest that this water channel has brought about the displacement at least 415m3 of the soils of the cultural deposit. The shallow bed of the channel still exposes the ancient potsherds that are embedded in the subsurface of the zone ‘C. The major impact caused by aquaturbation of the discharged water of the tank through this channel is the haploidization, (Buol et al 1973:91) the process that displaced the artefacts horizontally and vertically. The water pressure of the channel had pushed the soil at least a distance of 165 m. across the mound horizontally. The flooding led to a wash off of the banks resulting in soil erosion. This event caused a mix-up of the upper horizons of the sub-surface adjacent to the channel vertically.
The agents of faunalturbation at the site are termites and rats. Single rat species generally known as the kangaroo rat (Tatera indica ceylonica) lives in the site, and is making a decisive impact on soil mixing. Normally tunnels excavated by this rat extend nearly 3 meters horizontally and penetrate into the soil from 1 to 2.5 meters. The soils excavated by the rat accumulate around the mouth of the tunnel forming a low height soil heap (Fig. 4). Sometimes the ancient potsherds visible on the soil heaps are derived rom the sub-surface. During the seasonal ploughing for agriculture, such soil heaps become flattened causing the artefacts to have a horizontal distribution on the surface. Such articulated patches of artefacts resulting from this process, were observed in some places at Valamkatuyaya. The internal tunnelling network formed by this rat was observed in a soil profile, which had been exposed by treasure hunters in zone ‘D’. The tunnel network visible in the soil profile called krotovina structures (Wood & Lee Johnson 1982) had resulted in the distortion of layer and feature interfaces in some places of the deposit. This is mainly due to the infiltration of rainwater through the tunnel and the subsequent leaching into the internal soil layers.
Termite hills are a common feature at the site. They carry fine particles of the sub-soil causing the displacement of finer artefacts such as small beads. The construction of internal burrows occupies more space in the sub-surface while producing a greater degree of artefact displacement. The holes they dig are easy channels for the penetration of air and water into the subsoil thus being the major taphanomic agents of the fragile archaeological materials such as bones and metal objects.
Floralturbation or soil mixing generated by trees at the site is twofold. The first and the most destructive is the effect of tree fall. Such a tree fall in zone ‘D’ shows that an approximate area of 2m radius had been disturbed. A depression of half meter radius had formed as a result (Fig. 5). The artefacts contained in the sub-soil of the damaged area were irregularly dispersed in the surrounding area of the tree fall. The tree fall alters the local contours making a depression called cradle-knoll topography (e.g ..Blancaneaux 1973;Booji 1975). At this point of time cradleknoll topographies appear as manmade features with a considerable density of artefacts resulting from soil mixing. Most of the shallow depressions visible on the Valamkatuyaya mound are the outcome of intentional digging but some of them could be the remnants of the cradle-knoll topography generated by the tree falls of different periods (Fig. 5). The penetration of the tree roots into the soil affects the internal structure of the cultural deposit making a distortion in the soil horizons. This affects the confines of the internal arrangement of the artefacts but more harmful to the cultural materials is the rainwater and the air that infiltrates through root-holes which remain when the tree dies. Especially the large trees in zone ‘D’ are critical in this regard. The exposed soil profile 1 noted in zone ‘D’ exemplifies the degree of damage that could be expected from tree root penetration (Fig. 6).
This term defines soil mixing through soil gases or wind. The soil mixing caused by wind action is not explicitly visible and not easily quantifiable in the dry zone of Sri Lanka. But a closer observation may suggest that wind action results in the displacement of tiny artefacts such as beads. During the months of August and September the southern part of the island experiences a relatively high-speed infrequent wind blow ranging from 40 to 60 kmph. This speed is powerful enough to push the dusty soils on the surface to a greater distance. The effect of the wind upon the movement of lighter artefacts within a shorter distance may cause the distortion of fine arrangements of artefacts in their archaeological context. Longterm wind deposition of the dust is able to hide some artefacts on the surface while obliterating the artefact arrangement patterns. It was noted in the Valamkatuyaya mound that the ashes of burned land plots that were swept through high-speed winds deposited in some places formed an ash layer of several centimetres thickness.
This term is used to denote soil mixing resulting from the swelling and shrinkage of clay (Hole 1961). Dry zone soils in Sri Lanka are subject to a swelling – shrinkage movement seasonally. The presence of the expansible soil clays such as montmorilonite allows the soil to shrink and crack during dry periods. Such cracks get filled with surface soil carried through wind and animal traffic (Fig. 6). When rains start the cracks fill with water which infiltrates through the filling (Fig.7). This results in a distortion of the interfaces of the soil horizons and sometimes leads to artefact dislocation. Generally the hydro-dynamics of the surface runoff during the monsoon rains may push the surface materials to a certain distance. The cracks also provide easy paths for ants and other burrowing insects to penetrate into the sub-soil resulting in various complications that affect the cultural deposit.
Survey methodology is comprised of two aspects. The first is exemplified by the sampling procedure and the second is the data collection and recording methodology, These two are objectively interwoven but are treated separately,
SAMPLING: The stratified sampling technique was followed in sampling. The northern sector of the mound was divided into 4 zones as zone A, B, C and D. A comparatively high density of artefacts and a good surface visibility were available in zones C and D. These zones were again divided into 10 m x 10 m squares called fractions. The intensity of the observational unit was reduced to a 1 x 1 m cell called a ‘unit’. The total area sampled was 1300 m2 being 8.29% .of the entire sample universe of the northern sector. The intensity of the sample size was 1 m2 and it formed 1300 observational units.
RECORDING: The surface of each unit was digitally photographed using a Nikon camera with 8 megapixel resolution from 2m above the ground level. 1300 such photographs were finally joined together to digitally reconstruct the entire surface of the sampled area. The objective of this painstaking task of reconstruction of the surface on a computer screen was to map out the complete distribution patterns of the different artefact classes in order to detect the patterns discernible if any. The photographs of each cell were joined together using computer software called the ‘panoramic sorter’. The full image showing the entire sample area was saved as a jpeg file. Subsequently it was recalled in Arc View 3.1 software and saved as an image theme. Finally the separate artefact classes were digitised and saved as individual themes to produce a holistic view of the distribution pattern of the artefacts on the surface in the sampled area.
DATA COLLECTION: The dusty surface soil down to about 5 to 10 cm of each unit was subjected to screening using a 4 mm metal net. After screening the surface oil, the quantities of the different artefact classes were recorded in weights (potsherds) and counts (other artefacts including stone implements, grind stones, pestles, bricks, tiles, beads and metal objects). The provenance of artefacts in zones A and B was recorded following off-set measurements and were plotted on a scaled map.
SUB-SURFACE: A soil profile showing the sub-surface is exposed at a place in zone D. It was a robber pit. The walls were cleaned and the layer interfaces were drawn (fig. XXX). The exposed parts of the soil profile are incomplete.
PUBLIC INFORMATION: It is said that the Galpaya mound has been continuously destroyed by treasure hunters for at least the past 10 years. Most of the cultural deposits had been removed from the site. ‘This situation urged was to approach the villagers of the neighbourhood in order to collect information about what they had seem and what they had collected from the site. This exercise proved successful and some of the villagers generously helped us by even showing the artefacts derived from the site and now being held at their disposal. A small collection of iron implements, coins (probably late Roman), a terracotta Buddha statue and an inscribed pestle stone (black basalt, probably an import) were recorded through public information supplementing the assemblage of artefacts recovered from the surface of the site.
ANALYSIS Two kinds of analyses were carried out aiming at the objectives of (a) understanding the chronological continuation of the site and (b) understanding the activity areas of the northern sector of the mound. Owing to the disturbances that had occurred on the surface the artefacts had to undergo a process of screening.
ARTEFACT SCREENING: The complexity of the distribution of the artefacts over the mound comprises a broad spatial scattering and different frequencies of accumulation (fig. xxx) (Fig. 8). to overcome the data redundancy in sampling, the following steps were taken to screen the artefacts:
(i) Collected surface artefacts, as many as possible to meet this requirement, the surface soil down to 10 cm was screened on the basis of a 1 m. grid using a 3 mm steel net. (ii) Collected potsherds were recorded according to their weight with the respective grid as the provenance. Tile fragments and brickbats were counted and the provenance was recorded in a similar manner as the potsherds.
(iii) Potsherds were sorted accordirig to their anatomical categories (eg. Rims, necks, body parts, spouts, handles, bases etc.)
(iv) Body parts that bear designing were separated from the main assemblage
ANALYSIS: The majority of the artefacts comprised in the assemblage were the earthenware sherds. Among them nearly 2000 sherds were drawn for a comparative analysis of types. The total number of individual rim forms identified is eleven. These types fall into a timerange extending from 900 BCE to 1400 CE as shown in figure 4.1.
SPATIAL REFLECTIONS: A simple statistical analysis of the scattering of the brickbats and the tiles was conducted to spot the accumulation patterns that would be incorporated in the subsequent interpretations. Figure 4.1 shows a computer generated distribution lattice of ancient brickbats in zone ‘C of the mound. The accumulation of ancient tile pieces shown in Figure 8 to some extent further describes the picture that emanated from the previous analysis of the brickbats.
THE PHASES OF OCCUPATION A relative chronological sequence (RCS) (Somadeva 2006) has been formulated for the occupational history at the Valamkatuyaya mound through a comparative analysis of datable earthenware types sampled from the mound surface. Two databases have been utilized for the comparative analysis, the one established for the urban mound at Akurugoda in Tissamaharama (Weisshar 2001) and the other the chronological framework that had been formulated for the urban origins in the lower Kirindi Oya basin (Somadeva 2006).
RCS suggests a continuity of the pottery using cultural occupation at the mound from 900 BCE to 1400. The noteworthy fact is that surface finds show an occupation maximum in the period between 200 CE and 800 CE. On the contrary, the period from 900 BCE to 200BCE manifests a lesser chronological representation of the types in the surface pottery assemblage. This fact proposes two possibilities that would have influenced the period of the generation of this peculiarity. The first could be the result of a data recovery factory. A high degree of post-occupational disturbances that occurred at the site could have brought about a reduction of the density and composition of different pottery types on the surface. The second could be the evolutionary nature of cultural activities at the site. We do not intend here to take the results of our chronological analysis as a true reflection of the occupation at the mound in a quantitative sense. Besides, the chronological manifestation it provides is vital in a sense, to reconstruct the occupational history of the mound. The presence of stone implements in the surface assemblage is the strongest evidence that confirms the existence of the hunter-gatherers in the area. The occurrence of finely retouched microliths together with grind stones may signify an existence of a new advanced phase of hunter gatherers and perhaps some pottery associated with them would have been contemporaneous with them. The presence of BRW could be considered as a firm indicator of the proto and early historic occupations at the site. From the early historic period onwards the site has been occupied until the end of the fourteenth century CE, as indicated by the comparative ceramic analysis. The decline of the settlements in the area had occurred after the thirteenth century CE as a common phenomenon for the settlement history of the dry zone Sri Lanka (see Somadeva 2006). The graph based on the 14C dates, which reflects the settlement continuum in the Lower Kirindi Oya basin since the early first millennium BCE is reproduced here for comparison.
The artefact palimpsest recovered from the Valamkatuyaya mound has provided us with a ‘snap shot’, reflecting the diversity of the material culture, which consists of the archaeological deposit at the site. This diversity could be encountered from two directions, on the one hand, it is a manifestation of an increasing complexity of behaviour that prevailed at the site through time. On the other hand, the explicit difference of the techno-cultural characteristics (stone-implements querns, pottery, iron, glass beads etc.) could be considered as time indicators of longterm changes that perhaps, would have provided a velocity to that development.
Blancaneaux P 1973. Notes de pedalogie guyanaise. Les Djougoung-Pete due basin-versant experimental de la crique Gregoire. (Sinnamary-Guyane Francaise) (Notes on Guyana pedology. The “Djougoung-Pete” on experimental basin slope of the Gregoire Creek. Sinnamary, French Guyana) Cahiers O-K-S-T-O-M Series Pedologie 11: 29-42.
Booji A.H., 1975. Storm als oorzaak van microrelief in de ondergrond (Storms-causing a microrelief in subsoil.) Boor en Spade 19: 30-31 pp. Buddhadatta, P. Rev. (eds,) 1959 Mahavamsa, Colombo: Gunasena.
Clark R.H., & A.J. Schofield 1991. By experiment and calibration an integral approach to archaeology of the plough soil In Interpreting artefact scatters, contributions to plough zone archaeology. Schofield A.J. (eds.), 93-106 pp. Oxford: Oxbow books.
Cooray P.G. 1984 Introduction to the Geology of Sri Lanka. Colombo: National Museum. Deraniyagala PE.P 1958. An open-air habitation site of Homo sapien balangodensis. Spolia Zeylanica 28(2): 223- 61.
Hole F.D., 1961. A classification of pedoturbation and some other processes and factors of soil formation in relation to isotropism anisotropism. Soil Science 91:375-77.
Kent S., 1987. Understanding the use of space: an ethnoarchaeological approach. In Theory and Methods of the Activiry Area Research. Kent S. et al (ed), 1-63 pp. New York: Columbia University Press.
Lambrick G., 1977. Archaeology and Agriculture: a survey of modern civilization equipment and the problem of assessing plough-damage t o archaeological sites. Council for British Archaeology and Oxfordshire Archaeology unit.
Lewarch D.E., 1979. Effects of tillage. an artefact patterning: a preliminary assessment. In Cannon Reservoir Human Ecology Project: a regional approach to cultural continuing and change. O’Brien M.J. & R.E. Warren (eds), 101-149 pp. University of ebreska, Department of Anthropology, Technical Report No. 79:14.
Roper D 1976. Lateral Displacement of Artefacts due to Ploughing. American Antiquiry 41.372-75 pp.
Schiffer M.B., 1996. Formation process of the archaeological record. Salt Lake City: University of Uttha Press.
-1999. The material life of human beings. London: Routledge. Spaulding A., 1960. The Dimensions of Archaeology. In Essaya in the Science of Culture:In honour of Leslie A. White. Dole G. & R. Carneiro (eds). ew York: Crowell.
Trubowitz N.C., 1978. The persistence of settlement pattern in a cultivated fields. In Essays in memory of Marian E. White. Engelbrecht W & D Grayson (eds), 41-46 pp. Franklin Pierce College, Department of Anthropology Occassional Publications in Northeastern Anthropology 5.
Somadeva R. 2006. Urban Origins in Southern Sri Lanka. Upsala: Uppsala University. Somadeva R. et al 2007. The Galpaya Ancient Settlement Survey: The results of the first field season conducted in a nelwy discovered site in Ratnapura district of Sabaragamuva province of Sri Lanka. Journal of Indian Ocean Archaeology No.3: 1-14.
Somadeva R. et al 2008. The Galpaya Survey: results of the first field season. Occasional paper No.1. Colombo: Postgraduate Institute of Archaeology (.in press)
Weisshaar H.J., H. Roth, W Wijeyapala (eds) 2001. Ancient Ruhuna: Sri Lankan-German Archaeological Project in the Southern Province. Von Zabern: Maiz Am Rhein.
Wood R.W & D. Lee Johnson 1982. A Survey of Disturbance Processes in Archaeological Site Formation. In Advances in Archaeological Method and Theory. Selection for Students from Volume I through 4. Schiffer M.B. (eds.), 539-594 pp.
1. A natural cave shelter with a Brahmi inscription at Bambaragala monastery.
2. The greater part of the Valamkatuyaya mound that had been dug-out by treasure seekers. This is the main factor for the disturbance of the cultural deposit at the site and the recent influence that has made a decisive impact to modify the site as well. The dense scattering of artefacts visible on the ground throughout an area of 7 hectares is a result of such transformation and modification activities (Fig. 2 &
3). The arrangement of the artefact clusters and scatters, which are visible at present on the surface, may result in misleading observations unless attention is focused upon site formation processes through time. Apart from the conscious interferences, several other events have also been identified as agents of site formation including cultural and natural transformations, which are presented below. 3. A view of the Galpaya mound after clearance.
4. A tunnel excavated by a kangaroo rat seen clearly on the soil profile in zone ‘D’.
5 A location of a fallen tree in zone ‘D’.
6.A location which depicts the signs of Argilliturbation in zone ‘D’.
7.An accumulation of ash brought by the wind in zone ‘D’.
8.A terracotta plaque depicting a relief of a Buddha statue in a seated posture.
Raj Somadeva is a Senior Lecturer, Postgraduate Institute of Archaeology, Colombo, Sri Lanka.