Mobility strategies and site structure: a case study of Inamgaon

Mobility strategies and site structure: a case study of Inamgaon

Journal of Anthropological Archaeology 22 (2003) 105–125 Mobility strategies and site structure: a case study of Inamgaon...

747KB Sizes 0 Downloads 4 Views

Journal of Anthropological Archaeology 22 (2003) 105–125

Mobility strategies and site structure: a case study of Inamgaon Sheena Panja Department of Ancient History and Archaeology, Visva Bharati, Santiniketan, India Received 12 December 1997; revised 1 September 2002

Abstract Mobility change and its archaeological indicators is one aspect that has been dealt with extensively in archaeological literature. Archaeologically, movement change has been understand at two levels, at the level of the region and at the level of a single site. This present work is a methodological exercise which seeks to understand mobility change from the archaeological record of a single site, ie, Inamgaon by reanalysing earlier arguments and bringing some new methods in this respect using actualistic research as a heuristic device to understand archaeological patterning. Criteria used by researchers to understand mobility change like house shape, form, change in organic remains, maintenance behaviour, spatial patterning have been critically assessed and used in this respect. The Chalcolithic site of Inamgaon situated in the western part of India has been taken up as a case study for this purpose. Ó 2003 Elsevier Science (USA). All rights reserved.

Introduction Mobility change and its archaeological indicators is one aspect that has been dealt with extensively in archaeological literature. Mobility is not merely variable but multidimensional and archaeologists have to recognise these various forms of movement as the way people move influences their culture and society (Kelly, 1992). Numerous archaeological works have sought to deal with the problem of identifying mobility change from the ‘‘static’’ archaeological record. Researchers studying past material culture have used multiple archaeological criterion to distinguish between types of mobility as well as employed models derived from ethnographic or actualistic work to give meaning to archaeological patterning (Binford, 1987; Kent, 1984). BinfordÕs seminal work sought to develop conceptual tools to understand the differences between mobility/subsistence patterns and their roles in creating diverse forms of archaeological remains (Binford, 1983). Other works have also tried to understand movement and its material consequences

E-mail address: [email protected]

(for e.g., Cribb, 1991) through a detailed study of a particular group. Archaeologically, movement change has been understood at two levels, at the level of the region and at the level of a single site. A change in movement strategies would inevitably be reflected in distribution of settlements across the environment (Rafferty, 1985) as settlement pattern change at the regional level is by far the most important and most decisive indicator that is visible to the archaeologist. However apart from a region, archaeologists have used a variety of traits to provide evidence of movement change from a single site. One of the frequently discussed traits is site size; a sedentary settlement is supposed to be larger than a nonsedentary one. Another aspect which is discussed is house form. It is assumed that sedentary people would build more substantial houses than people who are nonsedentary. Sometimes house shape is taken as a criterion to distinguish between sedentary and nomadic groups, i.e., rectangular houses with sedentary groups and circular with nonsedentary groups. Storage as well as increased variety of artefacts are also taken to represent sedentism. Flora and fauna are often also used to indicate nonsedentariness in areas where seasonal indicators are

0278-4165/03/$ - see front matter Ó 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0278-4165(03)00005-9


S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

helpful in determining when a site was occupied. Another avenue of research into mobility lies through one aspect of site structure analysis, the spatial distribution of debris on the site taking into account formation processes (Binford, 1990; Kelly, 1992; Rafferty, 1985; Wilk, 1983). This article is a continuation of this methodological exercise which seeks to recognise and understand movement strategies from past material culture of a single site using actualistic research as a heuristic device to understand archaeological patterning evident in the archaeological record. The Chalcolithic site of Inamgaon situated in the western part of India (Fig. 1) has been taken as a case study for this purpose. Mobility can be understood at two levels, the region and the site. Research on Inamgaon was earlier carried out to identify mobility change across a region through archaeological and actualistic perspectives (Panja, 1999). This present work is a methodological exercise which seeks to understand mobility change from the archaeological record of a single site, i.e., Inamgaon by reanalysing earlier arguments and bringing some new methods in this respect.

Inamgaon The site of Inamgaon, a village in Shirur taluk, Poona District, Maharashtra State (Western India) is located

89 km east of the city of Pune (Fig. 1). The entire area falls in the semi-arid tract where annual rainfall is very variable and scarce. The ancient site 3 km away from the modern village is situated on the right bank of the river Ghod. This site consists of five mounds measuring 550  430 m of which INM-I was subjected to extensive excavations. The Chalcolithic culture is divided into three phases based primarily on pottery types, Malwa, early Jorwe and late Jorwe phases. The dates for the three phases are 1600–1400, 1400–1000, and 1000–700 B.C., respectively (uncalibrated dates). Multidisciplinary work was carried out on this site to throw light on subsistence, habitation pattern, as well as diet and diseases of the inhabitants of the past. The material remains on the site comprised mainly structures, mostly mud huts numbering 134 which have survived only in the form of floors which are destroyed by later pits. The cultural material recovered includes pottery, copper, terracotta, stone objects, chalcedony lithics, beads and bone objects. About 266 burials, some symbolic, have also been unearthed. The wealth of evidence gathered from systematic prolonged horizontal excavations aided by scientific analysis was used to reconstruct past lifeways at Inamgaon. The subsistence base was presumed to be subsistence agriculture, stock-raising, and hunting–fishing. It was presumed that sedentary agriculture was the main occupation of the Chalcolithic farmers till about the close of the second millennium B.C. (the late Jorwe

Fig. 1. Map of India with study area.

S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

period),when due to increasing aridity, the people changed to semi-nomadic pastoralism (for details see Dhavalikar, 1988; Dhavalikar et al., 1988). Environmental change was based on the presence of a sterile layer found at the Chalcolithic sites of Nevasa and Prakashe both in Maharashtra (in river valleys north of the site of Inamgaon) which proved to have been formed due to severe aridity (Mujumdar and Rajaguru, 1965) as well as scientific investigations in Rajasthan in western India (Krishnamurthy, 1981).

Reanalysis of earlier arguments Earlier researchers had postulated a mobility-subsistence change at Inamgaon on the basis of criteria like house shape, organic remains, house form, and other aspects (Dhavalikar, 1988). One of the main criterion used in analysing mobility change was house shape; based on ethnographic analogues it was presumed that a change from rectangular structures during the early Jorwe period to circular ones during the late Jorwe levels denoted a mobility shift from a settled life to a more nomadic one. It is true that some ethnographers and archaeologists have equated round houses with relatively impermanent settlement patterns and rectangular houses with relatively permanent ones (Binford, 1990), but others find that sedentary, semi-nomadic and nomadic peoples have both round and rectangular houses. However, we have to remember that functional attributes, environmental concerns, symbolic values amongst other factors can also determine house shape. It has been seen that house shapes can be conditioned by the nature of the family unit occupying it (single individual or family) or the variety of activities conducted inside the structure or the symbolic significance of particular shapes in a society (Flannery, 1972; Hunter-Anderson Rosalind, 1977). Thus, house shape cannot be taken as an indication of mobility without seeing it in perspective with other aspects of material culture. House form is another criterion through which earlier researchers have argued for mobility change at Inamgaon. It was deduced that the presence of large, well built houses in the earlier phases denoted permanent settlements while flimsy houses in the later phases signified a change to a semi-nomadic temporary existence (Dhavalikar, 1988). But this criterion is open to question as well. The protohistoric Hidatsa who live from April to November in the Missouri valley construct wellconstructed wooden houses that are reused over a period of years even though they do not settle the whole yearround. Moreover ethnographic work with the Dhangars in Pune district, Western India (Panja, 1996a) has shown that groups who are semi-sedentary and move part of the year to places nearby for pasture construct houses similar to the sedentary group who practice diurnal


mobility while semi-nomadic Dhangar groups who come to the same place and occupy the same dwellings often build very substantial structures. Therefore this trait cannot be taken as an absolute criterion to distinguish between different mobility strategies. Settlement planning has also been used to corroborate the hypothesis of mobility change at Inamgaon. It was argued that well-laid out houses in the earlier layers indicated a more stable existence while an irregular layout in the later levels (late Jorwe) denoted a change to a nonsedentary existence (Dhavalikar, 1988). It is questionable, though, whether it is of any real use in using this criteria to separate sedentary from nonsedentary occupations. Many nonsedentary groups live in settlements in which all houses are uniformly oriented, while sedentary people live in houses that have been built with no apparent central plan or uniformity of orientation (Flannery, 1972; Rafferty, 1985). Similarly, organic remains were also used by earlier researchers to substantiate the hypothesis of sedentism to semi-nomadism. In the early phases a predominance of grains and cattle signifying agricultural activity was linked to sedentism, while in the later levels, a decrease in botanical remains and an increase in sheep/goat bones was used to indicate a change to semi-nomadic pastoralism. But we must understand that subsistence change may not always be linked to mobility patterns; there are agriculturists who are mobile (Graham, 1989) and sheep/goat pastoralists who are sedentary. Moreover organic remains should be analysed with regard to the context in which they are found along with an analysis of post-depositional disturbances which affect the remains. For example, a predominance of sheep/goat remains found from an area, which has been used as a habitation zone probably reflects the subsistence of the people of the site. But similar bones found in an area that has been used for short-term camping purposes, do not indicate that this was the dominant subsistence pattern of the people; rather the number could be due to the use of the site by different individuals who kept sheep/goat and were using the area as a herding zone. Therefore the context of the record has to be taken into account in order to assess its role in understanding human behaviour. Earlier work at Inamgaon was thus carried out using isolated aspects of material culture using simplistic ethnographic analogies to interpret material culture patterning. It is suggested that such a one-to-one link between material culture and human behaviour could result in an inaccurate interpretation of past cultural life; moreover a nonholistic view of material culture can result in erroneous conclusions as the change in material culture at Inamgaon could well be due to other factors not necessarily related to mobility or cultural change. A critical and holistic analysis of the nature, context and formation of the archaeological record is necessary be-


S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

fore it is used to fathom human strategies. It is with this perspective in mind that analysis was undertaken to rethink the link between mobility change and the archaeological record at Inamgaon. Reanalysis of the site of Inamgaon It has been argued that earlier researchers did not critically interpret archaeological criteria for identifying different mobility strategies at Inamgaon. One of the most important arguments was related to the change in the shape and type of structures or ‘‘houses’’ on the site. About 134 ‘‘houses’’ (mud structures), both rectangular and circular have been excavated at the site of Inamgaon. To reanalyse earlier arguments, the first step is to analyse the ‘‘house’’ itself as a holistic phenomenon not its isolated characteristics (like shape or form) as earlier researchers had done. This implies analysing the facilities and artefacts within the structure itself. Facilities include fire-pits (these, as defined by the excavators seem to be big hearth-like structures), silos, and hearths the analysis of which can be used to understand the function of the house. For example, fire-pits, hearths, and silos can be used to indicate that the house was used as a domestic unit. The presence of only silos or big fire-pits not usually used for domestic purposes could be taken to indicate that the house was used for a nondomestic or public purpose. The fact that no facilities exist in houses could mean that they were not being used for full-time occupation but rather for short-term purposes. During the Malwa phase, out of 16 houses, seven had both fire-pits and silos, four had only fire-pits, one had

only silos, two had all facilities, two had none. Most of the houses probably functioned as domestic units as they had hearths or fire-pits and silos for storage. The average size of the houses was 8  5 m. During the early Jorwe period, amongst 29 houses about 18 had no facilities, of which six were disturbed, so about 41% of the houses had no facilities. Two houses had all the facilities, one each had fire-pits and silos and hearths and silos, two had silos only, one had hearths, and four had fire-pits. The average size of the houses was smaller 5  3 m. During the late Jorwe period more than 50% of the houses had no facilities. Amongst the others four had fire-pits, four had fire-pits and courtyards, two had silos only, and only one had a hearth and silo. Most of the structures had no facilities. The average size was the smallest. The analyses of house facilities seem to show that these structures functioned in diverse ways throughout the entire time span of the site. In the first cultural period, Malwa, houses have more facilities indicating they are used as domestic units. In the early Jorwe levels the houses have less facilities but are larger in size than the late Jorwe houses hence they could have been used as nondomestic units. In the last phase the structures had less facilities so they could have been used as short-term camps (Fig. 2). Examining earlier arguments linking shape (rectangular vs. circular) and movement change, we find that the facilities and artefacts in houses of similar shapes differ chronologically. In the early Jorwe phase, rectangular houses had facilities while circular houses had none, while in the late Jorwe phase circular houses had hearths and courtyards while rectangular houses

Fig. 2. Shape of houses and facilities.

S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

Fig. 3. Artefacts and house shape.

had few facilities (Fig. 3). Moreover if we look at the artefacts inside the structures, circular structures seem to have more of a variety of artefacts than rectangular structures in the early Jorwe phase, while in the late Jorwe phase there seems to be no difference in artefact numbers and variety between rectangular and circular structures (Fig. 4). Therefore houses which are similar in shape have different facilities and artefacts in different cultural phases, which could be due to the fact that the functions of the houses changed through time. Thus, shape of houses, the principal original argument for a shift from a sedentary to a mobile way of life, seems irrelevant to the reconstruction of mobility strategies, as the change in the later levels in house shape could well


mean a shift in the functional nature of the site, rather than a movement change. The ‘‘houses’’ at Inamgaon represent a complicated phenomenon, it seems that houses of diverse shape and form were playing different functional roles during different levels of the site. If we combine this with other forms of evidence a similar picture emerges about the site. As far as organic remains are concerned, during the Malwa period, the total number of winter crops (wheat, barley, common pea, chick pea or gram, grass pea) surpassed that of summer crops (horsegram and hyacinth bean) indicating some seasonality (Kajale, 1988, p. 802). In the early Jorwe period, summer crops and winter crops are present in equal amounts (Kajale, 1988) which shows that the population was staying for a considerable part of the year. In the late Jorwe phase palaebotanical analysis revealed fluctuating cropping patterns comprising both winter and summer crops, which can occur when groups from different geographical zones are occupying the site (Kajale, 1988). This, together with the lack of facilities in the structures seem to suggest that the site was probably being used for short-term purposes by diverse groups during this cultural period. From the percentage of botanical remains, it also seems that people with varying mobility strategies occupied the site differentially through time for different purposes. Another form of evidence for mobility change is the nature and density of artefacts found on the sites. It has been perceived that artefacts found at sedentary sites tend to be heavy as well as various and numerous (Rafferty, 1985). The increase in artefact numbers can be attributed to the increased time spent at the site as well as the tendency for people to accumulate more possessions when they no longer have to transport them frequently from place to place. If we look at the total number of artefacts from the site from standardised versions, artefacts are more in numbers in the later levels, 2132 as compared to 1103 and 1073 in the Malwa and early Jorwe levels, respectively (Table 1). From ar-

Fig. 4. Facilities in houses.


S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

Table 1 Standardized number of artefacts Level



Other artefacts

Late Jorwe Early Jorwe Malwa

4050 3979

47,454 25,732

2132 1073




Standardization per cubic meter deposit.

tefact counts we see therefore that it is difficult to postulate a change from sedentism to semi-nomadism of the population on the site as the artefacts increase during the later levels even with standardised numbers. The increase in artefacts in the later levels can however also be explained by the fact that repeated occupation of a place by various groups of people can generate a greater density and volume of cultural material (Macneish, 1973 quoted in Rafferty, 1985). The high density of artefacts could therefore be due to the fact that many groups were repeatedly using a part of this area for short-term purposes and the site of Inamgaon could have been served as a temporary camping ground for different people in the late Jorwe levels. From the preceeding analysis it seems that the history of Inamgaon seems to be a complex one. The site could have been used in different ways; for domestic, nondomestic or short- term purposes. Therefore change in structure shape, form as well as a shift in the nature of organic remains do not only signify a change in movement strategies. It could well mean a shift in the function of this particular area of the site. Change in material culture could have been functional not cultural. To understand this change we have to employ some other form of analysis. Archaeologists have used another criterion, the study of site structure or what is commonly known as intra-site spatial analysis to understand mobility change and it is to this that we now turn (Hitchcock, 1987; Kelly, 1992; Kent, 1992; Killion, 1990; Kroll and Price, 1991; Murray, 1980).

Intra-site spatial analysis Intra-site spatial studies encompasses a wide area of research, but put in a very simplistic way it seeks to understand human behaviour through the patterning of artefacts and ecofacts within a site. From the beginning archaeologists used intra-site patterning to deduce how and what a site or different parts of it was used for. In the initial stages a site map was scrutinised to ascertain the presence of an overall structure or pattern to the distribution of artefacts that could be used to substantiate the archaeological site as a camp or habitation. This approach was evident in works like Grahame ClarkÕs Starr Carr or LeakeyÕs interpre-

tation of the Oldowan sites at Olduvai Gorge (Kroll and Price, 1991). In the second stage there was a more conscious attempt to interpret the function of a site or structures. Leroi-GourhanÕs research at Pincevent in 1966 and J.N. HillÕs work on Broken K. Pueblo (Hill, 1970) are outstanding examples of this sort of work. Certain propositions regarding various activities about room types were drawn up from modern ethnographic analogues and these were tested in the archaeological context. Areas of activities were known as ‘‘activity areas’’ and the objects used for specific activities ‘‘tool kits.’’ These efforts aimed at identifying ‘‘activity areas’’ and ‘‘tool kits’’ became an important concern in archaeological research in the 60s and 70s. Coupled with this were improvements in statistical techniques to identify spatial patterning within sites. Definition of tool kits and recognition of spatially differentiated activity areas were usually sought through quantitative analyses of material recovered from ‘‘living floors ’’ (Carr, 1984). The earliest problems arose with the nature of sampling of the artefacts. It was also realised that tools are not functionally specific nor are activity areas spatially dispersed. Ethnographic work further confirmed this assumption about the multifunctional nature of tools and activity areas (OÕConnell, 1987). Similarly KentÕs work also tested the ad hoc propositions that most activity areas are sex specific or that most activity areas are monofunctional (Kent, 1984). This concept of understanding activity areas of the past was further systematically criticised by Schiffer (of course Ascher had also dealt with it earlier) who argued that one must take into account how the archaeological record was formed before identifying how space was partitioned on that basis. This involved considering natural processes which may have disturbed the original context of the record as well as human activity or cultural processes to determine whether the artefacts or structures were used for specific activities or represented a palimpsest of diverse activities. (Schiffer, 1987 this aspect has been dealt with in numerous publications; Hietala, 1984; Kroll and Price, 1991; Simek, 1989.) This concept was further expanded by Binford who argued that mere enumeration and mathematical expositions of formation processes failed to take into account the differential manner in which recognisable behaviours are integrated relative to other archaeological indicators of past behaviour. BinfordÕs Middle-range research was an advance on SchifferÕs ‘‘Behavioral Archaeology’’ and aimed at trying to understand why in different systems material culture is formed differently. This fact was then used to identify different subsistence-settlement systems (Binford, 1987). Following Binford similar works on intra-site analysis were undertaken to understand different subsistencesettlement systems, pastoral, sedentary, and semi-sed-

S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

entary agriculturists through spatial patterning across a site (Graham, 1989; Sadr, 1991; Simms, 1988). This sort of work was criticised on the point that examples from ethnographic situations are episodic and are not able to capture the long-term scale of the archaeological record (Stafford, 1985; Wandsnider, 1989, 1996). It has been observed that archaeological spatial analysis has been unproductive to date as the goal has been to ask specific functions to which specific locations within a site were put. Assemblages found at specific locations rarely reflect the functions of those intra-site locations. The change is to shift focus from the spatial arrangement of technology or faunal elements to the arrangement of one or more attributes demonstrated to be sensitive to organisational properties of deposition to understand the duration, frequency, and planning of land use over a region (Stafford, 1985; Wandsnider, 1989). Post-processual work on space and notions of space have questioned this premise of understanding functional activity areas, formation processes and identification of subsistence-settlement systems, as according to the proponents of this school of thought space is organised according to the meaning it has to the people using it (Hodder, 1990; Moore, 1986). Depositional practices are guided by the way in which items and substances are classified. Archaeological deposits are generated in accordance with cultural rules than natural laws and are arbitrary and conventional. They are open to manipulation and reinterpretation (Thomas, 1991). Refuse is a symbolically loaded category of material which may be used and located in meaningful ways. To the post-processualists, spatial variation in refuse disposal is likely to be the interaction between functional requirements and differing attitudes to various artefacts types and by-products. Thus it is difficult to generalise about activity areas and functional use of site as advocated by the post-processualists. Thus the aims of spatial analysis ranged from understanding activity areas, subsistence-settlement systems to cultural meanings and codes. Intra-site spatial analysis and mobility strategies Following the methodological premise of BinfordÕs middle range research, this work aims to use site structure to distinguish different types of mobility from the archaeological record, using actualistic situations as a heuristic model with which to understand intra-site patterning of material culture. In archaeological research spatial analysis has been used to understand the material consequences of diverse forms of mobility stategies. HitchcockÕs work on the Basarwa (Hitchcock, 1987) shows that sites occupied for lengthy periods tend to be cleaned up regularly. Specialised dumps and the distance from hearths and houses to dumping areas increase in sedentary situations as do


site size and storage features. As mobility decreases, internal site differentiation increases due to the use of specific areas for specific activities. While many conventions, i.e., social organisation and cultural conventions affect the use of space, the site structures among nonresidentially mobile horticulturists and settled foragers and pastoralists appear to be more internally differentiated than mobile people (Kelly, 1992; Kent, 1992). MurrayÕs work on sedentary and mobile populations show that sedentary and semi-sedentary population throw their discarded elements away from the living space hence there is no link between use and discard location. In migratory populations activity areas can be understood as use and discard locations are similar. GrahamÕs work amongst the Raramuri show that mobility amongst agriculturists is recognisable through the organisation and use of space at residential sites. Differences between sedentary agriculturists and mobile agriculturists were identifiable through the occurrence of permanent facilities, formalised activity areas, and more thorough site maintenance behaviour in sedentary groups as compared to mobile groups (Graham, 1989). Following this methodological approach the aim is to use spatial analysis to throw light on mobility strategies keeping in mind the long-term consequences of behaviour (Wandsnider, 1989, 1996). However this work is being done with one initial assumption in mind, that human behaviour was being dictated by ‘‘rational’’ notions of cleanliness and time spent at a site. But humans have their own notions of hygiene and thus different groups might maintain their space in different ways in different cultures. Moreover different cultures might have different notions of material culture, pottery or bone or ash and thus might discard refuse in different ways according to the perception they have about the objects used and discarded (Hodder, 1990; Moore, 1986). My point in this work is that function and symbol are inextricably linked in these elements of routinised practice. It is being argued here that ‘‘functional’’ considerations also condition use and refuse patterning as do ‘‘symbolic’’ perceptions of space. Every action has meaning, and even functional considerations is a meaning in itself. Meaning is not extra and above function, rather it is a part of it (Hodder, 1989). Therefore one consideration, in this case functional is being taken as a backdrop with which to compare the archaeological record. This functional notion can be critiqued if archaeological evidence points out to a different notion of refuse disposal.

Context of material culture Site structure is an important aspect to deduce mobility strategies from the archaeological record. Never-


S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

theless the emphasis on a particular aspect of site structure studies, be it activity area study or maintenance behaviour, can only be carried out depending on the nature and context of the archaeological record. Earlier researchers assumed that the artefacts found on living surfaces directly represent their original context of use. Many interpretations were put forward regarding the structures as well as the whole site with this assumption in mind. Moreover they had also postulated certain functional attributes to the site especially of the structures. This is in keeping with the trend in early spatial work that artefacts deposited reflect their place of use. But subsequent work has shown that artefacts are not always left where they were originally used but undergo certain modifications, both natural and cultural, which are termed as formation processes. Moreover a structure or a site used for many years might also have changed its character and context. The archaeological record rarely preserves episodic events,and is not conducive to ‘‘finegrained’’ behavioural interpretation. Rather it is a palimpsest involving differential use of artefacts, structures and place. So the context of the artefacts has to be understood to see if it can be used to understand activity areas or other aspects of the past. Thus these palimpsest deposits need to be studied from geological deposition and long-term behavioural dynamics which lead to re-use of facilities and artefacts. Whether discrete ‘‘occupation’’ of the site is distinguishable depends on rate and tempo of sedimentation. The archaeological record may be fine-grained as evidenced in thin separate lenses or coarse-grained resulting in a complex palimpsest (Petraglia, 1987). The archaeological record differs in terms of their inherent integrity; some may contain more research potential for activity areas while others may throw light on other questions. If a sedimentation event followed each occupation of a site after abandonment, then temporal resolution in a deposit would be episodic (Petraglia, 1987; Wandsnider, 1987). The Inamgaon deposits do not contain thin and welldefined lenses of material thus one has to investigate how far the archaeological record at Inamgaon is affected by natural and cultural processes. Can we talk of episodic behaviour of the past? How far do post-occupational natural factors distort the archaeological record? Moreover how far do these artefacts reflect in situ use rather than secondary dumping or abandonment processes? Natural formation processes Before we try to understand the cultural implications of the material record, we have to account for post-occupational disturbance caused by natural factors, called natural formation processes. This work is very limited as

the excavators do not always describe the trenches or houses in great detail. So information is gathered from examining the artefacts themselves combined with descriptions of trenches and houses from the fieldnotebooks, excavation reports, maps, and section drawings. A deposit, whether it is a floor or fill of a house or a layer of trash can contain the products of many differential depositional processes, both cultural and noncultural. It is sometimes difficult to discern accurately the processes it has undergone, due to the fact that different processes produce similar traces or later processes obliterate earlier ones. Nevertheless this part is just a preliminary attempt to understand natural formation processes to identify whether the ‘‘deposits’’ we are going to investigate are caused due to natural or human factors. There are many processes which transform the archaeological record which include fluvial activity, weathering, bioturbation amongst many others (Nash and Petraglia, 1987; Schiffer, 1987; Wilseden, 1982; Wood and Johnson, 1978). This analysis is at two levels, one is devoted to understanding the nature of structures and its associated artefacts; in the other, the artefacts themselves as well as the associated sediments inside and outside structures are taken as the basis of analysis. In the latter case artefacts are taken as the major focus of enquiry as information about the sediments is fragmentary. The analysis of structures aims to show whether the artefacts contained in the houses are contemporary with the structures or deposited after the structures were abandoned. This is understood from the nature of disturbance processes affecting the houses and the artefacts contained inside them. If similar processes are affecting both the houses and the artefacts we can say that the artefacts are contemporary with the structure. If not, we assume that the artefacts were deposited by other natural and cultural processes after the house was abandoned. The houses are in different stages of preservation. Many of the house floors are disturbed due to erosion, or bioturbation or burning. We find that on both disturbed and undisturbed floors, the number of artefacts are similar. The total artefacts in fact are more in number in disturbed houses than undisturbed houses, 4467 as opposed to 3969 (Table 2). Many of the artefacts in some of the structures do not seem to be contemporary with the working of the house but were deposited after the abandonment of the ‘‘original’’ structure as the specific disturbance processes affecting the houses do not seem to be affecting the artefacts (Table 3). The disturbance caused to the houses are not affecting the artefacts, so other processes are playing a role in the patterning of artefacts in the structures. Later pits dug into structures could also lead to infiltration of artefacts from later layers, hence we have a lot of artefacts in houses disturbed by later pits. But we see that both

S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125


Table 2 Artefacts in disturbed and undisturbed houses Artefacts

Malwa houses D

Early Jorwe houses

Late Jorwe houses






>10 10–20 >20

3 1 8

0 1 4

5 1 11

1 0 11

3 2 11

3 0 21








undisturbed and disturbed structures have a lot of artefacts, so pits are not the only factor responsible for artefact accumulation. How have these artefacts accumulated, through natural processes or human factors? From field observations, bioturbation, and fluvial abrasion were taken as major factors affecting the site. The information about the sediments was gathered from notebooks, section drawings, maps, and field observations on the site. For the artefacts, weathering, bioturbation, fluvial abrasion, rodent activity, corrosion, thermal fractures, and pitting were identified as the main factors affecting terracotta, stone, copper, stone artefacts, pottery, and bones. Attributes like size, rounding of edges due to abrasion, rodent marks, weathered and pitted surfaces, thermal cracks and fractures, and corrosion were sought to be identified on the artefacts to account whether it was formed by natural or cultural processes or both. There is evidence of disturbance processes affecting a few parts of the site like fluvial action, rodent activity, and bioturbation as gathered from the excavation reports and section drawings. Laboratory analysis however revealed that that most of the artefacts on the site of Inamgaon are fresh and less disturbed. In an overall analysis, a negligible amount, 4.19% in the structures, and 7.89% outside structures show signs of weathering or corrosion or rodent activity. There are few instances of other indicators like size sorting caused due to natural processes like rodent activities or fluvial action. But the assemblage size is not affected in parts of the site with evidence of natural disturbances. Therefore the spatial patterning seems to be caused more by the use and reuse of the site by humans than by natural factors. The ‘‘anamalous’’ artefacts could be natural intrusions in an

Table 3 Disturbed houses and artefacts House



134 105 129 71

Burnt Burnt Burnt Rodent

No sign of burning No sign of burning No sign of burning No size sorting or marks on artefacts

otherwise homogenous assemblage, or the evidence of different episodes of human activities. Thus these ‘‘disturbed’’ areas or artefacts have been excluded from the analysis of the spatial patterns across the site. Cultural formation processes of the archaeological record From an analysis of the artefacts and the general description of the deposits, it is surmised that the patterning seems to be primarily due to human factors rather than post-occupational disturbance factors. What is the nature of this human factor, does it represent in situ use areas or dumps and secondary refuse deposits? Ethnoarchaeology has shown that different activities are often performed at the same place and same activities at different places in a site. Activities do not necessarily produce co-variant sets of artefacts in proportion to the frequency of performance. Refuse produced by a specific activity need not be deposited at the point of production (Kent, 1984; OÕConnell, 1987; Yellen, 1977). Hence the notion of activity areas is a complex category which needs to be investigated, rather than assumed. To understand the nature of the deposit we have to take into account the type, variety, size, and completeness of the artefacts to demarcate primary from secondary refuse as well as to understand the nature of abandonment. These concepts of primary, secondary refuse, maintenance, dumps are open to critique as different cultures have different notions of hygiene. Nevertheless we operate with our own biases, and hence these concepts are being defined with how we understand things today. This can serve as a background to analyse varying notions of refuse in different cultures and through time. In this part, Schifferian notions of primary and secondary refuse and de facto refuse are being used to understand the nature of the deposit; i.e., as to whether is represents in situ use or secondary dumping (Schiffer, 1987). The second arena of analysis concerns the structures with the artefacts contained within them. We saw in the earlier section that many of the artefacts contained in the structures were deposited after their abandonment. So it is difficult in many instances to reconstruct the function of a house from its contents as it has undergone reuse in diverse ways. This section also aims to analyse if


S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

the houses have undergone reuse and if so for what purpose. How far can these structures be used to reconstruct episodic events? Can we talk of a single lifehistory of the structures? Method of analysis The information has been gathered from the excavation reports, unpublished field notebooks, maps, and diagrams of individual houses. The nature of artefacts has been arrived at from examining the artefacts themselves, coupled with information from notebooks and antiquity registers. Artefacts have been analysed on the basis of their size, type and completeness, and state of preservation (to denote reuse potential). A comparison of the artefacts enabled one to distinguish large from small as those less than 5 cm and those greater than 5 cm maximum dimension. This is not an absolute category but just a device to distinguish larger or smaller artefacts. The ethnographic examples which state dimensions that can be used (OÕConnell, 1987; Wandsnider, 1989) are not found useful here as the data is limited and very small artefacts or microrefuse were not collected by the excavators. So an average number based on the sizes of the artefacts has been used as a basis for comparison. Wherever possible some refitting studies were carried out with pottery and other clay and stone objects. The nature of the refuse (based on the plethora of archaeological literature on the subject), either primary, secondary or de facto as well as abandonment processes are being identified to see whether activity areas can be identified on the site of Inamgaon. Some of the terms and methods of analysis are outlined below. Primary refuse This is indicative of in situ activity areas and is characterised by small objects which are left behind after cleaning in maintained areas (Schiffer, 1987). Secondary refuse To distinguish secondary refuse, the density of artefacts, diversity of artefact types, size (more large), and low reuse potential (defined by incompleteness and state of the artefact) are taken into account (Douglas, 1994; Schiffer, 1987). These cannot be used to identify activity areas as they are redeposited refuse. De facto refuse This is left behind when sites are abandoned suddenly and can be used to identify activity areas. These are complete and still usable artefacts which would be usually curated in normal circumstances (Cameron and Tomka, 1993; Stevenson, 1982). These may also include artefacts which can be refitted.

Activity areas Activity areas have been identified as those which have distinct types of artefacts, pottery or tools or bones as well as other evidence of activity. For example tool making areas would have evidence of lithic debris. Craft areas would have debris as well as finished products. Facilities, like a kiln or anvils would point to the existence of an activity area. A butchering area would have the meat bearing bones with cut marks, bone fragments as well as lithics or other butchering implements. Reuse of houses Reuse of houses was judged in two ways, one in which disturbed floors or houses have numerous artefacts which do not seem to reflect any disturbance process thus implying that probably the artefacts were deposited after abandonment. The second instance is one in which houses which have facilities had secondary refuse which implied that houses which were used for habitation or other purposes are now being reused as dumps. Analysis of structures and artefacts Earlier researchers had used the nature of the houses and their contents to give functional interpretations of various parts of the site. In this part each of the houses and its contents will be taken up and analysed separately to understand how far the earlier interpretations were valid or not. Floor level in situ artefacts can give us an idea about the original function of the house. An analysis of floor level artefacts revealed that few of the houses had floor level artefacts (Table 5). Most of the data in the houses were fill data not found directly on the floor of the house. How far can this data tell us about the original function of the house? What is the nature of this material? Is it material which accumulated due to use or secondary dumping? Can it tell us about how the house was used in the past? From the analysis of the contents of the houses we see that most of the houses (61) seem to be used or reused as secondary refuse areas (Table 4). Many of the houses seem to have been reused for different purposes as evident from the analysis of the floors, the facilities, and the artefacts contained within it. Hence it is difficult to talk of episodic events in this case as the record is a cultural palimpsest formed of various activities. It might be more appropriate to consider the varieties of use of a structure as arrayed along a continuum. At one end is the status of full-time use and the other total abandonment. In between it might be used for (1) part-time living; (2) use as a storage area; (3) use as a dump site. In this case these structures or area around the structures seem to be dumps as evident from the size, variety, and

S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125


Table 4 Nature and reuse of structures Cultural periods

Reused as dump

Dump—no reuse

Activity area present

Disturbed with no artefact


Malwa Early Jorwe Late Jorwe

10 15 16

0 7 13

2 1 5

5 6 6

17 29 40







Table 5 Artefacts on floors Cultural period Late Jorwe Early Jorwe Malwa Total



Bone tools

6 6 2

12 2 1

3 1 0




state of the artefacts (Panja, 2000; Rothschild et al., 1993). An analysis of the artefacts outside the structures revealed that most of the artefacts were large and of a diverse variety (Figs. 6 and 9). In a few areas (Figs. 12 and 14) there seems to be some maintenance activity and use areas could be identified. The analysis shows that most of the artefacts discarded seem to be of large or diverse sizes and of different types. There are also very few complete artefacts thus denoting a lack of de facto refuse. Therefore it is difficult to talk of activity areas here as the material seems to be secondarily deposited as dumps. It is difficult to talk of episodic events in this case as the record here seems to be a cultural palimpsest formed of various activities.

Mobility and site structurean intra-site study of Inamgaon Refuse and mobility—an actualistic perspective (for details see Panja, 1996a,b) Since most of the artefactual material at Inamgaon (from the analysis done with the structures) seems to be in the form of secondary refuse, maintenance behaviour is taken as a focal point of enquiry in this regard. Actualistic work on refuse patterning amongst different groups can provide us with a guideline with which to understand past mobility strategies. It has been seen (Graham, 1989) that difference between sedentary and mobile societies are evident from many aspects. Activity areas are more specialised in sedentary groups than mobile ones. Moreover some aspects like storage facilities are more common amongst

Stone object



Total houses

7 7 1

3 0 0

1 0 0

40 29 17





sedentary groups. Refuse patterning can also be seen to differ in sedentary and mobile societies. Ethnographic work has shown that there are differences in space use and management of refuse between groups who are sedentary and those who are mobile (Murray, 1980; Sadr, 1991; Simms, 1988). These works have dealt with just two broad groups, sedentary and mobile. There has been no attempt to understand the various dimensions of mobility and refuse disposal practices. Thus actualistic work was undertaken to understand how sedentary and semi-nomadic groups use their space and particularly manage their refuse in order to provide a heuristic device with which to compare patterning in the archaeological record. Accordingly intra-site spatial work was undertaken on occupied sites of sedentary Dhangars and two abandoned sites of the semi-nomadic Dhangars and compared with the data from published work dealing with similar problems. In this work, four different dimensions of mobility, sedentary, semi-sedentary, semi-nomadic and nomadic (as gauged from ethnographic and published work) were taken up for analysis. Field work and literature survey (Graham, 1989; Killion, 1990; Panja, 1996a,b) revealed that artefact patterning in sedentary groups with diurnal mobility is characterised by maintained structures and the area around structures with definite dumping grounds away from the structures or on the periphery. As sedentary groups stay in one place for a greater length of time, in the long-term dumps will be large and placed in a certain place (Fig. 16). Semi-sedentary groups (where a section of the population migrates for short periods of time) would have smaller but similar dumps. Semi-nomadic mobility (for example the Dhangar groups) represent one form of movement


S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

where the groups have permanent places of residence to which they return annually. On the basis of fieldwork and published literature (Panja, 1996a,b; Simms, 1988) we assume that in the long-term semi-nomadic groups who return to the same place and use the same facilities will have maintained structures with bigger artefacts near the structures and a random scatter of small artefacts in the other areas (Fig. 17). Sites of nomadic groups who are continuously on the move and do not use the same place repeatedly even though the region of their movement remains the same have a random scatter of artefacts all over the site through time (Sadr, 1991; Wandsnider, 1992, per comm.) (Fig. 18). Intra-site spatial analysis at Inamgaon We saw in the earlier sections that most of the material discarded on the site of Inamgaon was of a secondary nature resulting from dumping activities rather than in situ use. The actualistic section showed how the patterning of refuse can throw some light on the nature of mobility strategies of the people. This is to be used as a heuristic model with which to analyse the nature of movement from the archaeological record. The entire analysis is based on the maps (Figs. 5–15).1 To designate dumps we have used the criteria stated earlier in the chapter, i.e., the presence of a high density and a diverse variety of broken artefacts mostly large in size. Activity areas have been demarcated as those zones where the cluster consists of a single variety of artefacts. Use areas as opposed to dumps can also be identified by the presence of a low density of artefacts, usually small in size. Method of analysis Artefacts based on size, completeness, and type have been analysed across the site through the layers. Completeness of artefacts is taken to indicate the nature of abandonment. The density, size, variety, and completeness of artefacts is used to distinguish secondary refuse from primary refuse. This patterning is aimed to answer questions: Where are the dumps placed? What is its relationship with activity areas? What is the relationship between different types of artefacts? The phase called the Malwa period (Layers 16 and 15), was excavated sparsely, but certain patterns emerge


The maps were generated through a programme written in QBasic which linked up Dbase III+ files with graphical images of trenches and structures. The houses have been all drawn as circular or rectangular even though the shape might fall in between these two broad categories. All houses are enclosed to understand spatial patterns even though they might not be perfect squares or circles. The scale is .

(Figs. 5–7). Artefacts are concentrated in definite areas, they are usually large and broken and are of varied type, hence we can call them as dumps. There are very few complete artefacts hence abandonment must have been planned. But this could also be due to natural processes as artefacts in the bottom layers tend to get crushed. Certain territories are being used to discard refuse especially near the structures. Most of the houses have hearths or silos or fire-pits but contain secondary refuse. This could mean that they functioned as independent households and were reused as dumps. The same areas were not used as dumps as then there would have been a cleared space and definite refuse disposal spots. The spots of dumping refuse were changing hence we have many areas with a cluster of artefacts. From ethnographic parallels we can assume that if a certain population was staying at a spot for a considerable period of time, there would be few dumps and few cleared areas especially around structures used as fulltime use-areas. The fact that there are many clusters of artefacts and reuse of structures shows that the same people or different groups were occupying structures periodically and reusing them differently. From the large size of the dumps and the presence of clusters of refuse we can surmise that the inhabitants were semi-sedentary as they were staying for longer periods of time. They were however not reoccupying the same spots. As in the earlier layers during the early Jorwe phase (Layers 11, 9, 8, 7, and 6) it seems that some of the structures are being abandoned and occupied by people who were using them as refuse areas hence it is difficult to distinguish episodic events, as the record is a palimpsest modified by groups of people) (Figs. 8–11). During this period few houses have fire-pits or hearths or courtyards. But some of the houses are large in size and possess silos and huge fire-pits which cannot be associated with individual families. So this part of the site could denote an area which was initially not used for habitation purposes but as store houses or craft areas or public purposes and later reused as dumps. Artefact patterning shows definite use of some area for specific activity purposes and others for dumping. The large size of the dumps and the presence of activity areas on this site where rate of deposition seems to be slow would mean the continued use of the site implying that people were staying there and using the place for a long period of time or that semi-sedentary people were reoccupying the same place again and again. The fact that some spots were reused also shows that other groups were abandoning the structures which were being reused for other purposes by the same or different people. Nevertheless some change could be discerned between the layers, layer 11 show random distributions, 8 and 9 show some definite areas of activity, 7 and 6 show clusters of all types of artefacts. Probably there was

S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125


Fig. 5. Artefact scatter—Layer 16.

Fig. 6. Size of artefacts—Layer 16.

some differential use of the site through time. In layers 8 and 9 probably some parts of the site were used as activity areas, while others were dumps, while in the 6 and 7th layers most of the area was reused as dumps. Taking into account the differential patterning through the layers, the large size of the dumps, the specific spots of the refuse areas, the reuse of structures, the presence of some areas of specific activities, it seemed that this site during this period was probably occupied by both sedentary and semi-sedentary groups for long periods of time.

In the late Jorwe period (Layers 4, 2, and 1) there is a dense concentration of artefacts near the structures while artefacts on the rest of the site are randomly scattered (Figs. 12–15). The dumps are smaller than the earlier layers. The size of the artefacts are mostly small so the area seems maintained and rapidly abandoned. Some areas can be identified as activity areas. In layer 4 we have smaller dumps, but in the later levels there is a random scatter of artefacts with a few clusters. Most of the houses do not have any facilities. Earlier researchers talk of a communal structure of society with


S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

Fig. 7. Artefact scatter—Layer 15.

Fig. 8. Artefact scatter—Layer 11.

each block of houses having a fire-pit or hearth but that is not evident on the maps. Some of these structures were reused. As stated earlier most probably, some of the structures were not used for habitation purposes but for temporary use by families or individuals who came for grazing, etc. Earlier researchers had postulated a change to a semi-nomadic lifestyle based on sheep/goat. But what does this nomadism mean, were the people coming back and occupying the same structures? From ethnographic analogies this would result in maintained structures and a more concentrated scatter of artefacts near structures.

Or where they coming back and occupying the same area but not the same place? That would result in a random scatter of artefacts throughout the site. The analysis shows that there are small dense concentrations of artefacts near structures while the rest of the site is maintained with few small artefacts. The dumps are small and hence meant that the people were staying for shorter periods of time. Through ethnographic analogy (Fig. 17) one can surmise that these were the remains of semi-nomadic or semi-sedentary people staying for short periods of time. There is a also a random patterning of artefacts in the later levels which through analogy

S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125


Fig. 9. Size of artefacts—Layer 11.

Fig. 10. Artefact scatter—Layer 8.

(Fig. 18) seem to be the remains of mobile nomadic groups who were using the site as short-term camps. Taking into account the small size of the dumps, the maintained area on the site, the presence of specific activity zones, the lack of facilities in the houses, it seems that both semi-nomadic and semi-sedentary groups were staying in a part of the site for short periods of time while other mobile groups were using the major part of the site as short-term camps (see Figs. 17, 18).

Mobility strategies and the site of Inamgaon Inamgaon seems to have been used and reused by diverse groups for multiples purposes. If we combine this information with the patterning evident from other aspects of material culture certain patterns emerge. In the Malwa period, refuse areas are marked out indicating a somewhat long periods of stay. Most of the structures had facilities and seemed to be reused as re-


S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

Fig. 11. Artefact scatter—Layer 6.

Fig. 12. Artefact scatter—Layer 4.

fuse areas implying that different or similar groups of people were using the houses as domestic units, abandoning then and reusing them in diverse ways including dumping spots. From botanical remains one finds the

evidence of winter crops pointing to some form of seasonality. This shows that some form of short-term semisedentary movements took place during the early period the site was occupied.

S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125


Fig. 13. Size of artefacts—Layer 4.

Fig. 14. Artefact scatter—Layer 2.

During the early Jorwe period, dumps are also concentrated in and around structures; certain areas show some specific activity. Houses have less facilities and are large in size and are also reused as refuse areas. During this period it seems that houses were used for nonhousehold activity. From the nature of the concentrated dumps, the

presence of few activity areas and the presence of both summer and winter crops we can surmise that people during this time were both sedentary and semi-sedentary. During the late Jorwe period the dumps are scattered and concentrated around structures. A part of the area is well maintained with small artefacts. The houses do not


S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

Fig. 15. Artefact scatter—Layer 1.

Fig. 16. Spatial patterning—sedentary groups. Fig. 17. Spatial patterning—semi-nomadic groups.

have facilities. Botanical remains decrease during this period but alternatively summer and winter crops predominate through the layers. Wild animals increase in this period (Pawankar, 1996). As the dumps are scattered people seem to be staying for shorter periods of time. Some parts of the site seem to used for short-term

purposes as there are few artefacts. The presence of diverse pottery and fluctuating cropping patterns point to the fact that there were diverse groups of people occupying the site. Thus it seems that a section of the population were semi-sedentary or semi-nomadic and were

S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

Fig. 18. Spatial patterning—nomadic groups.

staying for short periods of time while nomadic groups were using other parts of the site for short-term purposes. The history and formation of the site of Inamgaon seems to be composed of multiple events. The site was used and reused by different groups of people for diverse purposes creating a palimpsest composed of different types of record creating processes. Thus the material change in the later levels could well represent a differential use of the site rather than a change in subsistence and mobility. The change therefore could have been functional rather than cultural. Functional change therefore refers to lateral shifts in the location of various activities, giving the appearance of temporal change at a given location even though there is no overall change in subsistence strategy.Work on hunter-gatherers (Binford, 1983) and agriculturists (Anasazi in Schlanger, 1992) shows that as the residential focus of a settlement system moves laterally across the landscape, various places previously used from the former residential locus may change roles in a very systematic way. Former habitation loci may be used as bases for short stays and for logistical forays aimed at collecting materials, hunting, etc., so that they function as limited activity loci or they may be maintained as potential bases for agricultural operations, functioning now as seasonal loci (Schlanger, 1992). Similarly DavidsonÕs work on the Palaeolithic cave sites in the Levante region of eastern Spain, Parpallo, Les Mallaetes, Volcan shows that the change evident in faunal remains at these sites does not signify a shift in subsistence strategies due to environmental degradation, rather a


change in the functions of the sites in relation to the exploitation of a given set of resources (Davidson, 1983). The material record at Inamgaon does not seem to indicate that there was shift from a primarily sedentary lifestyle to a semi-nomadic way of life; rather the change in the archaeological remains convey that Inamgaon was a place that was used for diverse purposes by diverse groups of people from a domestic, to a nondomestic zone to a short-term camping ground. Regional analysis of sites across the landscape also did not identify any settlement pattern change at the level of the region. Based on actualistic models, the regional settlement pattern seems to indicate a semi-sedentary way of life where a section of the population were migrating for short distances and occupying one of these four places regularly (Panja, 1999). It is hoped that future detailed site specific studies will throw more light on this phenomenon at more Chalcolithic sites in the same region. However from the preceeding analysis, we can conclude that the changes in material culture at the site of Inamgaon are more effectively explained as shifts in the pattern of use of the site within semi-sedentary to seminomadic agropastoral adaptations rather than representing a more dramatic shift from sedentary agriculture to semi-nomadic pastoral adaptations. This functional change is reflected in material culture giving rise to what earlier researchers labelled as the semi-nomadic and pastoral late Jorwe period.

Conclusion Archaeologists have had difficulty in identifying different forms of mobility because we do not really understand the relationships between movement and material record. It is argued that mobility is diverse and varied and there is no definite and direct link between mobility patterns and the ‘‘static’’ remains that is generated as a consequence. It is argued here that a critical and holistic view is required to understand movement from the archaeological remains in order to deal with the problem of ‘‘equifinality’’ of material record (different human behaviour can generate similar archaeological remains). Indicators of movement cannot be understood in isolation, rather a multitude of criteria is necessary to recognise human mobility patterns with actualistic work as a heuristic device to understand the diversity of human behaviour and its material consequences. This article is an interpretive attempt to understand this problem of understanding the various dimensions of movement and material culture both from an archaeological and actualistic perspective. Here mobility has been understood within a rational and ecoutilitarian sense, but this exercise is at best an ideal type to be used in interpretive strategies for understanding


S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125

how cognitive, cultural or ‘‘nonrational’’ variable must be figured into explanatory models of past human behaviour. It is this hermeneutical exercise between ‘‘source-side’’ and ‘‘subject-side’’ research (Wylie, 1989, 1990) that will enable us to understand the diversity of human behaviour in the past.

References cited Binford, L.R., 1983. Working at Archaeology. Academic Press, New York. Binford, L.R., 1987. Researching ambiguity: frames of reference and site structure. In: Kent, S. (Ed.), Method and Theory for Activity Area Research: An Ethnoarchaeological Approach. Columbia University Press, New York, pp. 449–512. Binford, L.R., 1990. Mobility, housing and environment: a comparitive study. Journal of Anthropological Archaeology 46, 119–152. Cameron, C.M., Tomka, S.A., 1993. Abandonment of Settlements and Regions. Cambridge University Press, Cambridge. Carr, C., 1984. The nature of organization of intrasite archaeological records and spatial analytic approaches to their investigation. In: Schiffer, M.B. (Ed.), Advances in Archaeological Method and Theory, vol. 7. Academic Press, New York, pp. 103–211. Cribb, R., 1991. Nomads in Archaeology. Cambridge University Press, Cambridge. Davidson, I., 1983. Site variability and prehistoric economy in Levante. In: Bailey, G. (Ed.), Hunter-gatherer in Prehistory. Cambridge University Press, Cambridge, pp. 79–95. Dhavalikar, M.K., 1988. The First Farmers of the Deccan. Ravish Publishers, Pune. Dhavalikar, M.K., Sankalia, H.D., Ansari, Z.D., 1988. Excavations at Inamgaon, vols. I and II. Deccan College PostGraduate Research Institute, Pune. Douglas, W.C., 1994. Identification and assessment of secondary refuse aggregates. Journal of Archaeological Method and Theory 1, 41–68. Flannery, K.V., 1972. The origins of the village as a settlement type in Mesoamerica and the near east: a comparitive study. In: Ucko, P.J., Tringham, R., Dimbleby, G.W. (Eds.), Man, Settlement and Urbanism. Duckworth, London, pp. 23–53. Graham, M., 1989. Raramuri residential site structure: an ethnoarchaeological approach to settlement organization. Unpublished Ph.D Dissertation, University of New Mexico, Albuquerque, New Mexico. Hietala, H.J., 1984. Intrasite Spatial Analysis in Archaeology. Cambridge University Press, Cambridge. Hill, J.N., 1970. Broken K Pueblo: Prehistoric Social organization in the American Southwest. Anthropological papers of the University of Arizona 18, University of Arizona Press, Tucson. Hitchcock, R.K., 1987. Sedentism and site structure: organizational change in Kalahari Basarwa residential locations. In: Kent, S. (Ed.), Method and Theory for Activity Area Research. Columbia University Press, New York, pp. 374– 423.

Hodder, I., 1989. This is not an article about material culture as text. Journal of Anthropological Archaeology 8, 250– 269. Hodder, I., 1990. The Domestication of Europe. Basil Blackwell, Oxford. Hunter-Anderson Rosalind, L., 1977. A theoretical approach to the study of house form. In: Binford, L.R. (Ed.), For Theory Building in Archaeology, Essays on Faunal Remains, Aquatic Resources, Spatial Analysis, and Systemic Modeling. Academic Press, New York, pp. 287–317. Kajale, M.D., 1988. Plant economy. In: Dhavalikar, M.K., Sankalia, H.D., Ansari, Z.D. (Eds.), Excavations at Inamgaon, vol. I, Pt. ii. Deccan College Post-Graduate Research Institute, Pune, pp. 727–823. Kelly, R.L., 1992. Mobility/sedentism: concepts, archaeological measures, and effects. Annual Review of Anthropology 21, 43–66. Kent, S., 1984. Analyzing Activity Areas. University of New Mexico Press, Albuquerque. Kent, S., 1992. Studying variability in the archaeological record: an ethnoarchaeological model for distinguishing mobility patterns. American Antiquity 57 (4), 635–660. Killion, T.W., 1990. Cultivation intensity and residential site structure: an ethnoarchaeological examination of peasant agriculture in the Sierra De Los Tuxtlas, Veracruz, Mexico. Latin American Antiquity 1 (3), 191–215. Krishnamurthy, R.V. et al., 1981. Palaeo-climatic influences from the behaviour of radio-carbon dates of carbonates from sand dunes of Rajasthan. Proceedings of the Indian Academy of Sciences 90, 155–160. Kroll, E.M., Price, T.D., 1991. The Interpretation of Archaeological Spatial Patterning. Plenum Press, New York. Moore, H.L., 1986. Space, Text and Gender. Cambridge University Press, Cambridge. Mujumdar, G.G., Rajaguru, S.N., 1965. Comments on soils as environmental and chronological tool. In: Misra, V.N., Mate, M.S. (Eds.), Indian Prehistory—1964. Deccan College Post Graduate Research, Pune, pp. 248–253. Murray, P., 1980. Discard location: the ethnographic data. American Antiquity 45, 490–502. Nash, D.T., Petraglia, M.D. (Eds.), 1987. Natural Formation Processes and the Archaeological Record. BAR, International Series 352, Oxford. OÕConnell, J.F., 1987. Alyawara site structure and its archaeological implications. American Antiquity 52 (1), 74–108. Panja, S., 1996. Mobility strategies, site structure and subsistence-settlement organization: a case study of the Chalcolithic in the middle Bhima valley with special reference to Inamgaon. Unpublished Ph.D Thesis, University of Poona, Pune. Panja, S., 1996b. Mobility strategies, site structure and settlement organization—an actualistic perspective. Man and Environment 21 (1), 58–73. Panja, S., 2000. A re-analysis of houses: the evidence from Inamgaon, Maharashtra. Bulletin of the Deccan College Postgraduate and Research Institute 58, 93–104. Panja, S., 1999. Mobility and Subsistence Strategies at Inamgaon. A Chalcolithic Site in Western India. Asian Perspectives 38 (2), 154–186.

S. Panja / Journal of Anthropological Archaeology 22 (2003) 105–125 Pawankar, S., 1996. Archaeozoology at Inamgaon. Unpublished Ph.D Thesis, University of Poona, Pune. Petraglia, M., 1987. Site Formation processes at the Abri Dufaure: a study of upper palaeolithic rockshelter and hillslope deposits in southwestern France. Unpublished Ph.D. Dissertation, University of New Mexico, Albuquerque, New Mexico. Rafferty, J.E., 1985. The archaeological record on sedentariness: recognition, development, and implications. In: Schiffer, M.B. (Ed.), Advances in Archaeological Method and Theory, 8. Academic Press, New York, pp. 113–155. Rothschild, N.A., Mills, B.J., Ferguson, T.J., Dublin, S., 1993. Abandonment at Zuni farming villages. In: Cameron, C.A., Tomka, S.A. (Eds.), Abandonment of Settlements and Re- gions. Cambridge University Press, Cambridge, pp. 123–137. Sadr, K., 1991. The Development of Nomadism in Ancient Northeast Africa. Pennsylvania Press, Philadelphia. Schiffer, M.B., 1987. Formation Processes of the Archaeological Record. University of New Mexico Press, Albuquerque. Schlanger, S., 1992. Recognizing persistent places in Anasazi settlement systems. In: Rossignol, J., Wandsnider, L. (Eds.), Space, Time and Archaeological Landscapes. Plenum Press, New York, pp. 91–112. Simek, J.F., 1989. Structure and diversity in intrasite spatial analysis. In: Leonard, D.R., Jones, G.T. (Eds.), Quantifying Diversity in Archaeology. Cambridge University Press, Cambridge. Simms, S.R., 1988. The archaeological structure of a Bedouin camp. Journal of Archaeological Science 15, 197–211. Stafford, R., 1985. Hunter-gatherer settlement strategies: a regional perspective on intra-site spatial analysis. Paper presented at the Annual Meeting of the Society for American Archaeology in the Symposium Beyond Ethnoarchaeology: Adaptive Strategies of Forager-Collectors from the Archaeological Record. Denver. Stevenson, M.G., 1982. Toward an understanding of site abandonment behaviour: evidence from historic mining camps in the southwest Yukon. Journal of Anthropological Archaeology 1, 237–265.


Thomas, J., 1991. Rethinking the Neolithic. Cambridge University Press, Cambridge. Wandsnider, L.A., 1987. Natural formation procees experimentation and archaeological analysis. In: Nash, D.T., Petraglia, M.D. (Eds.), Natural Formation Processes and the Archaeological Record. BAR International Series 352, pp. 150–185. Wandsnider L.A., 1989. Long-term land use, formation processes and the structure of the archaeological landscape: a case study from southern Wyoming. Unpublished Ph.D Dissertation, University of New Mexico, Albuquerque, New Mexico. Wandsnider, L.A., 1992. The spatial dimension of time. In: Rossignol, J., Wandsnider, L. (Eds.), Space, Time and Archaeological Landscapes. Plenum Press, New York, pp. 257–282. Wandsnider, L.A., 1996. Describing and comparing archaeological spatial structures. Journal of Archaeological Method and Theory 3 (4), 319–383. Wilk, R., 1983. Little house in the jungle: the causes of variation in house size among modern Kekchi Maya. Journal of Anthropological Archaeology 2, 99–116. Wilseden, L.E., 1982. The study of impacts on archaeological sites. In: Schiffer, M.B. (Ed.), Advances in Archaeological Method and Theory, vol. 5. Academic Press, New York, pp. 51–96. Wood, W.R., Johnson, D.L., 1978. A survey of disturbance processes in archaeological site formation. In: Schiffer, M.B. (Ed.), Advances in Archaeological Method and Theory, vol. 1. Academic Press, New York, pp. 315–381. Wylie, A., 1989. Archaeological cables and tacking: the implications of practice for BernsteinÕs Ôoptions beyond objectivism and relativismÕ. Philosophy of Social Science 19, 1–18. Wylie, A., 1990. The interpretive dilemma. In: Pinsky, V., Wylie, A. (Eds.), Critical Traditions in Contemporary Archaeology. Cambridge University Press, Cambridge, pp. 18–28. Yellen, J., 1977. Archaeological Approaches to the Present. Academic Press, New York.