Traversing an Archaeological Landscape

Seven years ago, we began what we described as the study of the settlement archaeology and palaeoenvironment of a diagnostic micro-region of the Sri Lankan Dry Zone’. The project is now usually referred to as the SARCP (the Settlement Archaeology Project). Our initial objectives were, at one and the same time, modest and extremely ambitious. On the one hand, we were embarking on an exploratory probe into the little-known archaeological terrain surrounding the major urban center at Sigiriya and the ancient royal monastery complex at Dambulla; on the other, we were undertaking the

investigation of the total archaeological landscape’ of an area.

of about one thousand square kilometers, the scope of our study extending from the prehistory and palaeo-ecology of the area to its contemporary ethnography. The present report - together with its precursor, The Settle­ ment Archaeology of the Sigiriya-Dambulla Region - brings together the preliminary results of that undertaking. As one might have predicted, these results lie somewhere in between the two extremes of our objectives. As the illustration on the opposite page and other plans show, we now have from the Sigiriya-Dambulla region one of the best-documented archaeological landscapes in Sri Lanka. At the same time, we realize that we have only scratched the surface of this landscape and retrieved some fragments of the data it has to offer. Many of the major - and minor - research problems underlying our investigations remain unresolved. However, we have now co­ me to know at first hand the nature of the terrain, its research

Figure 1:1 SARCP archaeological sites and irrigation sites in the Sigiriya-Dambulla region.

The Sigiriya-Dambulla region: between the Anuradhapura and Polonnaruva systems and a gateway to the central mountains

Figure 1:3 The river basins of the Sigiriya-Dambulla region

potential, and the capabilities and limits of our investigative techniques and resources. In place of a relatively blank sheet in our understanding of the settlement network of the Sigiriya- Dambulla area through the various phases of its long history (Bandaranayake 1990), we now have a fairly complex picture of what there is, what there could be and also an acute sense of how little we know. As Mogren observes (see "Objectives, Methods, Constraints and Perspectives" this volume) in his overview of the project, emphasizing one of the laws of archaeological research: "...an interpretation of a site or a pattern, that is published or presented...in some form should always be seen as a mere take-off for further research, never as an ab­ solute fact, or termination of study. It must be under­ stood, therefore, that no matter how many times one returns to a given site, its does not represent a truly complete understanding of a past reality." Our surveys have documented so far a total of 195 archaeological sites and 203 irrigation tanks within the demarcated zone. This forms a complex, diachronic palimpsest, a superimposition, often one upon the other, of the remains of the human occupation of the area through several millennia. The project is contained within an area of about 40km from north to south and 30km from east to west, in the uppermost reaches of the Kiri Oya, Sigiri Oya, Mirisgoni Oya and Dambulu Oya basins. For a variety of reasons, such as the prevailing conditions in 1988 and 1989, the relative inaccessibility of certain sectors such as the northern half of the Kiri Oya basin, the constant presence of wild elephants in the more heavily-forested tracts, human and transport resource constraints and the concentration on other research interests, various sectors of the target area have been surveyed in different degrees of intensity.

The Kiri Oya basin

Thus, research in the Kiri Oya basin was more or less re­ stricted to a brief preliminary survey of its southern half in 1988 (Manatunga 1990); the excavations of the major iron production site at Alakolavava - Dehigaha-ala-kanda (KO. 14) in 1990 and 1991 (see Forenius and Solangaarachchi this volume); and the survey and trial excavations of the Vavala- Sigiriya canal and dam network in 1992 (see Myrdal-Runeb- « jer: "The Archaeology of Irrigated Agriculture" this volume)

Figure 1:4 Prehistoric sites.

and 1993. Rich in spring water resources, limestone quarries and, presumably, in high-quality ironore, the area seems to have formed an important industrial zone and a well-settled communications corridor between the Sigiriya and Dambulla area to the south and west and the Minneriya-Giritale and Parakrama Samudra ’sub-systems’ to the east. Further re­ search in the Kiri Oya basin lies outside the present program of the SARCP, but must remain an important unfinished task to be accomplished at some future date.

The Upper Sigiri Oya basin

The Sigiri Oya originates in the bed of the Sigiri Mahavava, as a small stream flowing northwards out of the breached bund of this once massive, man-made lake. Well beyond the greater Sigiriya area it ultimately joins the northeastward­ flowing Yan Oya, the country’s fifth longest river, and a major settlement zone of the protohistoric and Early and Mid­ dle Historic Periods in Sri Lanka. Limited surveys of the Sigiri Oya basin between 1988 and 1990 have revealed not only a typical tank village and tank cascade landscape (see Wickremesekara 1990a:95-101; 1990b:163-166), but also a series of riverside settlements of EHP-1 which have had to remain uninvestigated. It is only the Talkote complex (Mo- gren this volume), Tammannagala (Somadeva and Kasth- urisinghe this volume) and the prehistoric sites of Aligala and Potana (Adikari: "Approaches to the Prehistory of the Sigiriya-Dambulla Region" this volume; Karunaratne and Ad­ ikari: "Excavations at Aligala" this volume; Adikari: "Excavations at the Sigiri-Potana Cave Complex" this vo­ lume) that have been the subjects of trial excavations. The Sigiriya-Mapagala-Ramakale-Pidurangala complex and the Sigiriya hinterland briefly discussed below, dominate this ba- sin and remain the best-researched section of the entire study area.

Figure 1:5 Protohistoric - Early Historic (EHP 1) sites.

The Mirisgoni Oya and Dambulu Oya basins

The documentation and survey work in the Mirisgoni Oya and Dambulu Oya basins has not been as extensive as in the Sigiriya hinterland, not so much for reasons of inaccessibility or the difficulty of the terrain as in the Kiri Oya area, but rather for those of time and resource allocation. The early work on the study of the irrigation network (Wickremesekera)

could not be pursued, greater emphasis being laid on and resources allocated to the excavations of the important prehistoric site at Potana (which was in danger of being des­ troyed by modern construction work), the substantial settle­ ment site at Ibbankatuva, and the survey of iron production sites in the southern sector of the Mirisgoni Oya basin (Ka- runaratne and Mogren, to be published in Forenius and Sol- angaarachchi (eds.) in preparation). The Potana excavations, briefly described below (Adikari this volume) have produced a wealth of data, especially of faunal remains - nearly 300 kilograms - and human skeletal material from a prehistoric horizon. This rock shelter has also become a test site for our first steps in palaeobotanical studies (by T.R. Prematilake), and the experimental application of chemical analysis to soil as a prelude or complement to excavation (by A. Tantillage).

The Sigiriya Hinterland

The most intensively surveyed area is the immediate hinter­ land of the urban and palace complex at Sigiriya (fig. 1:8), including the great monasteries at Ramakale and Pidurangala (figs. 1:12 and 1:13), each covering an area of about one square kilometre, the irrigation network of the Sigiri Mahavava and the Sigiri-Vavala canal (Myrdal-Runebjer: "The Archaeology of Irrigated Agriculture" this volume; see also fig. 1:8), and the complex of archaeological sites around the ancient settlements at Talkote (Mogren this volume: "The Archaeology of Talkote").

From Prehistory to the Early Historic Period

A much more varied and intricate picture than was hitherto imagined has now emerged. We are now aware of the prehis­ toric occupation of the Sigiriya area, with nearly five prehis­ toric sites available within a 6km radius of Sigiriya itself. A substantial body of data relating to prehistoric habitat, tech­ nology, subsistence and lifeways, and even ritual practices has been gathered and will be the subject of on-going ana­ lysis and research (Adikari this volume: "Approaches to the Prehistory of the Sigiriya-Dambulla region"; Karunaratne and Adikari this volume: "Excavations at Aligala"; Adikari this volume: "Excavations at the Sigiri-Potana Cave Complex"). Aligala, within the eastern sector of the citadel area at Si­ giriya, is a site of primary importance, where we have a rare instance of an immediate succession from the prehistoric thr­ ough the protohistoric to the Early Historic Period.

Figure 1:7 ‘Potana Man’ - skeleton No. 2 from cave 1, context 10, Sigiri-Potana (MO. 14). Three ,4C dates for this context have a 5000 - 4500 BC range. There is no indication whether the skeleton is intrusive. Photo: CCF - Sigiriya Project.

Figure 1:8 Sigiriya Hinterland

Figure 1:9 The Aligala-Potana sequence.

This site became doubly important when the protohistoric settlement horizon here provided one of the earliest radio- metric dates for an Iron Age site in the country. The Aligala sequence is complemented by that from the rescue excavations at Potana.

Irrigation

Another important focus of research (see Myrdal-Runebjer: "The Archaeology of Irrigated Agriculture" this volume) has been the hitherto little-understood canal system lying to the south and east of the Sigiri Mahavava, the great relict man­ made lake, whose earth dam has now been found to have extended for a distance of 7km from the southern foot of the Sigiriya rock and the Mapagala fortress to Nikavatavana and beyond (Adikari 1993). This canal system seems to have its origins as far away as Vavala, 9.6km to the south of Sigiriya. It forms, therefore, an early example of a trans-basin canal, whose complex technology, function, chronology and man­ power inputs are continuing subjects of investigation. The Sigiri Mahavava seems to have been supplemented at its southernmost point by another canal which seems to have connected the Kuda Gona Oya to the great lake from a point 11km away.

The urban form

Equally significant has been the discovery of the full extent of the urban area of Sigiriya, now appearing as a walled city nearly 3km in length, with a series of suburban settlements (figs. 1:10 and 1:11; see also Karunaratne: "The Eastern Pr­ ecinct of the Sigiriya Complex" this volume). Judging by the pottery from these sites and the 14C dates from the large suburban ’village’ of Tammannagala (see Somadeva and Ka- sthurisinghe this volume), these settlements seem to date fr­ om the major urban period in the late 5th century AD, but at least some of them continued - as the city itself seems to have done - for some centuries after its abandonment as a royal capital. Sigiriya is one of the best-preserved urban forms of the first millennium in Asia, exemplifying the grand scale and vision of Sri Lankan city planning. We can now add to the discussion of this in the previous volume (Bandaranayake 19- 90.24-26). The city can now be seen as having an elongated rectangular form, nearly 3km from east to west and 1km from north to south, based on a square module of circa 170m and a concentric-cum-axial design. This gemometrical conception also accomodates organic and assymmetrical elements such as the Sigiriya rock itself, the terraced hill around the base of the rock, the Mapagala fortress, the partly man-made, partly natural topography of the lake, and the ’borrowed scenery’ of the

Figure 1:10 Suburban settlements of the 5th-7th century. The black dots indicate the existence of BRW utility EHP 1 settlements.

Figure 1:11 Schematic Plan of the Urban Center

Kandalama mountains and the Matale ranges.

The urban plan presents itself in at least five, hierarchically organized zonal divisions:

Level 5: the palace on the summit;

Level 4: the inner citadel or raj angaria',

Level 3: the ’inner city’ or ceremonial precinct to the east,

and the royal pleasure gardens to the west;

Level 2: the outer city and the Mapagala fortress;

Level 1: the suburbs and villages beyond the outer city

ramparts and the lake, the Sigiri Mahavava.

Buddhist monasteries Immediately contiguous to this urban landscape and extending beyond it are the remains of a large number of Buddhist monasteries belonging to various categories and from different periods of history. These monastery sites form the most visible and expressive aspects of premodem settlement in the Sigiriya-Dambulla region, as almost everywhere else in the Dry Zone. They occupy a dramatic and central position in an elaborate mosaic of multi-period village settlements, irriga­tion features, iron production centres, cemeteries, and fortifications. The monasteries were clearly principal foci of social organization in the countryside. Their role as centres of religion, as well as their socio-political and socioeconomic functions are well known and have been extensively researched but as yet mostly from literary and epigraphical sources.

Figure 1:12 Ramakale

Figure 1:13 pidurangala.

The ’pre-Kasyapa’ and ’post-Kasyapa’ history of Sigiriya, first as a rock-shelter monastery of the earliest phase of the Early Historical Period (EHP 1 - 3rd to 1st century BC) and then the establishment and development of the post-Kasyapan monastery of the Middle Historical Period (6th to 13th century AD) is complemented by the continuation and expansion of the grand monastery complexes at Pidurangala and Ramakale. Our surveys have also indicated a number of equally important and other less substantial monastic centres, many of which were previously unknown. These monastic sites are set amidst a large number of multi-period village settlements, the best-surveyed and most complex example being that of the Talkote complex described at some length below (Mogren thisvolume). Detailed investigations of selected monastic complexes, including field surveys and test excavations have been completed and will form the subject of a separate volume (in preparation - by Prishantha Gunawardena and the present au­ thor) and more than one dissertation. Particular attention has been focussed on the network of rock-shelter monasteries of the earliest phase of the Historical Period, which form an important component of the monastery sites in the area, and which were discussed in a preliminary way in the earlier report (Bandaranayake 1990:22-23). A general survey of the area indicates a clear hierarchy of sites and site catchment systems. In particular, excavations and detailed surveys have been carried out in rock-shelters at Sigiriya, Pidurangala and Dambulla. The underlying research objectives were:

Figure 1:15 Early Buddhist rock shelter complexes.

(a) the depositional and habitational sequence;

(b) the study of rock-shelter architecture;

(c) the extent and use of space within a rock shelter and its maximum number of occupants in a cave;

(d) the relationship between the cave itself and the sur­ rounding area; (e) the population parameters of monastic settlement. The analysis of the results of these studies is still on-going and will be completed and published in the forthcoming monograph.

Figure 1:14 Buddhist monasteries of the EHP 1 and Early Middle Historic Period.

Figure 1:16 Development of monumental structures.

Thematic studies

Period-based and thematic interests - such as prehistory, the early settlement period, iron production, irrigation systems, central place studies (particularly of the monastic centres in the area), the ethnoarchaeology of food production - have replaced the general field surveys of the first two seasons of the project. The major research problems that are being addressed include:

1. The descriptive reconstruction of the human settle­ ment in the area from prehistoric times onwards and an understanding of the major changes and transformations that took place;

2. The indications if any in the study area, of the beginnings of agriculture and settlement and of the development of major rural and urban settlements in the Protohistoric Period and in Phase I of the Early Historic Period.

3. The character and function of the monastery as a central place, its population, development history and material and other evidence of the exact nature of its relationship with the society around it.

4.The urban form at Sigiriya in the wider context of urbanization in Sri Lanka. *

5. The major changes that took place and the charac­ ter, chronology and causative factors of the gra­ dually decline of central places and of population in the period after about the 13th century.

6. The role played by iron production in the economy of the region and the chronology of this develop­ ment.

7. The character, chronology and technology of va­ rious types of irrigation systems and the manpower inputs involved in the construction of irrigation w- orks.

8. The usefulness of specific contemporary ethnog­ raphic studies in the interpretation of archaeologi cal remains, especially the ethno-archaeology of food production.

The archaeological scope of such questions, and the inves­ tigative resources required in addressing them, in the context of the study area itself and in terms of the level of development of Sri Lankan archaeology, defines the intellectual and institu­ tional horizons of the project. The present volume reflects the multi-disciplinary approach and the broad thematic and chronological range that has been achieved in SARCP. It dis­ plays the new perspectives and insights the project has brought to the study of this country’s rich and multi-faceted history, undertaken principally by a new generation of Sri Lankan archaeologists, with the collaboration and assistance of our Swedish colleagues.

Future Perspectives

The program continues to maintain, in a number of different ways, the momentum generated during the last seven years. Several research dissertations based on or deriving from SARCP fieldwork are in progress. They are scheduled to be completed in 1994-95 and to be published as project monographs. These descriptive and analytical studies will form the project’s most substantial research product. Theoretical syntheses relating the composite and diachronic data generated to broader patterns and long range processes of historical development, on a national, regional and global scale, are also being attempted - without losing the close focus on micro and macro-level fieldwork. In this con­ text, comparative studies in irrigation, palaeo-ecology, ur­ banization, iron production, and the social history of fortifications are underway, at least in an incipient or program­ matic form, and constitute some of the important areas of ongoing and future SARCP research, considerably expanding the project’s intellectual terrain. The potential for such re­ search in two areas of study was assessed recently in a field visit (led by the palaeo-ecologist, Prof. Urwe Miller, of the University of Stockholm) and in a paper read before the Sri Lanka Historical Association (Bandaranayake 1993). The lat­ ter discussed the possibilities of using data concerning irriga­ tion works and monumental remains, on a local or national scale, as useful indicators of cycles of economic development, which could at the same time be related to long-range global processes. The type of projection that might be obtained from the study of monumental remains in the study area were presented in the form of a diagram (Fig.1.16), generated from the combined and partly "guesstimated’ evaluation of three factors - (a) site area; (b) estimated quantities of material and/or labour input; (c) level of technology. In all the technical complexities of SARCP research the human dimensions of archaeology have not been forgotten. While settlement and landscape studies, like all other branches of the discipline, are primarily concerned with the tech­ nicalities involved in the retrieval and analysis of material residues, as well as wider interpretative and theoretical con­ cerns, they are ultimately concerned with people - in the past and in the present. The primary objectives of the program, such as the ’archaeology of the village’, of the neglected base of the social pyramid, the development of national research capabilities and the collegial relationship between Sri Lankan and Swedish archaeologists, have remained in the forefront of day-to-day work. We may use as an elegant and symbolic expression of this concern Maya Upananda’s evocative photograph of a terracotta souvenir sculpture from Sigiriya (Fig. 1.17) in which he has captured for us an intensely per­ sonal glance from the past. Fashioned out of clay, the simplest and most basic of materials taken from the homeground of both the farmer and the archaeologist, this figure expresses the imaginative and individualized sensibilities of an ancient society so profoundly concerned with its art that it wrote poetry about it - the Sigiri graffiti (Priyanka, this volume) - and produced sculptured souvenirs of the well-known Sigiriya paintings. This image reminds us that archaeology is as much about ourselves as it is about trying to understand the ’objective’ societal realities of the past.­

Figure 1:17 Terracotta souvenir sculpture from the Boulder Garden at Sigiriya. 7th-9th centuries AD. These figurines depict the apsaras, the celestrial nymphs in the Sigiriya paintings. They were probably made as souvenirs to be taken away by visitors to the city who came to see the abandoned rock, its palace and the paintings. Photo: Maya Upananda.

REFERENCES

Adikari, G. Preliminary report on ancient irrigation. Study of the Sigiri Mahavava. PGIAR Ar chive. Cat.No. 93/14.

Bandaranayake, S., M. Mogren and S. Epitawatte (eds.). 19- 90. The Settlement Archaeology of the Sigiriya-Dambulla Region. Colombo: PGIAR.

Bandaranayake, S. 1990. Approaches to the Settlement Ar­ chaeology of the Sigiriya-Dambulla Region. In Ban­ daranayake, etaZ. 1990:14-38.

Bandaranayake, S. 1993. Sri Lanka in the context of long range patterns of development in world history a note. Paper presented to Sri Lanka Historical Association Re­ search SeminarColombo.

Forenius, S. and R. Solangaarachchi (eds.). (in preparation). Approaches to the Ancient Iron Production of the Sigiriya-Dambulla Region. Colombo: PGIAR.

Karunaratne, P. and M. Mogren. The Spatial Aspect of Iron Production in the Sigiriya-Dambulla Region. In Forenius, S. and R. Solangaarachchi (eds.) (in preparation).

Prematilake, T.R. Palaeobotanical studies in the Sigiriya Region. Paper presented at the seminar on the Archaeology °f Sigiriya. CCF - Sigiriya Project. February 1994.

Thanthilage, A. Report on phosphate analysis for tracing set­ tlement patterns in the Sigiri-Potana cave complex. PGIAR Archives. Cat. No. 1992.

Wickremesekera, C. 1990a. The Sigiri Oya and Misrfsgoni Oya basins. In Bandaranayake, S., et al. 1990.94-104.

Wickremesekera, C. 1990b. A catalogue of villages. In Ban­ daranayake, et a/.1990:159-163.