Settlement Pattern and the Spatial Organization of Subsistence and Mortuary Practices in the

Mesolithic Ganges Valley, North-Central India

Author(s): Umesh C. Chattopadhyaya
Reviewed work(s):
Source: World Archaeology, Vol. 27, No. 3, Hunter-Gatherer Land Use (Feb., 1996), pp. 461-476
Published by: Taylor & Francis, Ltd.
Stable URL: http://www.jstor.org/stable/124936 .
Accessed: 16/10/2012 23:50

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to World
Archaeology.

http://www.jstor.org
Settlement pattern and the spatial
organization of subsistence and
mortuary practices in the
Mesolithic Ganges Valley, North-
Central India

Umesh C. Chattopadhyaya

Abstract

This articleevaluatesa cross-culturalarchaeologicalmodel linkingthe rise of formalcemeteries

amonghunter-gatherers to subsistenceandsettlementpatterns,usingthe Mesolithicof the Ganges
valley as a case study.Faunal data,includingageabledeerteeth, andgraveorientationin relationto
solar variationsuggestthat the sites of Mahadahaand Damdamawere logisticallyorganizedand
residentiallystable. The archaeologicalevidencethus supportsthe Saxe-Goldsteinformulationon
the interrelationshipbetween cemeteriesand corporategrouprightsto crucialresources(aquatic
resourcesin the Gangeticcase). It is suggestedthatthiseconomicappoachshouldbe complemented
by an understandingbasedon culturalbeliefsof past societies.

Keywords

Mesolithic;India;Gangesvalley;cemetery;settlementpattern;faunalremains.

Introduction

An important feature of the Mesolithic in India, as elsewhere, has been the location and
use of ecological settings seldom exploited by humans in the Pleistocene (Allchin and
Allchin 1982). This phenomenon, generally explained in terms of population rise due to
favourable environmental conditions coupled with technological changes, might have led,
in certain circumstances, to competition for critical resources, including land, water and
food, and changes in the settlement pattern. Formal cemeteries, according to some,
correlate with territorial behaviour by corporate groups who want to gain control over
critical resources (Bloch 1975; Chapman 1981; Goldstein 1981; Saxe 1970). An attempt is
made here to evaluate this cross-culturally generated model in relation to settlement

World Archaeology Vol. 27(3): 461-476 Hunter-GathererLand Use

? Routledge 1996 0043-8243
462 Umesh Chattopadhyaya

patterns and land-use strategies among Mesolithic hunter-fisher-gatherers of the Ganges

valley, north-central India.

The emergence of formal cemeteries: theoretical perspectives

Mortuary symbolism associated with formal disposal areas for the dead has been variously
interpreted in anthropology. There have been growing attempts not only in anthropology
but also in archaeology to provide interpretations with a specific economic dimension.
Saxe, for example, used anthropological data to postulate the following hypothesis: 'To
the degree that corporate group rights to use and/or control crucial but restricted resources
are attained and or legitimised by means of lineal descent from the dead (i.e. lineal ties to
ancestors), such groups will maintain formal disposal areas for the exclusive disposal of
their dead, and conversely' (Saxe 1970:119, emphasis added). By formal disposal area is
meant a permanent, specialized, bounded area- a 'cemetery', for example.
Testing this hypothesis with a world-wide sample of thirty societies, Goldstein concludes
that it does not work in both directions. However, she substantiates the converse part of
Saxe's hypothesis in the following words:
If a permanent, specialised bounded area for the exclusive disposal of the group's dead
exists, then it is likely that this represents a corporate group that has rights over the use
and/or control of crucial but restricted resources. This corporate control is most likely to
be attained and/or legitimised by means of lineal descent from the dead, either in terms
of an actual lineage or in the form of a strong, established tradition of the critical
resource passing from parent to offspring.
(Goldstein 1981: 61)
Goldstein further maintains that 'the more structured and formal the disposal area, the
fewer alternative explanations of social organisation apply, and conversely' (1981: 61).
In India the best-known ethnographic example of the relationship between formal
disposal areas for the dead and corporate group rights over critical resources can be found
among the Mundas of the Chhota Nagpur hills of southern Bihar (Hoffman, 1950). Land is
precious among these lineage-based groups and inherited within the family. Each clan has
its own Sasan, or formal disposal area for the dead, situated on one side of the village.
Large stone slabs, diri, erected on the Sasan allow Sasan-diri to be viewed as a territorial
marker: corporate group behaviour finds legitimation from dead ancestors (personal
communication from V.S. Sahay, Department of Social Anthropology, University of
Allahabad). However, it may be mentioned that Sasan-diri have an equally important
emotive role to play in the Munda society, a point I shall return to later.

The Ganges valley

Ecological setting
The central Ganges valley is characterized by tropical dry deciduous (savanna) forest
(Spate and Learmonth 1972). The climate varies considerably between the principal
seasons: winter (November-January), spring and dry summer (February-June) and the
 Subsistence and mortuarypractices in the Mesolithic Ganges valley 463

26?

^----^ V3 '^ - 30'

11^ L1 PC \
Allahabad /

Varanasi [1

- 25'

EASTERN VINDHYAS__ ~

30'

* Epipalaeolithic
;~\ ?, / ==^~* Mesolithic, non-geometric

r--
__^
_\ --/
* Son__^./
A
A
Mesolithic,
Mesolithic,
geometric
geometric (excavated)
High ground
-/)~~ ~~0 - .
50
.km
b . *

Figure I Map showing the distribution of representative prehistoric sites of the central Ganges
valley (after Sharma et al. 1980). DDM = Damdama; MDH = Mahadaha; SNR = Sarai-Nahar Rai.
464 Umesh Chattopadhyaya

monsoon period and autumn (July-October). The average rainfall is about 990 mm and
the effective temperature is as high as 17.2?C. The vegetation is the Phragmites-
Saccharum-Imperata tree and shrub savanna type (Dabadghao and Shankarnarayan
1973). These grassy forests were certainly more widespread in the past, and gave shelter to
a number of herbivores such as different species of deer, rhinoceros and wild boar.
However, the wild fauna has almost been wiped out due to large-scale cultivation. An
important feature of the area is the presence of large numbers of oxbow lakes, the isolated
meanders formed during the southward shifting of the river Ganges during the terminal
Pleistocene and Holocene (Chattopadhyaya 1995). The riparian environments contain
aquatic resources including tortoise and fish, and mammals like swamp deer and buffalo.
Pollen studies in the Ganges valley from the archaeological site of Mahadaha and a deep
profile from Khuilan oxbow lake suggest the presence of grassland, swamp and forest
during the period under consideration (Gupta 1976; Pant and Pant 1980). It has been
suggested that present ecological conditions in the Royal Chitwan National Park,
characterized by rich forest cover and a wide range of wild ungulates, are close to those
prevailing in the area under study during the early/middle Holocene (Chattopadhyaya
1988).

Archaeological investigations
The central plains of 'he Ganges valley are an important, but enigmatic, area in South
Asian prehistory. Archaeological investigations carried out by the University of Alla-
habad over the past twenty-five years have located six Epipalaeolithic and 194 Mesolithic
sites dating from the terminal Pleistocene to mid-Holocene, but no earlier Palaeolithic
sites (Sharma et al. 1980). The spatial distribution of the Mesolithic sites in the Ganges
valley shows two distinct clusters (Fig. 1). For a list of individual sites see Sharma et al.
(1980:123-31).
Most of the Mesolithic sites are close to oxbow lakes and the streams issuing from them
(Pal 1994). They fall into two categories: non-geometric microlithic (172 sites) and
geometric microlithic (twenty-two sites). The sites with geometric microliths are generally
larger. These two site categories should not necessarily be viewed as chronologically
distinct, as has been hypothesized (Sharma et al. 1980). Rather, they might indicate sites of
varying size and function used by the same groups (Chattopadhyaya 1988).

Excavations
The three excavated sites of Sarai-Nahar Rai (SNR), Mahadaha (MDH) and Damdama
(DDM) are among the larger sites of the region, and have yielded geometric microliths and
onsite human burial practices. Sarai-Nahar Rai and Mahadaha are located on oxbow
lakes; Damdama, on the other hand, is on a raised area at the confluence of two branches
of a tributaryof the Pili Nadi that flows into the river Sai (Pal 1992).
Sarai-NaharRai (1,800 m2)is divisible into two strata with an average total depth of 6 cm
(Sharma 1973, 1975). The excavations by the University of Allahabad unearthed from the
lower stratum a total of eleven graves containing fourteen individuals; one grave was a
quadruple burial. The site of Mahadaha (8,000 m2) is divisible into two broad areas: the
 Subsistence and mortuary practices in the Mesolithic Ganges valley 465

TableI Some attributesof the humanburialsfromthe excavatedMesolithicsites in the Ganges

Valley(fromKennedyet al. 1992;Sharma1975;Pal 1992)

Attribute Sarai-NaharRai Mahadaha Damdama

No. of graves 11 28 41
No. of doubleburials 0 2 5
No. of tripleburials 0 0 1
No. of quadrupleburials 1 0 0
No. of individuals 14 30 46
No. of males 9 17 24
No. of females 4 7 17
No. of children 1 3 0
% malesin total sexed adultsand
subadults 69.2 70.8 58.5

lake area towards the east, and the combined cemetery-cum-habitation area and
butchering/dumping complex; the average total thickness of the latter was 60cm. The
cemetery-cum-habitation area contained four occupational layers and produced twenty-
eight in situ graves with thirty individuals, two being double burials. Finally, Damdama
(8750m2) is divisible into ten occupational strata, the average total thickness of the
occupational deposit being 1.05m. The lower nine strata are definitely Mesolithic in
character (Pal 1992). A total of forty-one graves was located from different layers. While
most were single interments, five were double and one contained three individuals.
All three sites provide similar mortuary evidence: individuals were buried in shallow
rectangular graves, generally in an extended position. The presence of earthen tumuli on
the graves was demonstrated in some cases. Barring a few exceptions at Damdama, all the
graves excavated were aligned broadly west-east or east-west (in all such descriptions, the
direction given first indicates the direction of the head). There are always more male than
female burials (Table 1). A few children are represented at Mahadaha and Damdama.
According to the excavators, the presence of bone and antler ornaments in some male
graves and differentiation of grave goods at Mahadaha are indicative of social ranking
(Sharma et al. 1980). At Mahadaha thirty-five pit hearths, both plain and plastered, were
found in the cemetery-cum-habitation area. They contained burnt clay, ash and charred
animal bones and their positions suggest that they were associated with burial rituals. The
occurrence of a large charred bovid skull in one hearth supports this interpretation.
These Mesolithic sites have also yielded burnt plaster floors, charred seeds, animal
bones, and artefacts such as microliths, bone/antler tools, querns, mullers, ring stones,
sling stones, hammer stones and anvils. Microliths comprise retouched blades, points,
awls, lunates, triangles and trapezes. The Middle Ganges valley has no rocks; the nearest
source of chert, chalcedony and other cryptocrystalline raw materials is the middle Son
valley in the eastern Vindhyas, about 160km south of the main Gangetic settlements
(William MacCormack, Berkeley, personal communication). The diminutive size of the
microliths and the nearly exhausted cores at these sites indicate maximum utilization of the
'precious' raw materials (Sharma et al. 1980). The larger stone items like querns and
466 Umesh Chattopadhyaya

Table2 C14 dates from the Mesolithic Gangetic sites. OxA 1647 is an accelerator date; the rest are
conventional.

Site Laboratorynumber Date (BP, uncalibrated)

Sarai-Nahar Rai TF 1104 10,050 ? 110

TF 1356 & 1359 2860 ? 120
Mahadaha BS 136 4010 ? 120
BS 137 2880 +?250
BS 138 3840 ? 130
OxA 1647 6320 +?80

mullers were fashioned mostly on quartzite, the nearest source of which again is the
Vindhyas. It is most likely that such stones were obtained from the Vindhyas through some
kind of exchange network (Allchin and Allchin 1982: 72-3). The Ganges populations may
also have depended on Vindhyan societies for obtaining marriage partners, particularly
during the early phases of occupation (Chattopadhyaya 1988). It is therefore important to
recognize that, in order to facilitate exchange of the above items, Ganges societies had to
ensure economic productivity greater than their own normal requirements.

Chronology
A few conventional radiocarbon dates from Sarai-Nahar Rai and Mahadaha fall either in
the early or the late Holocene (Table 2; cf. Possehl and Rissman 1992). If the early date of
10,050 ?+110 BP from Sarai-Nahar Rai is unacceptable (Agrawal 1982), the late Holocene
dates are also questionable in view of the presence of Bos namadicus remains at the sites;
this species became extinct around the middle Holocene (Badam 1984: 344). To resolve
this problem, a charred animal bone fragment from Mahadaha stratum 3 was accelerator
dated, yielding an age of 6,320 ? 80 BP (OxA-1647). Thus, a middle Holocene date may
tentatively be suggested for Mahadaha. The other two sites may be similar because
detailed morphometric and pathological studies of the human skeletons from all three sites
show close genetic ties between the skeletal series: all the skeletons are thought to derive
from the same macropopulation, chronologically separated only by a few generations
(Kennedy 1984; Kennedy et al. 1992; Lukacs and Pal 1993). Burnt clay and soil samples
from Damdama are currently under analysis for thermoluminescence dating, and the
preliminary results tend to place the antiquity of the site between 7000 and 5000 years BP
(Lukacs and Pal 1993: 749).

The seasonal mobility hypothesis

The first excavation of Sarai-Nahar Rai led Sharma to hypothesize a summer migration of
Vindhyan groups into the Ganges valley in search of food and water, caused by seasonal
drought and local resource depletion in the rocky Vindhyas (Sharma 1973, 1975). He ruled
out any possibility of year-round occupation in the Ganges valley on two main grounds: the
absence of lithic raw materials in the entire Middle Gangetic plains, and the thin
occupational deposits at Sarai-Nahar Rai. Sharma's hypothesis is consistent with Binford's
 Subsistence and mortuarypractices in the Mesolithic Ganges valley 467

percent
XTToTC

I Mahadahalake area (N = 693)

30- .g. Mahadahacemetery and

30" - ^ habitationarea (N = 2420)
|
2.i I;0 . . Damdama1982-84 (N = 1769)
20
:: J|I I : - [[g}[]Sarai-NaharRai (N = 104)

10 E

7- B 0I0 i Ii Q 0 -i

Figure2 Faunalremainsfromthe Gangesvalleysites, basedon percentagesof NISP(= Numberof

IdentifiedSpecimens);conversionto MNI (= MinimumNumberof Individuals)generallyreduces
the tortoisepercentage.A few bones of hare, lizard,etc., are not included.

prediction of nomadism in low latitude regions characterized by high effective tempera-

tures (Binford 1980).
The excavations at Mahadaha and Damdama led subsequent workers to raise doubts
about Sharma's hypothesis (Varma 1981-2; Varma et al. 1985). Varma's arguments in
support of semi-permanent occupation at these two sites centred on (a) the greater
thickness of the cultural deposits and (b) the occurrence of large numbers of heavy,
non-transportable grindstones. Unfortunately, however, the arguments supporting either
view, though useful, are too general to be conclusive. I therefore started a fresh
investigation using the faunal data. In addition, mortuary remains have been used to
address the problem (Chattopadhyaya 1991). This allows subsistence and settlement
patterns to be considered, and their possible connection with onsite human burial practice.

Subsistence and settlement patterns

Faunal remains
Animal bones were recovered from Mahadaha, Damdama and Sarai-Nahar Rai. Analysis
of differential survival and representation of anatomical elements shows not only fairly
good preservation conditions (particularly for the ungulate remains) at Mahadaha and
Damdama, but also that the conventional recovery methods created minimal bias
(Chattopadhyaya 1991). The ungulate samples from these sites can therefore be regarded
as more or less representative, but the same cannot be maintained for Sarai-Nahar Rai
because of the comparatively small sample size.
468 Umesh Chattopadhyaya

Table3 Resourcediversityandnichewidthat the MesolithicGangeticsites

Mahadaha(lake) Mahadaha Damdama Sarai-NaharRai

(cemeterycum
habitation)

Resourcediversity 14 24 23 9
Nichewidth 5.86 5.66 4.55 3.99

All the animal taxa identified by me belong to the wild category. Figure 2 shows the
frequency of the most important taxa from Sarai-Nahar Rai, Mahadaha, and the 1982-4
excavation at Damdama. The high values of resource diversity and niche width at these sites
(Table 3) suggest a generalized economy, with however the specialized hunting of swamp
deer (Cervus duvauceli) and hog deer (Axis procinus), and an intensive exploitation of
aquatic resources (tortoise and at least two species of fish). It is particularlyargued that these
aquatic resourcds were both important and reliable; their underrepresentation on the sites
can be misleading due to taphonomic processes affecting the remains of smaller animals.
Botanical remains from Damdama indicate the presence of edible wild plants including wild
grasses, fat hen (Chenopodium cf. album), and others (Kajale 1990).
The faunal evidence provides an important clue to the settlement pattern, particularlyin
respect of whether the Ganges valley populations were sedentary at some sites, or were
residentially mobile as hypothesized by Sharma (1975) and predicted in Binford's (1980)
model for warm regions. The faunal assemblages from Mahadaha and Damdama contain a
large number of isolated teeth and a handful of ageable mandibles of hog deer and swamp
deer. Teeth provide useful information if the analyses are confined to unworn and slightly
worn molars only. The Mahadaha assemblage includes many such teeth. The ages at which
wear commences for different molars of hog deer and swamp deer are roughly known (Table
4), and they indicate a wide range of killing seasons. For example, taking April and July as
the months of birth of hog deer and swamp deer respectively, the unworn and slightly worn
lower permanent molars indicate a wide range of months for human occupations at
Mahadaha (Fig. 3). But even if we confine the analysis to fully erupted but unworn teeth,
then the mere presence of such lower molars of hog deer indicates August (M1), April (M2)
and December (M3) as the periods of occupation at Mahadaha. Likewise, July and January
are indicated by the molars of swamp deer. This evidence for site occupation in both summer
and winter is sufficient to challenge Sharma's hypothesis of summer visits by Vindhyan
groups.
In this connection, the presence at Mahadaha and Damdama of bandicoot rat (Bandicota
cf. bengalensis), a commensal species, is significant. It has been argued that commensal
animals cannot colonize a habitation site unless food is available throughout the year (Hesse
1979). The fact that the proportion of bandicoot remains increases from the early to the late
phases at both Mahadaha and Damdama may suggest increasingly year-round rather than
seasonal occupations at these sites (Chattopadhyaya 1991).

Mortuary evidence
Further support for this conclusion comes from the orientation of the graves. At all three
excavated sites there is a remarkably consistent west-east (or east-west) pattern. There is,
 Subsistence and mortuarypractices in the Mesolithic Ganges valley 469

Table4 Data on hog deer andswampdeer used for determiningseasonalityof archaeologicalsites

(afterMishra1982;Schaller1967)
Taxon Ages in months at which the permanent molars erupt Period of birth

M 1 M2 M3

Hog deer 4 12 20 late April

Swampdeer 6 12 24 July

however, a significant range of variation within this. A likely hypothesis is that graves were
aligned on sunrise or sunset at the time of death, something which is consistent with both
anthropological and archaeological evidence (Randsborg and Nybo 1986). The range of
solar variation across the horizon between summer and winter solstices changes marginally
from century to century (Astronomical Almanac 1988), and its present value at Mahadaha
and Damdama is approximately 53?. This figure was higher in the Holocene, but never
exceeded 56? in the mid-Holocene epoch (for details of these astronomical calculations,
see Chattopadhyaya 1991: 106-7). This means that at the three Gangetic sites, sunrise
during the summer and winter solstices would have occurred at angles of 62? and 118?from
the azimuth respectively. Similarly, sunset during the summer and winter solstices would
have occurred at angles of 298? and 242? respectively.
Seventy-four graves from the three sites have broadly west-east (or east-west)
orientation, but measurement of the angles from the azimuth was possible only in fifty-five
cases - ten from Sarai-Nahar Rai, twenty-four from Mahadaha, and twenty-one from
Damdama. For measuring these angles, the reference directions in the plans of in situ
burials were used. Fig. 4 shows the percentages of grave orientations in 5? intervals from
60? to 120?(with direction of sunrise as the possible basis for interment), corresponding to
240? to 300? (direction of sunset). These constitute the total range within which the burials
are expected to lie if both the solar orientation hypothesis and the assumption that deaths
were distributed through the year are correct. If the basis of interment was sunrise, then

Jan I Feb IMarl Apr 1Mayi JunI Jul IAug 1 Sep I Oct Nov Dec

HOG DEER - _.
Axis procinus

SWAMP DEER """".

Cervus duvauceli
= ?~~~~~~~~~~~~~~c~

M...... M2 --- M3
Figure 3 Seasons of death of hog deer and swamp deer at Mahadaha, based on fully erupted but
unwornand slightlyworn lower permanentmolars, using the dental crown height measurement
method.
470 Umesh Chattopadhyaya

WINTER HALF SUMMER HALF (basis of

------
|--"- -*14 | minterment:
ANGLE FROMTHE AZIMUTH sunset)
2
4045? 245--50 2.I250 15-60? 260-650 26 -700 270-75v 275-80 280-85 285-900 20-95 295-300 300-050
40

30- graves SNR(N= 10)

|~ . ti MDH (N 24)
20_- | DDM(N=21)

60-65? 65-70? 70-75? 75-80? 80-85> 85-90? 90-95? 95-1000 100-05? 105-10? 110-15? 115-20? 120-25?

ANGLE FROM THE AZIMUTH (basis of

interment:
SUMMER HALF WINTER HALF sunrise)
Figure 4 Histogram showing grave orientations in relation to the azimuth.

orientations around the ranges 60o--700/2400-250?and 110?-120?/290?-300?would imply

burial activities in summer and winter, respectively. The converse would be indicated if the
basis of interment was sunset. However, there may be a number of sources of error,
involving (i) the interment, (ii) the reference direction in the original drawings on which
the measurement of individual grave orientations are based, (iii) the identification of the
mean axis of graves/skeletons, and the actual measurement of the angle (the latter,
however, would not exceed 30', which is the least count of the measuring device), and (iv)
minor variations in geomagnetic North with time. Considering all these possibilities, a
marginal error may be expected in the computation of angles from the azimuth of grave
orientations and hence in the plotting of the histogram (Fig. 4). Eighty-three out of a total
of ninety individuals (in seventy-four graves) from the three sites are oriented broadly
west-east or east-west, and most of the fifty-five grave orientations that could be precisely
measured fall broadly within the calculated range of the annual solar path across the
horizon (Fig. 4). A strong correlation is therefore indicated between body/grave
orientation and sunrise/set.
Although one cannot be sure whether the basis of interment was sunrise or sunset, the
following inferences can be drawn from Fig. 4. (a) The pattern at Sarai-Nahar Rai implies
burial activity and hence site occupation in either the summer or the winter half of the year;
this is, however, inconclusive given the small sample size. (b) The even distribution at
Mahadaha suggests a more or less year-round pattern of occupation at the site. (c) The
Damdama evidence seems to follow the Mahadaha, rather than the Sarai-Nahar Rai,
pattern. This analysis provides additional evidence for residential stability at Mahadaha
 Subsistence and mortuarypractices in the Mesolithic Ganges valley 471

and Damdama. In both these cases, however, deaths were clearly more numerous during
autumn/spring since a greater number of graves are found aligned close to solar paths
around the equinoxes. This concentration may be explained as follows. It is likely that
during summer and winter a substantial part of the male population was engaged in
sex-specific activities, such as hunting or trading in distant areas. If such males died at a
considerable distance from the home-base, their chances of being buried at the site would
be considerably reduced. However, late spring is the period when hog deer move on to the
riverine plains near the settlements (Mishra 1982), and in late summer and autumn it is not
particularly feasible to travel long distances due to the monsoon conditions. During these
seasons more of the group might well be concentrated at the base camps.
One individual at Mahadaha, a sub-adult male, was buried in an atypical manner. The
positions of hands and feet show that they were tied together. The individual was denied
the usual extended position, and the direction of the face too deviates from the normal
pattern (Pal 1985). The explanation suggested here is that the person might have met his
death away from the site and that rigor mortis could have set in before the body's arrival in
the area of disposal. A number of such atypical cases, associated with male individuals, are
noted from Mahadaha and Damdama (Chattopadhyaya 1991:102-3; Pal 1992). Similar
evidence from a number of prehistoric sites in the New World (J.A. Brown, personal
communication) lends considerable support to the hypothesis that the mobility pattern
was logistically organized.

Skeletal element frequencies and site function

A logistic mobility pattern fits the archaeological evidence better than either residential
mobility or the kind of sedentism one normally associates with agriculturalists. It is
possible that some of the numerous small, non-geometric microlithic sites located in
riparian, grassland and forest environments were short-term camps used by task groups
from the larger permanent sites. Unfortunately, none of these small sites has so far been
excavated. Future studies will examine their possible functions through a study of skeletal
element frequencies, using the Percent Minimum Anatomical Units method (%MAU: for
details, see Binford 1978). Meanwhile, this method has been used on the swamp deer and
hog deer assemblages from Mahadaha and Damdama to see if any light could be shed on
site function and the logistical use of the landscape. The pattern of variation in element
frequencies might reveal the economic activities of hunters, i.e. whether the site was a
hunting camp, a kill-butchery site or a base camp.
Results from Mahadaha and Damdama are plotted in Fig. 5, which shows that these sites
generally resemble each other. The low representation of the vertebral column of the two
species at both sites reduces the possibility that they were seasonal hunting camps, such as
the Danish Mesolithic site of Ringkloster (Rowley-Conwy 1993). Again, the pattern at the
Gangetic sites is substantially different from those obtained from kill-butchery sites
described by Binford (1978). If the Gangetic sites were base camps, one would expect
higher representation of body parts carrying the most meat, such as upper limb bones,
pelvis, etc. However, it is the tarsal bones (astragalus and calcaneum) that are most
common, followed by metapodials and distal humeri; femur and pelvis, associated with
 472 Umesh Chattopadhyaya

20 40 60 80 100 20 40 60 80 100
mandible .---. '- ' mandible
maxilla - maxilla
'
atlas atlas
axis - axis
cervical 3-7 - K cervical 3-7
thoracic - thoracic
lumbar-- lumbar
scapula- scapula -
P humerus - _ P humerus
D humerus - D humerus
I radius/ulna - Pradius/ulna -

D radius - D radius -
pelvis - pelvis
P femur - P femur -
D fmur - femur
f r-
P tibia - _ _ P tibia -
D tibia - = D tibia -
- astragalus -
astragalus
calcaneum - calcaneum -alaeu
P m/carpal - Pm/carpal -
D m/carpal - D m/carpal -
P /tarsal - m/tarsal P -
D m/tarsal - D m/tarsal -
phalanx 1 - phalanx 1 -
phalanx 2 - phalanx 2 -
phalanx 3 - phalanx 3 -

DAMDAMA MAHADAHA

-?--e?-- Swamp Deer --- Hog Deer

Figure 5 Swamp deer and hog deer skeletal element representations at Mahadaha and Damdama,
using the %MAU method (Binford 1978).

high meat yield, are not well represented. It appears that the animals were hunted in
distant locations where certain parts of the body were consumed and the bulk of the carcass
was discarded; it was mostly the limb bones and antlers (often together with skulls) which
were transported to the permanent sites. Swamp deer are only slightly smaller than
European red deer, but hog deer are only slightly heavier than goat. Had the latter species
of deer been killed close to the site, the whole body could easily have been transported to
the site. Although humeri of both these species are well represented, the under-
representation of femur and pelvis is disturbing. So the above analysis shows that the
results from Mahadaha and Damdama do not truly match any of the three ideal site
categories, i.e. hunting camp, kill-butchery site and base camp.
 Subsistence and mortuary practices in the Mesolithic Ganges valley 473

It is possible that classification of archaeological sites by ethnographic analogy may be

inappropriate. Sites like Mahadaha and Damdama where numerous activities were carried
out, including those involving burials and ritual hearths containing charred animal bones,
might have been much more complex than the ethnoarchaeological examples described by
Binford (1978). It is also possible that the taphonomic models used hitherto may be
oversimplified, and that such processes are involved in the relative rarity of femur and
pelvis - despite the fact that neither differential recovery nor carnivore activities can be
shown to have exerted a major influence (Chattopadhyaya 1991).
An assumption made so far is that high meat-yielding parts originating at kill-butchery
sites or hunting camps necessarily end up at base camps. But there is the possibility that, in
spite of being primarily a base camp, a site might also function as a centre of secondary
distribution of certain prized items, including faunal packages, for exchange. This may be
applicable to the Gangetic sites where stone raw materials constituted a critical resource
which might have been procured in exchange for local products including foodstuffs. This
finds some support from the fact that most of the tortoise remains at the excavated
Gangetic sites are charred shells; the skeletal remains, more directly associated with meat,
are heavily under-represented. Food resources may, in other words, have been produced
in excess of normal requirements, to facilitate the acquisition of exotic stone, a critical
resource for Gangetic societies. As preservation at Mahadaha and Damdama is fairly
good, this peculiar aspect of bone representation cannot be explained by taphonomic
factors alone. However, this is just a hypothesis that needs to be tested against data from
other sites lying around the geographical boundaries between the Ganges valley and the
Vindhyas.

Overview and discussion

The above discussions suggest a pattern of logistic mobility in the Mesolithic Ganges
valley, with some sites with rich aquatic resource potential being inhabited on a more or
less year-round basis. The population would have varied through the year as people
undertook periodic journeys across the wider landscape for foraging activities including
hunting. Aquatic resources near the permanent sites, particularly tortoise and fish, were
the most dependable and predictable resources, and could have been exploited without
recourse to arduous trips.
At Mahadaha, burial activities spread through four cultural phases were however
confined to a single spatially distinct area. This area (405 m2) constitutes about 5 per cent of
the total activity area and forms an exclusive, well defined location at the centre of the site.
Certainly this cemetery was (a) the scene of year-round activities, as elaborated above, (b)
specialized in terms of ritual, and (c) bounded (being at the centre of the settlement). ]It
thus qualifies as 'formal'. Following the Saxe-Goldstein formulation discussed above, it
can be said that Mahadaha provides fairly good evidence for the existence of corporate
group rights over productive, reliable but restricted resources legitimized by lineal descent
from the dead.
The question, then, is what might these crucial resources have been. As discussed
earlier, two species of deer and aquatic resources like tortoise and fish constituted the
474 Umesh Chattopadhyaya

primary resources. Numerous lakes in the central Ganges valley might have favoured the
aggregation of swamp deer; as they do not make large-scale migrations, their location and
hunting might have been predictable at least in certain seasons. However, the aquatic
resources like tortoise and fish constitute not only a very good source of many essential
nutrients, but are, at the same time, highly productive and reliable resources. It seems
most probable that growing population in the Ganges valley during the Mesolithic
(Chattopadhyaya 1988) led to competition for aquatic resources. This is supported to some
extent by the evidence of warfare from certain human skeletons of Sarai-Nahar Rai
(Chattopadhyaya 1991; Sharma 1973). Under the circumstances, it was perhaps necessary
for people to act as corporate groups and to assert land claims by ritually signifying their
claim through mortuary symbolism.
Although there is a link between hunter-gatherer territoriality, corporate group descent
and formal cemeteries, the Saxe-Goldstein formulation, with its exclusive emphasis on the
economic dimension, presents only a partial picture of social reality. Mesolithic Gangetic
society is reflected in material symbolism incorporating both material (economic) and
non-material dimensions. These dimensions are 'written' in space through formal
cemeteries at some of the Gangetic sites. An economic explanation for the emergence of
formal cemeteries appears well founded. The danger, however, with exclusively economic
explanations lies in the fact that they neither explain the whole archaeological record, nor
do they address important non-economic aspects of society, whether this be today or 6,000
years ago. This may mislead policy makers currently concerned with the welfare of tribal
societies. For example, the present trend in India with regard to rehabilitation of tribal
groups has raised some sensitive questions (Rao 1994). Physical rehabilitation is not so
much the bone of contention, as this can open avenues for fresh economic opportunities.
The main objection to such programmes, for example from the Mundas of Chhota Nagpur
region mentioned above, centres not on physical, but on cultural rehabilitation (V.S.
Sahay, personal communication). The question of great significance to such societies is
how to transport the Sasan-diris, which are intimately connected with social wellbeing,
group identity and continuity. If archaeology is not to be divorced from the present,
theoretical formulations on the rise of formal cemeteries should look forward to cultural
and social implications as well as back to their origins.

Acknowledgements

I must thank Dr Peter Rowley-Conwy for his encouraging comments on earlier drafts of
this paper. I also thank the authorities of the Department of Ancient History, Culture and
Archaeology, University of Allahabad, particularly Professor V. D. Misra, for giving me
access to various kinds of database. I am also grateful to other members of the department,
A. Malaviya, L. K. Tiwari and A. P. Khatri for helping me in various ways. I also thank
Professor A. R. N. Srivastava and Dr V. S. Sahay of the Department of Social
Anthropology, University of Allahabad, for helping me with their expertise; I found Dr
Sahay's discussions on the Mundas particularly useful. I am thankful to the authorities of
the Mehta Research Institute, Allahabad, especially Dr Seshadi and Mr S. Varma, for
helping me with E-mail. Indrani was helpful, as always, in going through this paper in its
 Subsistence and mortuary practices in the Mesolithic Ganges valley 475

various stages of preparation. While several persons helped me towards the final produc-
tion of this paper, I alone take responsibility for its shortcomings.

References

Agrawal, D. P. 1982. The Archaeology of India. London: Curzon Press.

Allchin, B. and Allchin, R. 1982. The Rise of Civilization in India and Pakistan. Cambridge: Cam-
bridge University Press.
Astronomical Almanac 1988. (Preface by C.K. Roberts and A. Boksenberg). London: HMSO.
Badam, G. L. 1984. Holocene faunal material from India with special reference to domesticated
animals. In Animals and Archaeology 3: Early Herders and their Flocks (eds J. Clutton-Brock and
C. Grigson). BAR International Series 202: 339-53.
Binford, L. R. 1978. Nunamiut Ethnoarchaeology. New York: Academic Press.
Binford, L. R. 1980. Willow smoke and dogs' tails: hunter-gatherer settlement systems and arch-
aeological site formation. American Antiquity, 45: 4-20.
Bloch, M. 1975. Property and the end of affinity. In Marxist Analyses and Social Anthropology (ed.
M. Bloch). London: Malaby Press, pp. 203-28.
Chapman, R. 1981. The emergence of formal disposal areas and the 'problem' of megalithic tombs
in prehistoric Europe. In The Archaeology of Death (eds R. Chapman, I. Kinnes and K. Ran-
dsborg). Cambridge: Cambridge University Press, pp. 71-81.
Chattopadhyaya, U. C. 1988. Subsistence variability and complex social formations in prehistoric
Ganga valley: problems and prospect. Man and Environment, 12:135-52.
Chattopadhyaya, U. C. 1991. A Study of Subsistence and Settlement Patterns during the Late Pre-
history of Northcentral India. Unpublished PhD dissertation. University of Cambridge.
Chattopadhyaya, U. C. 1995. Application of satellite remote sensing to the study of mid-Ganga
palaeochannels and its archaeological implications. Project report submitted to the National
Remote Sensing Agency, Department of Space, Hyderabad.
Dabadghao, P. M. and Shankarnarayan, K. A. 1973. The Grass Cover of India. New Delhi: Indian
Council of Agricultural Research.
Goldstein, L. 1981. One-dimensional archaeology and multi-dimensional people: spatial organis-
ation and mortuary analysis. In The Archaeology of Death (eds. R. Chapman, I. Kinnes and K.
Randsborg). Cambridge: Cambridge University Press, pp. 53-69.
Gupta, H. P. 1976. Holocene palynology from meander lake in the Ganga valley, district Pra-
tapgarh, U.P. The Palaeobotanist, 25: 109-19.
Hesse, B. 1979. Rodent remains and sedentism in the Neolithic: evidence from Tepe Ganj Dareh,
Western Iran. Journal of Mammalogy, 60(4): 856-57.
Hoffman, J. 1950. Encyclopedia Mundarica. Patna: Government Printing Press (Reprint).
Kajale, M. D. 1990. Some initial observations on palaeobotanical evidence for Mesolithic plant
economy from excavations at Damdama, Pratapgarh, Uttar Pradesh. In Adaptation and Other
Essays (eds N. C. Ghosh and S. Chakrabarti). Santiniketan: Visva-Bharati, pp. 98-102.
Kennedy, K. A. R. 1984. Biological adaptations and affinities of Mesolithic South Asians. In The
People of South Asia (ed. J. R. Lukacs). New York: Plenum, pp. 29-57.
Kennedy, K. A. R., Lukacs, J. R., Pastor, R. F., Johnston, T. L., Lovell, N. C., Pal, J. N., Hemp-
hill, B. E. and Burrow, C. B. 1992. Human Skeletal Remains from Mahadaha: A Gangetic Mesoli-
thic Site Ithaca, NY: Cornell University, South Asian Occasional Papers and Theses No.11.
476 Umesh Chattopadhyaya

Lukacs, J. R. and Pal, J. N. 1993. Mesolithic subsistence in north India: inferences from the dental
attributes. CurrentAnthropology, 34(5): 745-65.
Mishra, H. M. 1982. The Ecology and Behaviour of Chital (Axis axis) in the Royal Chitwan National
Park, Nepal (With Comparative Studies of Hog Deer (Axis porcinus), Sambar (Cervus unicolor) and
Barking Deer (Muntiachus muntjak). Unpublished PhD dissertation. University of Edinburgh.
Pal, J. N. 1985. Some new light on the Mesolithic burial practices of the Ganga valley: evidence from
Mahadaha, Pratapgarh, Uttar Pradesh. Man and Environment, 9: 28-37.
Pal, J. N. 1992. Mesolithic human burials in the Ganga plain, north India. Man and Environment,
17(2): 35-44.
Pal, J. N. 1994. Mesolithic settlements in the Ganga plain. Man and Environment, 19: 91-101.
Pant, D. D. and Pant, R. 1980. Preliminary observations on pollen flora of Chopani-Mando
(Vindhyas) and Mahadaha (Ganga valley). History and Archaeology, 1: 229-30.
Possehl, G. L. and Rissman, P. C. 1992. The chronology of prehistoric India: from earliest times to
the Iron Age. In Chronologies in Old World Archaeology 1: India (ed. R. W. Ehrich). Chicago,
Chicago University Press, pp. 465-90.
Randsborg, K. and Nybo, C. 1986. The coffin and the sun. Acta Archaeologica, 55:161-84.
Rao, N. 1994. Subsistence and associated settlement patterns in Central India: an ethnoarchaeologi-
cal analysis. In Living Traditions: Studies in the Ethnoarchaeology of South Asia (ed. B. Allchin).
New Delhi: Oxford & IBH, pp. 143-68.
Rowley-Conwy, P. 1993. Season and reason: the case for a regional interpretation of Mesolithic
settlement patterns. In Hunting and Animal Exploitation in the Later Palaeolithic and Mesolithic of
Eurasia (eds G. L. Peterkin, H. Bricker and P. Mellars). Archaeological Papers of the American
Anthropological Association 4:179-88.
Saxe, A. A. 1970. Social Dimensions of Mortuary Practices. Unpublished PhD dissertation.
University of Michigan.
Sharma, G. R. 1973. Mesolithic lake cultures in the Ganga valley. Proceedings of the Prehistoric
Society, 39: 129-46.
Sharma, G. R. 1975. Seasonal migrations and Mesolithic lake cultures of the Ganga valley. In K. C.
Chattopadhyaya Memorial Volume. Allahabad: University of Allahabad, pp. 1-20.
Sharma, G. R., Misra, V. D., Mandal, D., Misra, B. B. and Pal, J. N. 1980. Beginnings of
Agriculture. Allahabad: Abinash Prakashan.
Spate, O. H. K. and Learmonth, A. T. A. 1972. India and Pakistan: A General and Regional
Geography. London: Methuen.
Varma, R. K. 1981-2. The Mesolithic cultures of India. Puratattva, 13-14: 27-36.
Varma, R. K., Misra, V. D., Pandey, J. N. and Pal, J. N. 1985. A preliminary report on the
excavations at Damdama (1982-84). Man and Environment, 9: 45-65.