19th Ave New York, NY 95822, USA

Colonization of Time


Colonization of Time: some problem of identifying archaeological sites in modified historic landscapes, a case study in Sri Lanka


Identification and interpretation of archaeological sites might have embodied three major problems. First exemplifies by the understanding of the functional aspect of sites. Some single sites reflect different functions of different periods. Some sites are multi-functional and belonged to a single period of time. Determination of function of an archaeological site for identification is a complicated exercise and most of the times it remains incomplete and theoretically malsounded. Second problem signifies by the anomalies confronted by the site formation process. A site can be transformed through number of human activities. Notable among them in agricultural societies are ploughing and tilling. Long-term ploughing and tilling may cause a considerable effect to the archaeological content of a given site. Most crucial is the complete disappearance or displacement of the surface artifacts. Third problem is the degree of influence by the archaeologist’s idiosyncrasy on his definition about what is an archaeological site.

  1. Introduction

The history of the archaeological site surveys in Sri Lanka extends to the late nineteenth century (cf. Parker 1884; Nevill 1886; Hocart 1924). The early works are mere attempts of descriptive registration of archaeologically potential locations of different parts of the country especially in the north central and southern and southeastern Sri Lanka. A considerably long time has elapsed until 1970s to refocus the archaeological concern in the country to question the space related aspects of the ancient human behavior within a broader spatial framework. Within that temporal hiatus of nearly a century, the archaeology of the country was lagged behind with a bearing of ‘monument based’ research interest. This monothetic intellectual framework should be understood within the complex socio-political atmosphere of the mid twentieth century Sri Lanka. The contemporary held colonial and post-colonial resistance became its zenith during the mid twentieth century, which country’s history and archaeology played a significant role within it in an implicit manner (cf. Harischandra xxxx). Persuasion to study ‘privileged relics’ (cf. Bapty 2002?) of the past in the country was one of a passive motivations of that time, which was influential to strengthen the nationalistic ideology. Nationalism was a global tendency in the development of archaeology during the nineteenth century. I am not intending here to make an assessment upon this unidirectional development of the subject in Sri Lanka but worthwhile to quote Gellner’s proverbial phrase to generalize the possible theoretical consequences of this trend. He states; ‘nationalist ideology suffered from pervasive false consciousness. It claimed to protect an old folk society while in fact helping to build up an anonymous mass society’ (1983:124 emphasis added).

After 1970s, several independent research projects based on regional scale spatial studies appeared (Solheim & Deraniyagala 1972; Ragupathy 1987; Dissanayake xxxx; Vidanapatirana xxxx; Rambukwella xxxxx; Senanayake; Bandaranayake et al 1992; 1994; Somadeva 2003). Implementation of regional scale archaeological site surveys aiming at a better understanding of regional archaeological landscapes was encountered. In all respect, it was a new experience in archaeological research of the country, which transformed the traditional concern to a new conceptual ground. Beside this novelty of the experience, apart from very few studies, the rest show a necessity of having more work on their theoretical and methodological foundations. The principle methodological criterion of most of the major work up to date on the archaeological site surveys in Sri Lanka was perhaps a linear apprehension of surface scatters of artifacts in observed locations (cf. Manatunga 1990; Mogren 1990; Wickramasekara 1990). For me, it seems that such a linearity of the view is derived from the conviction of the observers about the possible direct relationship between artifact formation and their find-spots. This approach is theoretically polemical and subjected to a wider criticism in recent discussions in field archaeology (Schofield 1991).

  1. The structure
  2. What we see?: the polemics

Generally the identification of places of archaeological significance is based on one’s observations of artifact scatters in a given location unless the ruins of above ground monuments are not present. In some of the previous works, the common logic of the observations that carry forward to establish the identity of ‘a site’ is the assessment made 3 on the basis of quantitative or qualitative characteristics of artifact assemblages. The recent global research literature has pointed out the fact that the existence of artifacts on the surface along is not the only valid criterion to identify an archaeological site and has emphasized the theoretical limitations of the quantitative assessments of the surface artifact assamblages (cf. Wilkinson 1982; Simms 1988; Oake & Shennan 1989) .

The main focus of the criticism on the determination of a site upon the direct relationship between artifacts and space relationship grounded on the morphological characteristics of the artifact arrangements on the surface. Beyond the issues related to the site formation process (Schiffer 1987), no adequate methodological framework has been developed to trace the behavioral complexities that resulted the formation of material culture in the archaeological records. Therefore to reach a behavioral interpretation from the artifact scatters might be extremely fractious effort due to their fragmentary, in proportionate and taphanomically transformed nature.

The common generalization of our field observations is grounded on two polar ends of the quantative representation of the artifacts i.e (a) the high density of artifacts as a firm reflection of an existence of a site and (b) the availability of less artifacts as off-site accumulations or ‘background noises’ (Schofield 1991: 4) that has no direct spatial relevance to an archaeological context. Three specific terminologies were used in two previous major works in Sri Lanka to denote the high-density occurrences of surface artifacts i.e ‘major pottery sites’, ‘presence of fair quantity of potsherds’ (cf. Manatunaga 1990) and ‘unusual concentration of potsherds’ (Wickremasekara 1990). Use of these terminologies might show us a partial view of perceiving ‘an archaeological site’ within a narrow confinement of artifact pattern recognition. There is no reference available so far in the archaeological literature in the country to show that the low-density occurrences have been subjected to the identification as the places of archaeological significance.

Other than the indication of a human habitation, a high-density occurrence of surface artifacts can be resulted by a number of taphanomical processes. Sometimes such an occurrence can be an off-site accumulation or a background noise. Several ethnographic studies shows that material discard tend to occur more frequently away from habitation areas (cf. Gould 1967; Binford 1978). In such case, an attempt is needed to distinguish between off-site discarded assemblages and the in-site assemblages. Internal pattern recognition of the artifact organization within a site might help to solve this problem (cf. Binford 1983). The result of the trend surface analysis of the scatter of ancient potshreds at the habitation mound in Kirindagodana in the lower Kirindi Oya basin has showed the archaeological potentials of having pattern recognition in a surface assemblage. The surface artifacts scattered over 1250m2 at the site had been recorded following a 1x 1 meter grid. The distribution of the artifact counts in each grid was simulated as a contour map (In this case the x and y were the distance from the main reference point of the excavation and z value represented by the number of sherds in each grid) (fig. 1). The excavation at the site suggested that the area of high concentration of surface artifacts was the nucleus section of activities of the habitation. 14C dates pointed out that this site was occupied in the 350 BCE (cf. Somadeva 2005).

High-density occurrences of artifacts, which are secondary in deposition, posit another problem of inferring their in situ origin. According to my experience in Kirindi Oya survey, the only possible way of doing that without damaging the internal deposit is to study the history of the recent land use patterns of the area under survey. It may provide the information about various episodes of population mobilization and that information would be helpful to infer the rate of landscape modification occurred in the area. But it is difficult to obtain reliable information in this regard unless the relevant details are recorded in a legitimate manner (eg. regional land survey, regional administrative authority etc.).

Development activities are also a crucial factor of changing the appearance of landscapes in dry zone Sri Lanka causing unrecoverable damages to the surface artifact patterning especial during the new road constructions and canal building. In 1999, I observed a considerable amount of Black and Red ware sherds scattered along the new gravel road lying towards Ikkapallama from Debaravava in Tissamaharama. The density of the scatter was approximately 4 to 6 shreds per square meter. This is a significant rate of concern of identifying a site in the dry zone Sri Lanka if the deposition was not affected by the post-disturbances. Further observations confirm that those shreds came with the soil transported from a place about 15km away from the location for the purpose of the construction of the road. Number of such cases has repeatedly suggest that the presence of a high-density accumulation of surface artifacts along is not a sufficient indicator to confirm an existence of a ‘site’ with a greater degree of precision.

The second point in the previous works that I questioned here is the unspecified use of two notions i.e (a) settlement and (b) cluste.

  1. Sigiriya & Kirindi Oya surveys

Sigiriya is a village lies in the midst of the vast hinterland of the dry zone in the north central province in Sri Lanka (see Map 1). This village had this name after the fifth century city complex and the rock fortress of that name situated in the vicinity (Bandaranayake xxxx). Sigiriya hinterland is moist by several seasonal water streams. The fertile reddish brown soil and the micro distribution of its sub types have profoundly influenced to keep the earth in a suitable state for agriculture. Historical chronicles mention names of several drought resistant cereals such as finger millet (Elucene coracana) cultivated in the land plots of the dry zone (Siriweera xxxx). Wet rice (Oriza sativa sp.) cultivation is successful in this region only under the use of artificial irrigation methods. Archaeological research suggest that this hinterland was occupied by the prehistoric hunters gatherers fishermen communities at least 14 000 years ago (Adikari xxx). Some of the excavation results show that the transition from prehistory to early farming in this area was not a gradual transformation but perhaps a ‘cultural leap’ as similar to that of in the other parts of the island. However the point of departure of this process is still uncertain (Deraniyagala; Caswell). The continuity of the agro-pastoral settlements in the Sigirya hinterland extends up to the present and perhaps some short-term discontinuities and disequilibria can also be inferred from the archaeological evidence (cf. Bandaranayake xxx; Ranawella xxx).

A pioneering study of ancient settlements in Sigiriya hinterland was carried out in early 1990s (Bandaranayake et al 1990; 1994). It has unveiled the spatial extent of the archaeological landscape of the area and made an overview of the potentials of future research on settlement archaeology of the region and marks a turning point of the conceptual and methodological development of the archaeological research in Sri Lanka as well.

Lower Kirindi Oya basin survey was carried out by the author and the team of the Postgraduate Institute Archaeology for four years since 1999 aiming at a better understanding of the ancient settlement development in the area that lead to the urbanization in the late first millennium BCE. 300km2 around the Tissamaharama town has been surveyed following the field walking method. 127 locations of archaeological significance have been reported (Sinclair & Somadeva et al 2004; Somadeva 2005).

A new research project has been launched in 2004 in an area covering 5km radius from the Sigiriya rock (Map 1). The objective of this project was to fix a time series database for the previously identified sites in the landscape vicinity to the fifth century royal mansion at Sigiriya. It was hypothesized that this time series database would provide an evolutionary perspective to the settlements that has been identified during the previous survey. It is also inferred that such a perspective would enable archaeologist to simulate the continuity and changes of the cultural development in the Sigiriya hinterland and it would offer a basis to understand the fact that, in a long-term perspective, the dynamism of socio-economic dynamism of the rise of an urban agglomeration in Sigiriya around fifth century CE (Bandaranayake xxxx).

In several occasions of the year 2004, approximately 32km2 has been resurveyed by field walking. The resolution of the field observation was increased in to 10×10 meters grid. The objective of vesting much time on the re-survey was to; (a). record the GPS locations of the gravity centers of the sites (b). measure the approximate extent of the artifact scatterings of each site and (c). collect a site representative sample of surface finds.

76 localities identified as ‘ancient settlements’ during the previous survey (Mogran 1991) have re-visited (table 1). A relative chronological sequence has formulated on the basis of the potsherds in the sample through a comparison with the dated types of other sites in Sri Lanka ( for this methodology see, Somadeva 2005 forth coming). The extent of the artifact scattering and the degree of diversity of pottery types and other artifacts exist in the observed location were considered as the archaeological reflectors of a site.

The major observation captured during the re-survey was that the inspected locations are not homogeneous both time and function. Beyond this complexity, it was also witnessed that number of factors have been affected the formation of sites in that area and more influential is the periodic agricultural activities. Forest clearance, burning and tilling are crucial to the displacement of surface artifacts. This complexity of site formation suggests a necessity of a high-resolution field methodology to identify the gravity centers (probable in situ locations) of the sites. Moreover one needs a working hypothesis to contextualize his or her field observations. More useful in this regard is to divorce from rigid cultural generalizations and turn into a consideration of the stimulus of human agency upon the perceptions of ‘the landscape’.

The parameters of fixing a scrutinized field methodology might have require knowledge of at least four factors including (a) the physical characteristics of the terrain (slope gradient, soil friability etc.); (b) nature of the regional taphanomic processes and the degree of their influence to the landscape; (c) recent history of the landscape and (d)

Plate. xxx A temporary hut constructed in a cultivation land at Sigiriya: the material, space and design reflects the emphasis of human agency within the wider landscape (Photo: author)

degree of the surface visibility. These four are, in certain extent, interrelated at some points but their influence to the identification of archaeological sites could be discussed separately.

In the dry zone Sri Lanka, the dominant topographic feature is the undulating terrain. This terrain type consists of natural high grounds, which appear as small mounds. Such were major attractions of human occupation since late Iron Age in Sri Lanka. My Lower Kirindi Oya basin (LKB) survey in the Hambantota district has explicitly shown the credibility of this phenomenon. All sites recovered within the limits of floodplain of Kirindi Oya are on natural mounds. Such features are naturally protective against the floods and watchful places of surrounding landscape. Both in Sigiriya hinterland and the LKB, the earliest agro-pastoral settlements scattered either in the outer floodplain or in the low-height hill terraces (Somadeva 2005 forth coming). The migration into the floodplains was stimulated by a number of factors including population growth and the acquisition of technological advances especially in flood control and tilling deep alluvial soils. It seems that the floodplain occupation was a later development in Sri Lanka. LKB survey suggests that the early agro-pastoral settlements emerged in the hilly sector and the outer floodplain in late 1000 BCE or early 900 BCE (Somadeva 2005). According to the 14C dates assigned to the lowest levels of the urban mound at Akurugoda, the earliest dates of the floodplain occupation could not extend beyond 450 BCE (Weisshaar 2001). Recognition of this contrast of the landscape preference of the early settlers might beacon the archaeologist to infer about archaeologically potential areas of a given landscape. The archaeological use of the understanding the viable landscape facets of human activities in a given area is it helps to distinguish between the off-site accumulation of artifacts and the original locations of the ancient human activities.

My experience in the Sigiriya re-survey and the lower Kirindi Oya survey was the regional taphanomic processes are sometimes associated with the physical characteristics of the terrain. More pronounced natural taphanomic element in the dry zone is the soil erosion. Tilling the soil for cultivation begins immediately before the monsoon rains and sometimes the surface runoff of rainwater causes destructive erosion in the mound slopes. The surface artifact displacement is invariably associated with this process and the result is a new pattern of artifact arrangement at the site itself or an emergence of an off-site accumulation. If our identification of an archaeological site constructed on a quantitative.

basis, less consideration on the regional taphanomic processes would make false observation that direct us to some misleading interpretations.

The human activities that may alter the surface artifacts of a given location are a crucial determinant. More effective is the modern day re-occupation. Most of the locations of the modern settlements in the LKB have a close spatial proximity to the ancient sites or if not in sometimes were re-occupied in the same locations. The influence of the modern settlers to re-organize the surface patterns of artifact distribution cannot be listed with a

Fig.xxx Exposed earth during the land preparation for cultivation in the mound slops in the dry zone has a threat of accelerated soil erosion caused by the high dynamic surface run-off results by the monsoon rains.

great precision. It depends on their views that they carry upon the remnants of the past. My interviews with the present villages in the LKB and Sigiriya hinterland suggest that most of them have a neutral attitude towards the surface artifacts that they observe within their home gardens and cultivation lands. They think that such belonged to a remote period but have no value to the present. This indistinct attitude persuades them to act upon the archaeologically potential places in a careless manner. Some people tend to wipe out such evidence consciously due to fear of the legal provisions imposed on the archaeological sites in the Antiquity ordinance. I have made a very rear observation during the fieldwork in Huruluvava area in the north central dry zone in the year 2005 about the public perception of the artifacts. In that area, villagers believe that the ancient potsherds carry a magical or medicinal power and use such when one had a serpent bite. They rub potsherds with water and apply on the surface of the wound. I have no confidence to believe whether they have positive consequences of such a treatment but the conscious collection of artifacts under similar circumstances may also cause a considerable displacement of surface scatters.

Understanding ancient land use patterns would also useful to a great extent to identify the ancient sites through field observations. Theoretically this would be polemical but the Sigiriya resurvey shows that there is a possibility to simulate the ancient land use patterns

through a comparative analysis of the site content, especially the surface artifacts. According to this method, different types of rim shreds sampled from 6 adjacent locations have been analyzed. The results show a considerably high variability in two sites in comparison with the other locations. It was also observed that the extent of the artifact scatter of these two locations is greater than that of other four locations. The other peculiarity observed was the two locations that have a high variability of rim shared are associated with two small man made reservoirs. Other locations are visible along a small seasonal water stream. A clue is derived from the ethnographic observation I made about the present day land use pattern of the village farmers in the Sigiriya hinterland. Some of them are still living close to the ancient reservoirs. They cultivate cleared land plots in the nearby forests. During the rainy seasons and immediately aftermath they moved to the lands close to the adjacent water stream for cultivation. Temporary wooden huts construct within the cleared areas to get shelter during the daytime activities in the cultivation lands. This movement results not only a modification of the landscape but also add material debris to the surroundings. In late October 2004, I observed a temporary hut in the Sigiriya hinterland to inspect the volume and the diversity of the durable materials that such a short-term temporary habitation contains. The observation results are as follows.

  1. a clay pot to keep drinking water
  2. a small clay vessel uses to boil water
  3. a clay vessel uses to boil rice
  4. a clay vessel uses to cook a curry
  5. two aluminum cups and two aluminum plates

is the theoretical preconceptions of the human spatial behavior would not be always helpful to contextualize the field observations. The human choice of the selection of a location in a given space might have guided by different factors. Beyond that, within a single factor, different cultures may act in distinct forms according to their cultural norms and values.


Butzer, K. W. 1982. Archaeology as Human Ecology Methods and theory for a contextual approach. Cambridge: Cambridge University Press.

Chang, K. C. 1967a. Rethinking Archaeology. New York: Random House.

– 1967b. Major Aspects of the Interrelationship of Archaeology and Ethnology. In Current Anthropology 8:227-43. Binford, L. 1967. Reply to K.C. Chang’s “ Major Aspects of the Interrelationship of Archaeology and Ethnology”. In Current Anthropology 8: 234.

– 1968. Review of K.C. Chang’s Rethinking Archaeology. In Ethnohistory 15: 422-426.

Schiffer, M. B. 1987. Formation processes of the Archaeological Record. Albuquerque: University of New Mexico Press.

Stein, J. K. 1976. Deposits for archaeologists. In Advances in Archaeological Method and Theory 11: 337-395.

Triggre, B.

Rossignol, J. 1992. Concepts, methods, and theory building: A landscape approach. In Space, Time, and Archaeological Landscapes, Rossignol, J. & L. Wandsnider (eds), 3-20. New York and London: Plenum Press.

Dunell, R.

Tilley, C. Y. 1994. A phenomenology of landscape: Places, Paths and Monuments. Oxford: Berg.

MacCord, H., 1988. Where do you draw the line?. In American Society for Conservation Archaeology Report 15 (1): 6-15.

Doelle, W. 1975. Prehistoric resource exploitation within the Conoco Florence Project. In Archaeological Series no. 62. Tempe: Arizona State Museum.

– 1977. A multiple survey strategy for cultural resource management studies. In Conservation Archaeology: A guide for Cultural Resource Management Studies. Schiffer, M. B. & G. J. Gumerman (eds), 201-209. New York: Academic Press.

Plog, S., F. Plog. & W. Wait. 1978. Decision making in modern surveys. In Advances in Archaeological Method and Theory 1: 383-421.

Gellner, E. 1983. Nations and Nationalism. Ithaca: Cornell University Press.

Fox, C. 1922. The Archaeology of the Cambridge Region. Cambridge: Cambridge University Press

Related Posts

error: Content is protected !!