reconsidering south indian ashmounds
DESCRIPTION
essayTRANSCRIPT
Journal of Anthropological Archaeology 23 (2004) 309–330
www.elsevier.com/locate/jaa
Landscape, monumental architecture, and ritual:a reconsideration of the South Indian ashmounds
Peter G. Johansen*
Department of Anthropology, University of Chicago, 1126 East 59th Street, Chicago, IL 60637
Received 1 March 2004; revised 30 April 2004
Abstract
During the South Indian Neolithic period (3000–1200 BC), the agro-pastoral inhabitants of the South Deccan/North
Dharwar region constructed large mounded features by heaping and burning accumulations of cattle dung. These �ash-mound� features were comprised of a myriad of variegated strata of ash, vitrified dung, and other culturally modified sed-
iments, many of which reached monumental proportions. Ashmounds have been the subject of considerable debate since
coming to the attention of scholars in the early 19th century. Current debate has centered largely on the function and
spatial context of these features in relation to Neolithic settlement. This article examines the South Indian ashmounds
as monumental forms of architecture and the loci of ritual and ceremonial activity within the context of Neolithic
agro-pastoral landscape production. By situating ashmound construction within the social rhythm of cattle pastoralism
and carefully examining the emplotment, depositional histories, and post-Neolithic afterlives of these unique features this
paper argues that social practices likely originating in quotidian activities were gradually transformed into regular, public
ceremonial activities producing monumental forms, relating and reinforcing socio-symbolically charged information.
� 2004 Elsevier Inc. All rights reserved.
Keywords: Ashmounds; South Indian Neolithic; Landscape; Monumentality; Ritual; Social memory
The Indian �ashmound problem� has been the subject
of discussion and debate in South Asian archaeology for
more than 150 years. Ashmounds are large mounded
features comprised of stratified deposits of decompos-
ing, burned and vitrified cow dung and other culturally
modified soils bearing a variety of artifacts. Constructed
primarily during the South Indian Neolithic Period (cir-
ca 3000–1200 BC), these features vary greatly in size
with recorded surface areas ranging from 28m2 to as
much as 4951m2 and heights from 1.5 to 10m. To date,
more than 100 ashmound sites have been documented
0278-4165/$ - see front matter � 2004 Elsevier Inc. All rights reserve
doi:10.1016/j.jaa.2004.05.003
* Corresponding author.
E-mail address: [email protected].
within the South Deccan/North Dharwar region of
southern India (Paddayya, 2001), yet only a small num-
ber have been subjected to systematic archaeological
investigation (Fig. 1).
The �ashmound problem� refers to the longstanding
and dynamic debate surrounding the temporal and cau-
sal origins of these archaeological features. Since their
�rediscovery� in the early 19th century, most analyses
have been directed primarily towards understanding
the function and formation of this unique class of mate-
rial remains. Explanations have ranged from local ac-
counts such as those attributing specific ashmounds to
the remains of the monkey-king Vali, demon rakshasas
or mass human immolation, to functional interpreta-
tions such as refuse dumps, cattle pens or the location
d.
Fig. 1. Locations of South Indian Ashmound Sites for which published data are available.
310 P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330
of iron smelting activities (Allchin, 1963; Paddayya,
1991; Rami Reddy, 1990). Much recent controversy ap-
pears to originate from debates over cultural formation
processes and the relation of ashmounds to Neolithic
domestic settlements and sedentism (Allchin, 1963;
Korisettar et al., 2002; Paddayya, 2001).
This article posits a re-evaluation of the �ashmound
problem� employing an interpretive strategy that ex-
plores ashmounds as both monumental forms of
architecture and the location of ritual and ceremonial
activity within a Neolithic agro-pastoral landscape. Cul-
tural landscapes are spatial and temporal fields of action
in which material and conceptual contexts are
constructed and negotiated through the processual artic-
ulation of social action, structure and the physical envi-
ronment (Lycett, 2001; Morrison, n.d.; Smith, 2003). As
unique and important places within a Neolithic South
Indian landscape, ashmounds are examined, within the
context of available data, from a range of spatial, tem-
poral, and behavioral scales and contexts. While the
argument presented here is focused on enabling a clearer
understanding of ashmound features during the South
Indian Neolithic period, it is intended to contribute to
a broader anthropological discussion of archaeological
approaches to past historical processes involving the
production of cultural landscapes, monumentality, and
ritual architecture.
This article begins with a discussion of the South In-
dian �ashmound problem� followed by an examination of
the depositional structure, form and location of ash-
mound features. Ashmounds are then considered within
a Neolithic agro-pastoral landscape by examining regio-
nal archaeological data from which inferences on past
land-use, economy, and lifeway can be made. This is fol-
lowed by an analysis of ashmounds employing several
visual variables of perception to demonstrate their mon-
umentality and potential as socio-symbolic media in a
Neolithic South Indian cultural landscape. Ashmounds
are then examined using a number of variables of rit-
ual architecture to explore the behavioral implications
P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330 311
involved in their construction, use and maintenance.
Finally, an examination of the continued use of ashmo-
unds as monumental and ritual places in the production
of cultural landscapes during the subsequent Iron Age
(1200–400 BC) demonstrates both subtle and dramatic
shifts of practice and meaning as ashmounds were aban-
doned, reoccupied, reused, reinterpreted, and recon-
structed with vestiges of prior meanings surviving yet
often in greatly altered ways.
The ashmound problem
The �ashmound problem� has deep roots in the his-
tory of South Asian archaeology extending from the
early 19th century to the present day. The main dimen-
sions of the �ashmound problem,� as it is referred in
South Indian archaeological literature, center on debate
over the reasons for and dating of their construction and
use. A detailed treatment of its history is beyond the
scope of this discussion but excellent summaries are
found in Allchin (1963) and Paddayya (1991). In brief,
19th century exploration of ashmounds in the region
of present-day northeastern Karnataka and western
Andhra Pradesh by colonial surveyors and administra-
tors may be characterized as a debate on the historical
origins and cultural formation processes of ashmound
features. Early hypotheses attributing their formation
to natural geological processes were soon dispelled when
stratified cultural materials (ceramics, lithics, and fauna)
were found in early excavations (Newbold, 1843, pp.
129–131). The debate then centered on functional inter-
pretations, such as that ashmound deposits were ancient
or medieval funeral pyres (Newbold, 1843; Sewell, 1899),
the result of industrial activities (lime, brick, glass or
gold working) or accidentally burned accumulations of
cattle dung (Bruce Foote, 1979).
Bruce Foote�s extensive work on the geology and
archaeology of the region during the late 19th century
led to the discovery and documentation of numerous
new ashmound sites and to two crucial observations
on their form and location. His first observation was
the siliceous content of ashmound matrixes including
distinct traces of straw in the deposits (Bruce Foote,
1916; 1979, pp. 92–95). This led Bruce Foote to conclude
that ashmound matrixes were composed primarily of
fired cattle dung. This was later substantiated by two
independent chemical analyses confirming the high silica
content of specimens taken from the Wandali ashmound
(53.1–66.2%) (Bruce Foote, 1979, p. 95; Munn, 1921,
p. 7). The second observation was the discovery and
identification of many Neolithic objects, such as ground
stone celts, �mealing stones,� �rubbing stones,� and pot-
tery during the course of surface survey and excavations
at many ashmound sites (Bruce Foote, 1979, pp. 79–91).
These observations led to two related conclusions; that
the construction of ashmounds took place during the
Neolithic period and that many ashmounds were sur-
rounded by significant scatters of occupational debris.
Despite these findings, both the age and formation
postulates of the Bruce Foote hypothesis were chal-
lenged in the 20th century. Speculative conclusions, la-
ter largely rejected on practical and empirical grounds,
were made by both Woolley (1940) and Yazdani (1936)
who sought to explain the accumulation of such large
quantities of dung as fuel for gold or iron work-
ing activities. Despite the repeated rejection of these
hypotheses by chemical and technical analyses (e.g.,
Zeuner, 1960), this type of explanation has continued
to resurface periodically throughout the remainder of
the century (e.g., Rami Reddy, 1976, 1990; Sundara,
1971). Rami Reddy (1976) has also challenged the Neo-
lithic dating of the ashmounds based on the presence of
two small iron objects and significant deposits of Iron
Age ceramics from the upper levels of his excavations
at the site of Palavoy.
Since the 1950�s, survey and excavation by a number
of scholars (e.g., Allchin, 1961, 1963; Korisettar et al.,
2002; Mujumdar and Rajaguru, 1966; Paddayya, 1973,
1991, 1998; Rami Reddy, 1976; Sundara, 1971) have ex-
panded both the archaeological database and the param-
eters of the ashmound debate. However, with few
exceptions (e.g., Rami Reddy, 1976; Sundara, 1971),
the basic tenets of the Bruce Foote hypothesis remain
firmly established. Allchin�s (1961, 1963) survey and exca-vations at the site of Utnur has led him to conclude that
ashmounds were the remains of cattle pens which had
been regularly and perhaps ritually burned over the
course of their many years of use. Allchin�s (1963) conclu-sions were based on the presence of regular lines of post-
holes in the earliest layers followed by a berm of dung
built around the periphery of the Utnur ashmound.
Paddayya (1991) posited that these accumulations of
dung and subsequent burning were likely the result of
the efforts of the Neolithic inhabitants of adjacent settle-
ments to keep their communities clean of the vermin
associated with animal fecal matter. Following the large
horizontal excavations at Budihal-S, Paddayya (1998,
2001) argues that ashmounds were Neolithic dung refuse
piles appended to cattle pens located within pastoral vil-
lage sites. He agrees with Allchin�s assignment of a pos-
sible ritual function for the ashmounds although specific
details beyond his consideration of the cyclical and epi-
sodic burning of the dung are not offered (Paddayya,
1973, 1991). Based on the result of decades of survey
and excavation Paddayya (1998, 2001) argues that ash-
mounds are central features located within sedentary
Neolithic settlements.
Recent field reconnaissance, surface and subsurface
sampling of a number of Neolithic sites by Korisettar
et al. (2002) has led this group of scholars to argue that
ashmounds are found within a range of sites related to
312 P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330
Neolithic pastoral activities, yet none of which should be
considered permanent, year-round settlements. They base
these conclusions (1) on the low densities or absence of
occupational debris surrounding many ashmounds, (2)
environmental and topographical similarities of non-ash-
mound settlement sites vs. variety in ashmound site lo-
cales, and (3) the contrast of thicker and more extensive
archaeological deposits in non-ashmound settlements
with the thinner occupational deposits at ashmound sites.
Ashmound deposit formation
Ashmounds are large mounded features comprised of
stratified deposits of vitrified, carbonized, and decom-
posing cow dung mixed with layers and lenses of other
culturally modified soils. Most layers contain a variety
of Neolithic artifact types, primarily lithic material, pot-
tery sherds, and faunal remains. Paddayya (1991,
1993a,b), Rami Reddy (1990, 1976) and to a limited de-
gree Allchin (1963) all point to the fact that in virtually
every case in which the areas surrounding ashmound
sites have been subjected to even the most cursory at-
tempts at surface survey, these mounded features are
generally at the center of dense scatters of occupational
debris (cf., Korisettar et al., 2002). At Budihal-S, the
only large scale horizontal excavation of an occupa-
tional zone surrounding ashmound features, Paddayya
(1993b) uncovered the remains of 10 circular house
floors, a large butchering floor, hearths, and other
domestic features. Based on this evidence it may be con-
cluded that ashmound deposits are generally situated
within the archaeological remains of Neolithic settle-
ments. The size, duration, and periodicity of occupation
at many of the under explored sites remains an open
question awaiting more systematic research.
Ashmounds vary considerably in terms of vertical and
horizontal dimensions, the result of both Neolithic con-
struction and post-Neolithic impacts. Bruce Foote
(1916) and later Allchin (1963) noted two categories of
ashmound remains; large mounded features with verti-
cally extensive dimensions and lower, flatter mounds of-
ten characterized by a �vallum� or berm of ashmound
material surrounding their perimeter.1 Many of the for-
mer category of ashmounds consist of upper and lower
sections.2 Two of the most extensively excavated ash-
mound sites; Budihal-S (Paddayya, 1998) and Utnur3
1 Bruce Foote�s (1916, p. 91) cinder mounds and cinder
camps.2 e.g., Budihal-S, Gadiganuru, Kupgal, and Utnur.3 Despite the presence of 2m of recently deposited sediments
on the upper mound at Utnur, examination of the sections and
map from Allchin�s (1961) report documents differences in
deposit depths and paleo-surface topography of more than 1m
between the upper and lower sections of the mound.
(Allchin, 1961) have demonstrated the existence of upper
and lower ashmounds sections.Atboth sites the lower sec-
tions consist of large flat open areas with layers of burned
and decomposing dung and constructed barriers along
their perimeters. Allchin (1961, 1963) and Paddayya
(1998) each conclude that these are the remains of Neo-
lithic cattle pen enclosures. A stock enclosure has also
been inferred from the excavated remains of one of the
Halikallu ashmounds originally designated a �cindercamp� by Bruce Foote (1979; Krishna Sastry, 1979).
Paddayya (1991, p. 590) posits that the regular pres-
ence of occupational debris (i.e., lithics, pottery, and fau-
nal remains) throughout all of the dung ash deposits in
the upper sections of ashmounds indicates that dung
was deposited as secondary refuse with cultural material
adhering to it as it was removed from its initial point of
deposition by the cattle. In the profiles of the ashy and vit-
rified deposits at Budihal-S, he has observed the outlines
of individual piling episodes (Paddayya, 1991, p. 587).
Stratigraphic profiles from Budihal-S, Thanmandi Than-
da, Wandali, Kudatini, and Kupgal all demonstrate the
vertical and horizontal heterogeneity of the ashmound
deposits (Fig. 2). There are layers comprised entirely of
small lenses of soft dung ash intermixed with deposits
of grey culturally modified soils, as well as large layers
of vitrified and decomposed dung that are horizontally
and vertically discontinuous (Paddayya, 1993a, p. 79).
The powdered ash layers at many ashmound sites indi-
cate multiple burning episodes, while the analysis of the
vitrified layers (Mujumdar and Rajaguru, 1966; Zeuner,
1960) demonstrates the occurrence of large single epi-
sodes of burning at temperatures in excess of 1200 �C.Archaeologists also report thin and horizontally exten-
sive lenses comprised of almost culturally sterile soils
from several ashmounds (Table 1). Another structural
feature reported from systematic ashmound excavations
and otherwise exposed sections is that of rammed earth
or clay platforms at the foundation of certain ashmounds
(Table 1). At sites with two or more excavated ashmo-
unds, the basal rammed earth platform is present only
in a single mound. This may indicate that the construc-
tion of these platforms was a temporally restricted prac-
tice that previous or subsequent mounds lacked.
From a consideration of these depositional strata it is
possible to discern several distinct formation patterns.
The first entails periods of small-scale dumping of dung
and dirt that were burned frequently at low temperatures.
Second, there are periods of larger scale accumulation
punctuated by less frequent high temperature burning
(vitrification). Third, there is evidence for periods marked
by the capping of ash layers with very thin culturally
sterile soil or clay in the upper mounded sections of
ashmounds. In addition, the depositional histories of
the lower sections of the mounds at excavated sites
appear to be the result of cattle pen construction and
maintenance (cf., Allchin, 1961, 1963; Paddayya, 1998).
Fig. 2. Profile illustrating ashmound stratigraphy from the site of Kupgal (photo courtesy of Carla Sinopoli).
P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330 313
These observations seem to indicate that ashmound
construction activities were carried out regularly and
repeatedly yet with differential building rhythm and
tempo throughout much of the South Indian Neolithic.
By tempo (Binford, 1982), I am referring largely to the
frequency of deposition whereas rhythm indicates attri-
butes of depositional activities such as their duration,
sequencing, and repetition. The C14 dates from exca-
vated ashmound contexts indicate that the activities in-
volved in their construction occurred throughout much
of the Neolithic period (see Fig. 3). Unfortunately, very
few C14 dates exist from excavated ashmound contexts
and only in a single case, Budihal-S, is there a strati-
graphic sequence for a single ashmound. If the anoma-
lously late date from layer 3 is discarded (as suggested
by Paddayya, 1999), then these data taken together
with the depositional nature of the strata (at Budihal-
S and all other profiled ashmounds) are strongly
suggestive of upper mound construction activities (i.e.,
accumulation and burning) that were temporally
slow, punctuated by other more dramatic and rapid
processes.
Table 1
Summary of frequently discussed ashmound sites and their characteristics (based on data from Allchin, 1963; Korisettar et al., 2001;
Krishna Sastry, 1979; Mujumdar and Rajaguru, 1966; Paddayya, 1991; Possehl, 1989; Rami Reddy, 1976; Shah, 1973)
Ashmound site Available C14
dates
(Calibrated)
Number of
ashmounds
at site
Basal
rammed-earth
feature present
Thin sterile
lenses present
Flora Fauna
Budihal-S 1400–2500 BC 3–4 Not reported Yes Horse gram,
hyacinth, barley,
jubejube, cherry,
emblic myrobalam
Antelope, black buck,
buffalo, cattle, fowl, nilgai,
sheep/goat, tortoise
Hulikallu Not reported 2 Not reported Not reported Not reported Cattle
Kakkera Not reported 2 Yes Not reported Not reported Not reported
Kodekal 2893 BC 1 Yes Yes Jubejube Buffalo, cattle, dog,
fowl, sheep/goat
Kudatini Not reported 1 Possible Yes Not reported Cattle, sheep/goat
Kupgal Not reported 3 Yes Yes Not reported Not reported
Mallur Not reported 2 Yes Not reported Not reported Cattle
Palavoy 1680–2278 BC 4 Yes Not reported Jubejube Cattle, deer, pig,
sheep/goat
Thanmandi Not reported 1 Yes Not reported Not reported Not reported
Thanda
Utnur 2333–2850 BC 1 Not reported Yes Not reported Cattle, deer, goat, tortoise
Wandalli Not reported 1 Not reported Not reported Not reported Not reported
Fig. 3. Radio-carbon dates from the South Indian Neolithic Period.
314 P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330
The production of a South Indian Neolithic landscape:
economy and human ecology
This paper examines ashmound features as impor-
tant monumental places, integral parts of a Neo-
lithic South Indian cultural landscape. The landscape
of the South Indian Neolithic was something both
inhabited and conceptualized by its prehistoric occu-
pants; a multitude of interconnected places in which
specific economic practices were conducted and social
and ideological relations mediated, maintained, modi-
fied, and reinvented. Landscape production involves
social and spatial practice, perception, and conception
as critical moments within historically and culturally
P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330 315
unique fields of social action (Harvey, 1989; Lefebvre,
1991; Soja, 1985). It does not simply entail the �con-struction� or �fabrication� of things in space but rather
the active configuration of social relations and forms
through dynamic and historically contingent processes.
These processes are both material and ideological and
articulate the natural environment with human knowl-
edge, technology, and labour. Networks of meaning,
involved in the social relations of spatial production,
inhere to multi-scalar spatial forms (e.g., landscapes,
buildings, fields, villages, and monuments) differen-
tially enabling and constraining activities and idea-
tional and ideological understandings of the cultural
landscape.
One means of beginning to examine the production
of a past cultural landscape is through an exploration
of the economic and ecological attributes involved with
land use inferred from an archaeological landscape. As
ashmound construction was inextricably bound to pas-
toralist elements of the Neolithic economy and subsis-
tence system, an examination of the tempo of land-use
(Binford, 1982; Wandsnider, 1992) and its connection
with the rhythm of activities involved in ashmound for-
mation is a useful point of embarkation. The tempo of
land or locale-use refers to the frequency with which a
place is utilized (Binford, 1982; Wandsnider, 1992) as
well as the nature of that usage, while rhythm refers to
the nature of a use activity�s temporality; its duration,
repetition, sequencing, and cycling. The following dis-
cussion examines archaeological and paleo-environmen-
tal data through the application and integration of the
spatial concepts of lifespace (Binford, 1983) and land-
scape element (Wandsnider, 1998).
Landscape element denotes an area of space ‘‘that is
homogenous and can be uniquely characterized’’
(Wandsnider, 1998, p. 22). These are spatial locations
with physical and conceptual attributes which may be
chosen for a variety of human uses or avoidance. A life-
space refers to the space within a landscape element that
is brought into use by its human occupants (Binford,
1983; Wandsnider, 1998, p. 22). The selection and loca-
tion of lifespaces are subject to the spatial and temporal
requirements associated with the human activities con-
ducted within them. Thus the nature, character, and
temporal duration of lifespaces are contingent upon a
variety of factors such as environmental, ecological, so-
cial or cosmological constraints and allowances that a
landscape element, or configuration thereof, hold for
the technological, economic, social, and even religious
needs of the groups of people occupying and embodying
a larger regional landscape over time. The human inte-
gration of landscape element configuration with life-
spaces result in structures of occupation and use
which, over time, leave patterned archaeological remains
such as artifact and feature distributions (Wandsnider,
1998, p. 23). This structure can be empirically observed
in the archaeological record of the South Indian Neo-
lithic at several scales of analysis such as the region, site,
feature, and assemblage.
Physical setting
Ashmound sites of the South Indian Neolithic are lo-
cated in north-eastern Karnataka state and western
Andhra Pradesh—the South Deccan/North Dharwar re-
gion (Fig. 1). This region is cross-cut by the upper
courses of the Bhima, Krishna, and Tungabhadra, three
major, shallow, wide, and slow moving rivers which flow
in a generally south-easterly direction towards the Bay
of Bengal. The physical landscape and geology of the re-
gion is characterized by a relatively flat to undulating
terrain that is regularly traversed by granite-gneiss hills
and hill chains (Paddayya, 1991, p. 573). Between the
basalt deposits in the dolerite dykes and the Deccan
Trap-topped ingersols, the quartz available in the Dhar-
war deposits and the chert, chalcedony, and quartzite
available in nodule form in the rivers, there was abun-
dant lithic raw material for the typical Neolithic ground
and pecked stone industries (Allchin, 1963; Paddayya,
1973). Within this geological region the two primary
types of rock formations—Dharwar schists and quartzes
and Archaean granites and gneiss—generally produce
two distinct types of soils as they erode. The Dharwar
produce arable �black cotton� soils and the Archaean
granites produce a red sandy to loamy soil (Allchin,
1963, p. 8). The latter predominates in the hilly tracts se-
lected by the Neolithic builders of ashmounds for site
location, while the former are found primarily in lower
lying areas, especially around the major rivers and gen-
erally away from most Neolithic sites. This focus on set-
tlements away from the region�s most arable land is
consistent with the emphasis on pastoralism and the
practice of low-risk, rain-fed agriculture present in the
Neolithic economy of the region.
The South Deccan/North Dharwar region is charac-
terized today by a semi-arid climate with an annual rain-
fall that generally does not exceed 50–60cm and falls
between June and August during the southwest mon-
soon (Paddayya, 1973, p. 4). The region�s semi-arid cli-
mate and seasonal rainfall patterns have created a
floral cover characterized by thorn and scrub bush for-
ests dominated by species such as Acacia, Zysiphus,
and Dalbergia, which are interspersed with large tracts
of savanna grasslands (Rami Reddy, 1976, p. 114).
Paleoenvironmental reconstructions of the South Indian
Neolithic based primarily on paleosol analysis at the
ashmound site of Kupgal by Mujumdar and Rajaguru
(1966), palynological analysis of a marine core (SK 27
B/8) extracted from the inner continental shelf in coastal
Karnataka (Caratini et al., 1991), and the recovery of
botanical material from the excavated sites in the region
(Mittre and Ravi, 1990), indicate that the environment
316 P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330
during the Neolithic was slightly wetter and more humid
than that of the region today.
Dimensions of agro-pastoral land-use and the South
Indian Neolithic landscape
Archaeological and environmental data demonstrate
that the inhabitants of Neolithic settlements in the South
Deccan/North Dharwar region were engaged in a mixed
subsistence economy comprised of livestock herding,
agriculture, and the exploitation of wild flora and fauna.
Bioarchaeological analyses of subsistence remains at a
wide variety of Neolithic sites have demonstrated the
presence of an assortment of domesticated and wild
plant and animal resources (Tables 2 and 3).4 At the site
of Budihal-S, the most extensively excavated Neolithic
site, two large ashmounds, a large animal butchering
surface (Paddayya, 1993a, p. 285) and at least one large
cattle pen associated with Ashmound I have been ex-
posed (Paddayya, 1998; Paddayya et al., 1995). These
features, together with the presence of large amounts
of Bos indicus bone and open-mouthed jar sherds of
red/grey coarse ware—interpreted by Allchin (1963) as
milking jars—attest to a production emphasis on pasto-
ral products such as milk and meat. Specimens of
domestic and wild plant species (see Tables 1 and 2) were
also recovered from excavations (Paddayya, 1993a, pp.
284-285, 2001, p. 213). The site is also peppered with
the remains of broken saddle querns and rubbing stones,
ethnographically, and archaeologically associated with
domestic grain processing. These data are suggestive of
a subsistence system with generalized production units.
The argument for a production emphasis on cattle
pastoralism for ashmound settlements is substantiated
by multiple lines of archaeological evidence. Docu-
mented faunal remains from excavated Neolithic sites
presented in Table 3 demonstrate that bones of the
domesticated species Bos indicus5 clearly dominate all
of the faunal assemblages of both ashmound and non-
ashmound sites. Faunal remains excavated from the
Budihal-S butchering surface were 95% Bos indicus while
the remaining 5% were comprised of sheep/goat and
wild animal species (Paddayya et al., 1995, p. 29). The
butchering function of this feature is indicated by the
taphonomy of the bones, the presence of large roasting
pits filled with charcoal, burned bone and ash, and
4 Based on their recent sampling of several South Indian
Neolithic sites Korisettar et al. (2002) argue that domesticated
plant species are largely absent from ashmound sites.5 Recent sampling of many Neolithic sites by Korisettar
et al. (2001) demonstrates difficulties in differentiating between
the bone remains of the species Bos indicus and domesticated
water buffalo, Bubalis bubalis. This may indicate that earlier
identifications of large domesticated ruminants are likewise
obscured such as those presented in Table 3.
numerous heavy chopping tools and large ‘‘knife-like’’
chert blades (Paddayya et al., 1995, p. 28). The bones
were deposited in clusters, possessed abundant cut
marks and were larger than those present in the sur-
rounding houses. Skulls and epiphyses of Bos indicus
long bones were frequent elements deposited in this fea-
ture, suggesting the preliminary butchering of whole car-
casses. If the presence of wide necked ceramic jars which
are ubiquitous at ashmound sites are considered func-
tionally related to milking activities then a productive
emphasis on the dung, meat, and milk products of cattle
can be inferred from the available archaeological
evidence.
Ashmounds themselves are strong indicators of a
production emphasis on pastoral production. If the mas-
sive volume of many of these features and their location
at more than 100 separate sites in the region are consid-
ered together with other serious structural investments
such as the cattle pens excavated at Budihal-S, Hulik-
allu, and Utnur and the large butchering floor at the for-
mer site an emphasis on pastoral production at many
Neolithic settlements appears empirically substantiated
(Table 4). The partially excavated (873m2) cattle pen
adjacent to Ashmound I at Budihal-S covers an area
of approximately 3000m2 (Paddayya, 2001) while the
total size of the butchering floor which consisted of a pre-
pared surface of ash, calcium carbonate, and gravel 2–
5cm thick, is estimated at 250m2 (129m2 of which has
been excavated). Given the spatial investment detailed
above for the three features associated with pastoral pro-
duction, the investment in houses of wattle-and-daub
(approximately 3–4m diameter) pale in comparison.
The high degree of investment in these features is further
suggestive of at least a semi-sedentary form of pastoral-
ism, one more consistent with an agro-pastoral lifeway.
The surveys of Allchin (1963), Paddayya (1973,
1991), and Rami Reddy (1976, 1990) display a number
of patterns in the selection of landscape elements for set-
tlement activity which may be used to suggest a larger
notion of lifespace beyond the traditional confines of
the �archaeological site,� unfolding the terrain surround-
ing a settlement into a regional-scale Neolithic land-
scape. While prehistoric grazing does not produce
archaeologically visible indicators beyond the features
that humans build to facilitate this activity, Paddayya
(1991) suggests that large, flat open areas surrounding
ashmound sites were regularly engaged by pastoralists
to graze their herds during the period of an adjacent
site�s occupation. Occupation of the region subsequent
to the Neolithic, and especially that associated with
the intensive development over the last 50 years has
likely erased most traces of ephemeral pastoral features
beyond ashmound sites (Paddayya, 1996). Another
interpretation is that the selection and occupation of
ashmound settlements (many of which were at least of
a seasonally sedentary nature) were made towards the
Table 2
Archaeobotanical remains of domesticated and wild plant species from South Deccan/North Dharwar Neolithic sites (based on data
from Devaraj et al., 1995; Fuller, 2003; Kajale, 1989; Korisettar et al., 2002; Murty, 1989; Paddayya, 2001; Venkatasubbaiah and
Kajale, 1991)
Common name Species Site
Domestic species
Millets Finger milleta Eleusine coracana Hallur, Paiyampalli, Watgal
Kodo millet Paspalum scrobiculatum Hallur
Foxtail millet Setaria verticillata Hallur, Hanumantaraopeta, Hattibelagallu,
Hiregudda, Kurugodu, Sanganakallu, Tekkalakota,
Velpumandugu
Browntop millet Brachiaria ramosa Hallur, Hanumantaraopeta, Hattibelagallu, Hiregudda,
Kurugodu, Sanganakallu, Tekkalakota, Velpumandugu
Pulses Horse gram Macrotyloma uniflorum Budihal-S, Hallur, Paiyampalli, Sangankallu,
Tekkalakota, Watgal
Green gram (mung) Vigna radiata Hallur, Hanumantaraopeta, Hattibelagallu,
Hiregudda, Sanganakallu, Tekkalakota, Paiyampalli
Black gram Vigna mungo Hallur, Hanumantaraopeta
Pigeon Pea Cajanus cajan Peddamudiyam, Sanganakallu
Hyacinth bean Lablab purpureus Budihal-S, Hallur, Sanganakallu
Large cereals Wheats Triticum sp. Hallur, Hanumantaraopeta, Hiregudda, Sanganakallu
Barley Horduem vulgare Budihal-S, Hanumantaraopeta, Hattibelagallu,
Hiregudda, Kurugodu, Sanganakallu, Tekkalakota
Wild species
Fruits Indian Jubejube Zizyphus jubea Budihal-S, Hallur, Kodekal, Palavoy, Sanganakallu,
Tekkalakota, Hiregudda
Indian Cherry/Sebestem Plum Cordia sp. Budihal-S
Emblic myrobalam Phyllanthus sp. Budihal-S
Betel Nut Areca catechu Watgal
a Fuller (1999, 2003; Korisettar et al., 2002) considers all identifications of finger millet, save a single specimen from Hallur, to be
misidentifications by previous research based on morphological attributes.
Table 3
Percentages of NISP results from South Deccan/North Dharwar Neolithic sites. (Based on data from Allchin, 1961; Monahan, In
press; Nagaraja Rao, 1971; Rami Reddy, 1976; Sastri et al., 1984; Shah, 1973)
Species % Hallur Kodekal Palavoy Piklihal Sangankallu Veerapuram VMS-110
Domestic
Cattle 94.0 59.9 95.8 74.6 94.5 70.29 51.06
Sheep/goat 3.2 6.5 2.7 18.3 1.0 6.27 34.04
Dog 1.6 2.6 0 0.6 0.5 1.6 0
Piga 0.3 0 0.3 0 0 0.8 2.13
Buffaloa 0 3.2 0 4.4 0 0 0
Wild
Antelope/deer 0.8 21.1 1.2 0 1.6 20.9 8.52
Tortoise 0 0 0 1.8 0 0 0
Rodent 0 2.6 0 0 2.1 0.8 0
Other 0 3.9 0 0 0 0 04.26
Total 100 100 100 100 100 100 100
a Korisettar et al., 2002, p. 190 point to the difficulty in determining wild from domestic specimens of buffalo (Bubalus bubalus) and
pig (Sus scrofa) found in Neolithic sites.
P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330 317
Table 4
Surface area and volume estimates from a selection of ashmounds and other Neolithic features (compiled from data in Allchin, 1963;
Paddayya, 1998, 2001)
Site Feature Surface area estimate (m2) Volume estimate (m3)
Budihal-S Neolithic circular house 12.5 22
Butchering floor 250 N/A
Ashmound I: upper mound 2000 4019
Ashmound I: stock enclosure 3000 N/A
Hulikallu Ashmound I 2449 9797
Kodekal Ashsmound 871 2323
Kudatini Ashmound 1295 8635
Wandalli Ashmound 2590 12,089
318 P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330
provisioning of pasturage adequate to the needs of com-
munity herds without entailing a mobility strategy that
would require long distance ephemeral facilities.
Agricultural practices are another aspect of the Neo-
lithic land-use system for which direct evidence exists in
the form of artifacts (querns, rubber stones) and more
limitedly from macrobotanical remains (Table 2). The
location of Neolithic settlements; ashmound and other,
are generally on or adjacent to large topographical fea-
tures. These micro-regional landscape elements provide
some of the lowest risk locations for rain-fed agricul-
tural practices. Many of the most frequently docu-
mented Neolithic domestic plant species6 were
drought-resistant crops that grow well in the red sandy
loam of the Archaean deposits (Fuller, 2003; Mittre and
Ravi, 1990, p. 102) and are well suited to the monsoon
drainage patterns of the outcrop topography. It is likely
that at least some of the area surrounding ashmound
sites was used for pulse and millet cultivation. As the
difference in physical characteristics between archaeo-
logically visible Neolithic settlements with and without
ashmounds differ only slightly (some non-ashmound
sites are located in better soil regimes), the selection of
landscape element for ashmound sites by Neolithic
agro-pastoralists appears to be based primarily on the
availability first, of abundant pasture (Paddayya,
1991) and second, on topographical features conducive
to rain-fed agricultural practices. The interpretation of
these communities as engaged in a mixed agro-pastoral
lifeway in which production units were jointly involved
in pastoral and agricultural activities suggests a lack of
conflict regarding land-use between these two subsis-
tence pursuits. This suggests a tempo of Neolithic site
and land-use in which the organization of settlement
and subsistence practices were based on a sedentary or
at least semi-sedentary pattern of site occupation.
6 Especially Fuller�s, 2003 �basic Neolithic package� of pulsesand millets—i.e., Brachia ramose, Setaria verticillata, Macroty-
loma uniflorum, and Vigna radiate.
A survey of the available archaeological data on the
landscape elements selected for ashmound sites (and
most non-ashmound settlements) demonstrates a num-
ber of interesting similarities regarding their location.
These include locations that are within 1–2km of sec-
ondary and tertiary tributaries of the region�s major
drainages, locations on or beneath large granitic out-
crops and, in many cases, near the 500m contour level
(Morrison, n.d.; Paddayya, 1973; Venkatasubbaiah
et al., 1992). Paddayya (1991, p. 586) reports three gen-
eral observations from his sample of eight ashmounds
in the Shorapur doab: (1) all are close to perennial
sources of water, i.e., springs or streams, (2) all have
substantial occupational debris surrounding them, and
(3) all are located proximal to large open spaces suit-
able for extensive cattle grazing activities. Paddayya
(1991) takes these observations as an indication that
ashmounds formed a central part of Neolithic habita-
tion sites and not simply sites in and of themselves.
This conclusion is generally corroborated (i.e., observa-
tions 1 and 2) by the findings of both Rami Reddy
(1976) and Allchin (1963) although in the case of the
latter researcher, ashmound sites are interpreted as rep-
resenting temporary camps either for the nocturnal
penning or domestication of cattle (Allchin and All-
chin, 1974).
Korisettar et al. (2002) dispute the year-round
occupation of ashmound sites based on their own
observations of low density distributions of occupa-
tional debris surrounding many of the sites they have
revisited. They argue that occupational debris sur-
rounding ashmounds vary in density and spread from
clearly evident (i.e., at Budihal-S, Kupgal, and Pala-
voy) to sparse (i.e., Kudatini and Utnur) reflecting
extended-stay and short term encampments (Koriset-
tar et al., 2002, pp. 212–213). Their observations
are compelling as is the argument for a stronger ana-
lytic focus on site formation processes, yet such a
conclusion in light of the findings from Budihal-S
would require further serious systematic research on
seasonality, (see Fuller et al., 2001) site structure
and punctuated site abandonment (Graham, 1993).
P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330 319
Further systematic surface collection and horizontal
excavations at a variety of ashmound sites such as
those conducted at Budihal-S would do much to re-
solve this issue.7
Ashmounds, monumentality, and ritual in Neolithic South
Indian agro-pastoral communities
Monumentality
Previous research on the �ashmound problem� haslargely produced functional explanations for the con-
struction of these features based on economic activities
(i.e., refuse dumps, stock enclosures, and smelting facil-
ities). In general, researchers agree that ashmounds
were formed as the result of pastoral activities. They
diverge on the issue of how, why, and when these fea-
tures were formed. While Paddayya (1991, 1998, 2001)
suggests a possible interpretation of ashmounds as
monuments and locations of ceremonial activity, and
Allchin (1963) considers some of the ritual implications
that are ethnographically associated with cattle dung
and fire in India, neither has approached these ques-
tions by exploring the socio-symbolic structure of built
forms. Many ashmound features were important, mon-
umental places within the cultural landscape of South
India�s Neolithic agro-pastoralist inhabitants. They
were built with the intent of expressing a specific range
of meaning and engendering specific sets of actions and
reactions.
Monuments are public structures designed and built,
in scale and detail, to be both non-prosaic and clearly
recognizable forms of the built environment (Moore,
1996, p. 92). Their character is at once ordered, commu-
nicative and symbolic, with powerful affectual qualities.
Monuments are saturated with a ‘‘horizon of meaning,’’
in which any one of several meanings may enable or con-
strain the thought and action of interacting subjects
based on a range of spatial, temporal, and social circum-
stances (Lefebvre, 1991, p. 222). The production of mon-
umental space is a transformative process in which
material, symbols and signs are exchanged, symbolically
grounding a given perceptual order (e.g., possible com-
binations of the cosmological, political, and social) to
a set of material practices within a conceptually
7 At present, Korisettar et al.�s (2002) argument is based
primarily on field reconnaissance and paleo-botanical sampling
at a small group of sites and the depth of cultural deposits at
ashmound and non-ashmound Neolithic sites. It should be
noted that little systematic work (i.e., surface collection and
documentation) on determining the expanse of many Neolithic
sites in the region has been undertaken (see Sinopoli and
Morrison, 1992 for a notable exception).
established social order (Lefebvre, 1991, pp. 216–217;
Moore, 1996, p. 97).
Monuments are symbolically charged communicative
media that condense complex and dynamic networks of
meaning critical to the mediation of social relations in
human communities (Lawrence and Low, 1990, p. 466;
Lefebvre, 1991, p. 227). If ashmounds are understood
as monumental forms of architecture it must be recog-
nized that much of the specificities of meaning involved
with their construction and use are well beyond the
reach of contemporary analysis. This analysis is
concerned with an understanding of monumentality cen-
tered on how meaning is conveyed through monumental
architecture rather than identifying the precise nature
and range of past meanings. In other words, it is the goal
of this analysis to identify ashmounds as symbolically
charged monuments, devices through which social rela-
tions were mediated, at least in part, within a ritual for-
um of community action. The focus is therefore on
cultural formation processes involved in the production
of ashmounds as features and places within a Neolithic
cultural landscape; especially their socio-economic and
socio-symbolic contexts.
Before exploring specific behavioral implications in-
volved with the construction of ashmounds, it is neces-
sary to demonstrate that ashmounds are in fact
examples of monumental architecture, capable of con-
veying a range of socio-symbolic meaning in a clear
and legible manner. This will be accomplished using four
visual dimensions of perception (taken from Higuchi,
1983, p. 183,8 and first applied to the study of monu-
mentality in archaeology by Moore, 1996): clarity of
form, contrast with background, prominence, and suffi-
ciency of mass to emphasize presence. Moore (1996, p.
97) explores these dimensions to qualitatively assess
how monumental architecture is used to communicate
legible meanings regarding social relationships ‘‘not only
because of their scale but because of their functional
unity and visual prominence.’’ It is crucial to note that
the analysis of these qualities serves only to identify
monumental architecture from the quotidian. The pro-
duction, mediation, and contestation of the range of
meanings a monumental space may embody is histori-
cally contingent upon unique circumstances through
which individual and group subjectivities are created
(Smith, 2003). Past meaning cannot be read or perceived
from a monument without access to the social context of
its production.
Ashmounds are clearly recognizable architectural
forms. Even today, more than 3000 years after their final
period of construction, ashmounds are recognized
(where site destruction has not erased or obscured their
form) as a class of cultural features by residents and
8 Originally designed by Lynch (1960).
9 Surface area estimates were calculated using the following
formula: pr1r2. For volume estimates the formula (4/3) pr1r2h/2was employed.
10 i.e., estimated to have a 3.5m diameter, 2m high walls,
and a 0.5m conical roof.
320 P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330
researchers of the region alike. Excavations at Budihal-S
(Paddayya, 1998), Utnur (Allchin, 1961), and Hulikallu
(Krishna Sastry, 1979) display features such as low
perimeter embankments of rubble, earth, and vitrified
dung, rows of post-holes, trenching, and prepared sur-
faces of earth and dung which suggest that the remains
of the lower elevation ashmound areas were once stock
enclosures. The higher elevation sections of ashmounds
are large mounds that can still be clearly observed today
without the aid of excavation or careful scrutiny (Figs.
4A and B). The clarity of these forms across the Neo-
lithic landscape of South India would have ensured the
legibility of the range of meaning these structures were
intended to convey. This does not imply that these
meanings were universally understood and accepted
and may well have been sites of contestation and
resistance.
Ashmound features also contrast with their natural
background. In many recorded instances ashmounds
and their surrounding settlements are situated in locally
prominent points on the landscape. These include on the
top of promontories or natural platforms, at the foot of
granite inselburgs or clusters of small rocky outcrops, or
in small valley passes between hills. The massive ash-
mound at Kudatini is one case of the latter, in which
the feature stands prominently in the middle of a small
pass between two hills (Fig. 5).
The prominence of a built form can also be exam-
ined by measuring the visual angle of incidence with
which an observer encounters a structure (Moore,
1996, p. 98). An angle of incidence is a measurement
of the slope between the top of a structure and the eyes
of a person standing at the closest point of viewing
(Moore, 1996, p. 105). Fig. 6 illustrates the calculated
angles of incidence of four of the ashmounds discussed
in this text for a human observer with a height of
approximately 1.75m. Moore (1996, p. 105) calculates
that the closest viewing point for disturbed Andean
mounds is half of the distance of the mound�s width.
The normal angle of incidence involved in walking on
flat to undulating terrain is between 10� and 15� belowthe horizon or 0� (Higuchi, 1983, p. 46) and a normal
line of sight is approximately �10� when standing still
(Moore, 1996, p. 98). The visually prominent nature
from a vertical perspective of the Kudatini, Wandalli,
Utnur, and Budihal-S ashmounds is detailed in Fig.
6. Given the normal line of site when standing still,
an individual would have to tilt his/her neck upwards
between approximately 14� and 27� to fully view these
ashmounds from viewpoints approximately half their
width away from the mound. Given their truly massive
dimensions in comparison with those of other Neolithic
structures (i.e., Neolithic houses) (Table 4) it is almost
certain that when approaching a settlement a prehis-
toric observer would have viewed the ashmound prior
to other cultural features. The monumentality of
ashmounds would have served to identify specific
places in the landscape conveying information regard-
ing the social identity of communities. Within a settle-
ment, ashmounds would have served as a constant
reminder of community social relations.
Ashmounds vary greatly in size, due in part to mil-
lennia of destruction from human activity such as ash
mining for building material and agricultural develop-
ment (Paddayya, 1996) (Fig. 7), but also from the
duration and intensity of activities involved in their
construction. Fig. 8 displays a sample of surface areas
from 22 sites for which adequate data are available.
Fig. 9 displays rough volumetric estimates for 20 of
the same ashmounds.9 Given that the construction
material for these mounds are at their source individual
patties of dung, even lower end volume estimates such
as the ashmound at Kodekal (2323m3) are strong indi-
cators of the substantial mass of these mounds. Given
that the (empty) volume of the next largest known Neo-
lithic structure in the region; a large Neolithic house10
is approximately 22m3, the mass of most ashmounds
were more than sufficient to emphasize their presence
in the settlements and landscape within which they
were situated (Table 4).
Previous researchers have not considered the possi-
bility that ashmounds were used for manuring activities
during the Neolithic and that the stronger productive
emphasis on pastoralism during these times simply cre-
ated an excess of dung. Perhaps due to its association
both as a by-product of cattle (the most significant focus
of the Neolithic economy) and the fertility of land, the
collection, piling, and burning of dung was transformed
from a prosaic maintenance activity into a cyclical cere-
monial practice based on the ritualized destruction of a
highly valued and sacrilized substance. While it is clear
that at many sites ashmound construction achieved
monumental dimensions, and likely that the rhythm of
repetitive ritual behavior led to these results, ashmound
construction almost certainly originated in quotidian
behavior associated with stock enclosure maintenance.
The early and in some cases mid use-lives of many of
these features would not necessarily have had permanent
monumental dimensions. In fact, ashmound dimensions
may have expanded and contracted on a regular basis as
dung deposits of variable sizes were differentially depos-
ited and later burned throughout the course of their con-
struction. It was through the differential rhythm and
tempo of dung collection, piling, small scale burning,
larger-scale, higher temperature vitrification, and
Fig. 4. (A) Kupgal Ashmound (photo courtesy of Carla Sinopoli). (B) Gadiganuru Ashmound (photo by author).
P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330 321
capping with sterile sediments that these monuments
were produced. By virtue of their visually conspicuous
and functionally integrated nature, these features
marked agro-pastoral settlements and locales in the
Neolithic landscape and served to commemorate and
memorialize communal ritual.
Fig. 5. Kudatini Ashmound (photo courtesy of Carla Sinopoli).
322 P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330
Ritual architecture
Given an understanding of the socio-symbolic, com-
municative structure of monumental architecture and its
potential to convey condensed and complex networks of
social and/or cosmological meaning (Moore, 1996, pp.
95–97), what sort of behavioral implications can be asso-
ciated with the construction of ashmounds by the Neo-
lithic agro-pastoralists of the South Deccan/North
Dharwar region of South India? I argue that ritual
behavior directly correlated with the importance of
cattle pastoralism in the economic lifeway of these com-
munities was responsible for the construction and main-
tenance of ashmound features.
Ritual is a processual and strategic mode of human
behavior. In a general sense ritual behavior is a means
of engagement with some form of authoritative order
or reality that is seen to both profoundly affect yet tran-
scend present circumstances (Bell, 1997, p. 169). This
engagement is accomplished through ‘‘deliberate and
meaningful’’ practices and activities sanctioned and nat-
uralized through a degree of social consensus granting
the ritual act ‘‘special privilege’’ from other more mun-
dane and prosaic activities (Bell, 1997, pp. 166, 167).
High degrees of formalism, performance, adherence to
tradition and rules, and socio-symbolic content are attri-
butes common to much ethnographically and histori-
cally observed ritual practice (Bell, 1997, pp. 93–169).
Ritual action is often a highly formalized mode of
communication through which social and cosmological
orders are conveyed and social relationships reproduced
and altered (Turner, 1967, p. 95). As a formalized and
redundant form of behavior, ritual action is often �objec-tified� in the form of material culture and constructed
space (Moore, 1996, pp. 136–139). The latter serves as
the loci for regular, repeated, and often ceremonial
expressions of socio-symbolically charged information
such as those associated with community integration.
These built forms (structures and spaces) are themselves
intentionally formalized to avoid ambiguities in the
meanings they convey and as such can often be discerned
from other (profane) spaces and structures (Moore,
1996, p. 137). However, as ritual behavior is temporally
discontinuous, ritual spaces, and structures are some-
times shared with those of more profane activities
(Moore, 1996, p. 13) and as such can be difficult to iden-
tify archaeologically. Yet specific kinds of ritual action,
i.e., that which is public, ceremonial, and repetitive—of-
ten leave structured archaeological remains that are un-
ique with respect to other spatial forms (Moore, 1996,
p. 139). These differences are often recognizable by phys-
ical qualities such as size, design, construction, and loca-
tion (Moore, 1996, p. 139), and are all measurable
attributes by which monumental architecture is distin-
guished from other structural forms and uses of space
within the structured remains of past cultural landscapes.
Fig. 6. Profiles with angles of incidence for four ashmounds.
P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330 323
Ashmounds were ritual lifespaces constructed by the
gradual and formalized performance of ritual activity.
Through the accumulation, burning, and periodic cap-
ping of dung with culturally sterile soils ashmounds ac-
quired monumental architectural form at many sites.
While the origin of these features was likely quotidian
activities (i.e., cattle penning and dung disposal/storage)
at some point these practices became secured to a sym-
bolic system embedded in a uniquely Neolithic, South
Indian conception of the world expressed as formalized
and repetitive communal ritual. In the following discus-
sion I employ five architectural variables (after Moore,
1996, pp. 139–167) to examine ritual behavior through
the available archaeological evidence of ashmound re-
mains, arguing that ashmounds were sites of public,
repetitive, and ceremonial expressions of ritual action.
These variables are: permanence, scale, centrality, ubiq-
uity, and visibility. Each are employed to examine a
range of social behavior in an effort to implicate ritual
practices involved with the construction and use of ash-
mounds and should not be taken to represent universal
criteria for designating a ritual type or form of architec-
ture. Ashmounds as structures are juxtaposed with other
built forms in this archaeological landscape and exam-
ined within the communities in which they were an inte-
gral part. This demonstrates that ashmounds were
monumental architectural forms designed to mediate so-
cial and perhaps cosmological meaning in a ritually
communicative space. As Budihal-S is the only ash-
mound site which has been subjected to multi-season,
large scale horizontal excavations, much of the focus
of the following discussion is on this site.
Permanence and scale
Moore (1996, p. 139) considers the variable of perma-
nence as an archaeological measure of the expected tem-
poral length of a ritual structure�s intended use-life by its
Fig. 7. Southern side of the Gadiganuru Ashmound illustrating destructive impact to the site from sediment mining (photo by author).
Fig. 8. Frequencies of surface area estimates for a sample of
ashmounds.
Fig. 9. Frequencies of volume estimates for a sample of
ashmounds.
324 P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330
architects. Attributes of importance for consideration
within this variable are (1) quality of building material,
(2) construction method, and (3) duration of use. Ash-
mounds were constructed primarily of cattle dung and
culturally modified soils. The lower sections of exca-
vated ashmounds consist of prepared surfaces with
enclosure walls constructed of dung, soil, wood, and un-
shaped sandstone blocks (Allchin, 1961; Paddayya,
1998). The upper portions of ash mounds, with vertically
monumental proportions, were constructed of piled and
burned dung. As such, construction of these mounds
would have consisted of at least two tempos.
The lower section of the mound would have been
constructed and maintained expediently, largely in keep-
ing with the needs of enclosing cattle and presumably
renovated as needed. This would have entailed the con-
struction of perimeter enclosures and embankments and
their periodic maintenance. Maintenance to the interior
surface would have required the regular removal of
P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330 325
excess dung and periodic resurfacing. Within the lower
section of Ashmound I at Budihal-S there is a circular
platform of sandstone blocks nine meters in diameter.
This platform is located in the center of the enclosure
amidst a small cluster of three child and a single cattle
burial as well as a concentration of beads, chert blades,
and knives and cattle and sheep/goat bone (Paddayya,
1998, p. 150). The presence of the platform, burials,
and artifacts, in addition to this area being the location
of intensive burning activity is strongly suggestive of
community ceremonial activity.
The upper sections of the mounds were constructed of
incrementally deposited loads of dungwhich were episod-
ically burned, gradually increasing the horizontal and
vertical dimensions of the mounds over an extended per-
iod of time. The use of a building material so closely asso-
ciated with a community�s lifeway and survival to
construct structures of such massive dimensions is also
suggestive of a high cultural value. As discussed above,
the profiles of all excavated ashmounds indicated that
there were at least two differential tempos to the burnings;
frequent low temperature burnings of thin lenses or layers
of dung and less frequent high temperature burnings of
thick layers resulting in large deposits (as much as 1m
thick) of vitrified strata (Fig. 2). These depositional activ-
ities were interspersed with episodes in which many of the
mounds were capped by culturally sterile soils (see Table
1). Both the high temperature burnings and the capping
episodes served to strengthen and preserve the structural
integrity of the mounds enhancing their permanence as
monumental places. What exactly the differences in
depositional tempo had to do with the rhythm of ritual
activities involved in the production of ashmounds is
uncertain. However, it does demonstrate that there were
a variety of activities involved in ashmound construction,
use and maintenance and that these activities were struc-
tured, repetitive, cyclical, and public.
Ashmounds are clearly in a size class of their own in
comparison to other structures in Neolithic settlements.
Compare the surface area of the largest house at Budi-
hal-S at 12.5m2 with that of Ashmound I (upper
mounded section) at 2000m2 and Ashmound II at
1256m2 (see Table 4). Given the size of many of these
features and their accretional construction it is clear that
the builders of these mounds intended them to be per-
manent structures. The C14 date sequence from Ash-
mound-I at Budihal-S suggests that its construction
use and maintenance continued for as much as 300–
400 years, however, such a conclusion is tentative. Fur-
ther comments on the temporal duration of construction
on a particular ashmound is difficult as so few C14 dates
are available (Fig. 3). The tempo of this construction
was clearly episodic taking place on intra-seasonal, sea-
sonal, generational or inter-generational scales. The
cyclical burning of the dung and the capping of layers
with soils extracted presumably from beyond the zone
of human habitation also suggests the systematic prac-
tice of ritual activity that was public, repetitive, and
ceremonial.
Centrality and ubiquity
Revisiting dozens of previously recorded sites, Pad-
dayya (1991) established that ashmounds are almost al-
ways at the center of intensive scatters of Neolithic and
in some cases post-Neolithic occupational debris (cf.,
Korisettar et al., 2002). His horizontal excavation at
Budihal-S documented the centrality of these features
in relation to the surrounding settlement area. Budi-
hal-S is a large Neolithic site consisting of four �locali-ties� of intensive occupational debris spread out over
an area approximately 12ha (Paddayya, 1998). At the
center of at least two of these localities are ashmounds
(a third locality has a large central deposit of ash that
has been largely destroyed, a fourth locality appears less
certain). Excavations in three of the localities exposed
stratified Neolithic occupational remains adjacent to
the ashmounds and ash deposits. At locality I, the re-
mains of 10 circular houses were excavated in the area
directly south of the ashmound and on either side of
the large animal butchering floor (Paddayya et al.,
1995, p. 25). Limited excavations at Hulikallu (Krishna
Sastry, 1979, p. 49) also exposed the remains of a sizable
habitation area proximal to the ashmounds including at
least one circular house floor. At many other ashmound
sites surface remains of occupational debris are scattered
over adjacent areas (e.g., Gadiganuru, Kurekuppa;
see Allchin, 1963 and Paddayya, 1991 for lists of sites).
The central position of ashmounds within many Neo-
lithic settlements is a further indication of their impor-
tance in the social regimes of these communities. There
are also cases where ashmounds or large deposits of ash-
mound materials are found on landforms adjacent to
settlement sites such as at Sanganakallu (Korisettar et
al., 2002) and VMS-110.
While many ashmound settlements contain only a
single mound, at several sites there are as many as four
(Table 1). Whether these mounds were constructed and
used simultaneously is uncertain, but if they were this
may indicate that the ritual activity associated with their
maintenance was oriented towards specific community
groups such kin-group affiliations; however, this remains
speculative. The absence of ashmounds at many
Neolithic sites—e.g., Hallur, Maski, Tekkalakota, Vee-
rapuram, Watgal—may indicate that ritual activity asso-
ciated with cattle production was restricted to particular
communities or that ashmounds at theses sites have sub-
sequently been destroyed. It should be noted that during
the excavation of Neolithic Watgal a rammed earth fea-
ture surrounded (but not mounded) by a large and dense
concentration ash lensing was exposed (Devaraj et al.,
1995).
326 P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330
Visibility
The consideration of the visual accessibility of ritual
performance at ashmound features contributes to a bet-
ter understanding of the public nature of ceremonial life
at these places (see Fogelin, 200311). The scale of ashmo-
unds and their central location proximal to otherwise
vertically undifferentiated settlements suggests an equal-
ity of visual access to ceremonial activity within the ash-
mound precinct. At least one activity, the burning of the
mound, would have been visible to all in the community.
However, there is the possibility that ceremonial activity
within the enclosed lower mound sections, such as pos-
tulated earlier for the circular sandstone platform at
Budihal-S, may have been visually obscured to those
on the exterior of the enclosure by perishable materials
used in wall construction. Lines of post-holes around
the perimeter of the lower mound at Utnur suggest
the presence of such a vision restricting wall (Allchin,
1961, pp. 66–68)). Large patches of burned surfaces at
Budihal-S in isolated areas (such as that surrounding
the platform) indicate that pyrotechnic activity was
not restricted to the upper ashmound. However, the
large surface area within the enclosure could have held
hundreds of people at a time. And while it may be con-
cluded that the enclosure walls functioned to keep cattle
in, it cannot be determined that they also functioned to
restrict access to ceremonial activities within its confines.
Ritual forms of communication that are public, for-
malized, and repetitive require the structured organiza-
tion of space. The structured remains of ashmounds,
which are monumental, permanent, highly visible, and
central to areas populated during the South Indian Neo-
lithic suggest that these structural spaces were the loca-
tion of public, formalized, and repetitive communal
ritual. Ashmounds were monumental forms of architec-
ture built in part through ritual processes intended to
transmit socio-symbolically charged information likely
concerning group integration and social reproduction.
The ritual and monumental afterlives of ashmounds in
post-Neolithic South India
The activities involved with ashmound construction
appear to have begun at some point during the mid cen-
turies of the third millennium BC and endured for sev-
eral hundred years at many sites across South India
(Fig. 1). Following a shift in economic emphasis from
11 Fogelin (2003) employs horizontal sight line angles to
examine visibility between circumambulatory and assembly
areas in Early Historic period Indian rock-cut chaitya and open
air stupa complexes. This analysis has effectively demonstrated
visual barriers between two spatially segregated areas of ritual
practice in Early Historic Buddhist architecture.
a strong productive concentration on pastoralism in
Neolithic times to either a more balanced mix of agricul-
ture and animal husbandry or a stronger reliance on the
former during the subsequent Iron-Age (1200–400
BC), the practicality of creating monuments employing
a material with such valuable economic utility as cow
dung gradually became obsolete. Monuments continued
to be erected on the regional landscape, however, con-
structed instead primarily of stone and earth (i.e.,
megaliths).
Despite significant changes in social organization,
economy, and landscape production with the transition
from the Neolithic period to the Iron-Age (see Brubaker,
2001; Moorti, 1994), many ashmounds continued to be
important monumental places involved in ritual activi-
ties central to Iron-Age landscape production. Yet with
this transition, the social lives of ashmounds were signif-
icantly transformed, often in form and almost certainly
in terms of meaning. Despite this transformation, there
are clear indications that the ritual and monumental nat-
ure of ashmounds as significant places in the cultural
landscape of the Neolithic were embedded in the social
memory of Iron-Age societies. Memory is a spatially
contextualized mode of retaining and reproducing sen-
sory and mental impressions (Alcock, 2001; Bachelard,
1964). Social memory involves the transmitted memories
of groups of people; memory which serves in part to
construct group and individual identities and subjectivi-
ties tying the present to the past (Alcock, 1993, 2001;
Bradley, 1987; Connerton, 1989). The performance of
socio-symbolic action through formalized repetitive rit-
ual practice is an important mode of transmission for
group traditions (Bell, 1997, pp. 167–169).
During the Iron-Age there are a number of different
archaeologically visible ritual practices that involve ash-
mounds in the production of monumental places within
the cultural landscape. These include (1) the continua-
tion of ashmound formation in a manner consistent with
the Neolithic period, (2) the occupation, reoccupation or
re-use of locales with ashmounds, including, (a) occupa-
tions not including the construction of megalithic mon-
uments, (b) the construction of megalithic monuments
adjacent to existent ashmounds, and (c) the incorpora-
tion of ashmounds into expansive megalithic complexes,
and (3) the recycling of ashmound material in megalithic
memorials.
Abundant deposits of Iron-Age pottery from early in
the stratigraphic sequence at the site of Palavoy, as well
as an especially late C14 date from layer 7 in Ashmound
I, indicate that the practice of building ashmounds con-
tinued into the Iron-Age (Rami Reddy, 1976). The
abundant presence of Iron-Age pottery in the surface
scatters of occupational debris surrounding many ash-
mounds, as well as the occurrence of megaliths (e.g., dis-
tributions of stone circles, dolmens, and menhirs of
variable size and extent with or without interred human
P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330 327
remains) at or around many of these sites suggest a con-
tinuity of occupation or at least re-occupation and reuse
of these places for monumental building activities (All-
chin, 1963; Rami Reddy, 1976; Sundara, 1971). During
the course of the Vijayanagara Metropolitan Survey
every ashmound or ashmound-like deposit observed
was directly proximal to either surface scatters of Iron-
Age and Early Historic cultural material or standing
megaliths (Morrison, n.d.). Surface collections from
one such site, VMS-634 has yielded a ceramic assem-
blage dominated by Iron Age/Early Historic types
(Johansen, 2003).
Perhaps the most interesting continuity of ritual
practice is the erection of a massive megalithic monu-
ment on top of and around an ashmound in the Shora-
pur doab just north of the town of Shahpur (Fig. 1).
Meadows Taylor (1853, pp. 393-396; 1862) reported that
the 20m diameter mound at this site was encircled by
eight perimeters of large standing stones (some as tall
as 3m) and that the mound itself was faced with flat
stones and capped with a layer of soil with a circle of
standing stones on its summit (Fig. 10). Ashmound
deposits (up to 3m thick) were discovered when he exca-
vated the mound looking for tomb cysts. Meadows Tay-
lor (1853, 1862, p. 396) believed, according to one of the
Fig. 10. Plan and Section of the Shapur Ashmou
competing theories of his day, that the powdered and
vitrified ash in these deposits were the result of large-
scale human cremation. In the same general area where
Meadows Taylor made his observations, Paddayya
(1973) reports the remains of a large stone circle mega-
lith on top of the Shakapur ashmound. Allchin (1963,
p. 68) also reports a large stone circle of basalt and
gneiss boulders on the top of a large ashmound on the
Hanamsagar–Kodekap road.
To these cases must also be added the Iron-Age �ashcircle graves�; a megalithic construction found at a few
sites close to ashmound locations in which circular sur-
face deposits of dung ash enclose both stone circle and
dolmen megaliths. This unique category of megalith
has been observed in subtly different forms at the sites
of Rajankolur, Dimanhal (Paddayya, 1973), Chikka
Benekal, Piklihal, and Lingsugar (Allchin, 1960, 1963)
and Billamrayan Gudda (Munn, 1935) (see Fig. 1). At
each of these locations it appears that deposits of pow-
dered and vitrified dung from nearby ashmounds were
incorporated into the construction of the later Iron-
Age monuments, although this dung ash may have been
processed during the Iron-Age. Finally, the recent exca-
vation of an Iron-Age four-legged terracotta sarcopha-
gus burial from occupational deposits north-east of the
nd-Megalith (after Meadows Taylor, 1862).
328 P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330
massive ashmound at Kudatini (Boivin et al., 2002) is
another dramatic example of ritual and memorial conti-
nuity of place bridging the Neolithic and Iron-Age
periods.
Clearly there exists in this regional transition of mon-
umental practice and communal ritual a transmutation
of social memory from Neolithic to Iron Age cultural
landscape production. Ashmounds have formed an inte-
gral part of the experience and perception of those
inhabiting the cultural landscapes of the South Dec-
can/North Dharwar region from the Neolithic and
Iron-Age through to the present day. The archaeological
evidence for Iron-Age incorporation of the space and
material of Neolithic ashmound monuments into similar
and very different forms of landscape production dem-
onstrates a spatial and temporal continuity of social
importance associated with very special places in very
differently constituted social orders. The continuity of
ritual and monumental emphasis on special points on
the cultural landscape demonstrates the fluid nature of
cultural change in this dynamic regional landscape.
Conclusion
Neolithic ashmounds were embedded in an agro-pas-
toral landscape in which small village communities
emphasized the production of pastoral products. Sites
and settlements marked by ashmounds were located in
similar landscape elements ecologically favorable to pas-
toralism and small-scale agriculture. Ashmounds were
constructed incrementally, synchronized with the social
and ritual rhythm of cattle keeping. Within years or gen-
erations many of these mounds had acquired the dimen-
sional attributes of monumental form.
The recognition of ashmounds as monumental archi-
tecture entails an understanding of the socio-symbolic
potential of built form. Based upon a number of visual
dimensions of perception, this examination of ashmo-
unds demonstrates their monumentality and the com-
municative structure of their form as a unique class of
features. The proportions of many ashmounds ensured
that these were the most prominent structures on the
Neolithic landscape. Within settlements or more ephem-
eral encampments, ashmounds were visually unavoid-
able and served to constantly reinforce complex
networks of socio-symbolic meaning. While no attempt
to understand the specificities of this possible range of
meanings is made, certain behavioral implications in-
volved in ashmound construction inferred from the
archaeological record suggest an origin in ritual action.
A close examination of ashmound deposits illustrates
the cyclical and repetitive rhythm of activities involved
in their construction. In the upper sections of excavated
ashmounds this included the collection of cow dung, its
deposition in a central location within agro-pastoral
settlements and locales, its subsequent burning and the
capping of some of these episodes with culturally sterile
soils. The regular and formalized nature of these deposi-
tional episodes suggests an interpretation that is consis-
tent with the objectification of ritual action in the
production of these places. Examining ashmounds using
a set of dimensions designed to infer specific kinds of rit-
ual activity in built form, demonstrates the possibility
that these features were the location of regular public
ceremonial activity associated with cycles of pastoral
production. The accretional tempo of ashmound con-
struction and use and their monumental form likely
served to continually reinforce socio-symbolically
charged information conveyed during regular episodes
of ritual practice.
A closer analysis of the activities involved in
ashmound construction, use and maintenance poses a
variety of new questions about Neolithic ecology, envi-
ronment, society, economy, and ritual. An attempt has
been made to explain their uniqueness and ubiquity in
the prehistoric South Indian landscape as intersections
for a complex of dynamic cultural interactions rather
than single sphere use facilities like cattle pens or refuse
dumps. Crucially, this re-visitation of the �ashmound
problem� has demonstrated the certainty that much
more work needs to be undertaken in the directions of
data collection, analysis, and theory building before
the full explanatory potential of the ashmounds is even
close to being broached.
Acknowledgments
This work was originally a Master�s thesis completed
by the author in the spring of 2000 for the University of
Chicago�s Department of Anthropology. The argument
presented here is built on archaeological data carefully
collected by many researchers but especially by F.R. All-
chin and K. Paddayya. A great debt is owed to them for
their exhaustive and challenging research. Discussions
with Professor K. Paddayya while he was a visiting Ful-
bright scholar at the University of Michigan in 1999
were also an invaluable resource in formulating and
researching this paper. Comments from Andrew Bauer,
Radhika Bauer, Kathleen Morrison (thesis supervisor),
Sandra Morrison, Carla Sinopoli, and Adam T. Smith
were very helpful and greatly appreciated. I thank John
O�Shea and an anonymous reviewer for their very valu-
able comments during the review process. All responsi-
bility for errors and opinions are my own.
References
Alcock, S., 2001. The reconfiguration of memory in the Roman
Empire. In: Alcock, S., D�Altroy, T., Morrison, K., Carla
P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330 329
Sinopoli (Eds.), Imperial Designs: Comparative Dynamics
of Early Empires. University of Cambridge Press,
Cambridge.
Alcock, S., 1993. Graecia Capta: The landscapes of Roman
Greece. Cambridge University Press, Cambridge.
Allchin, F.R., 1963. Neolithic Cattle-Keepers of South India.
Cambridge University Press, Cambridge.
Allchin, F.R., 1961. Utnur Excavations. Andhra Pradesh
Government Archaeological Series No. 5. Government of
Andhra Pradesh, Hyderabad.
Allchin, F.R., 1960. Piklihal Excavations. Government of
Andhra Pradesh, Hyderabad.
Allchin, F.R., Allchin, B., 1974. Some new thoughts on Indian
cattle. In: van Lohuizen-de Leeuw, J.E., Ubaghs, J.M.
(Eds.), South Asian Archaeology 1973. E.J. Brill, Lieden,
pp. 71–77.
Bachelard, G., 1964. The Poetics of Space. Beacon Press,
Boston.
Bell, C., 1997. Ritual: Perspectives and Dimensions. Oxford
University Press, Oxford, New York.
Binford, L., 1982. The archaeology of place. Journal of
Anthropological Archaeology 1, 5–31.
Binford, L., 1983. In Pursuit of the Past. Thames and Hudson,
New York.
Boivin, N., Korisettar, R., Venkatasubbaiah, P.C., Lewis, H.,
Havanur, D., Malagyannavar, K., Chincholi, S., 2002.
Exploring Neolithic and Megalithic south India: the Bellary
District archaeological project. Antiquity 76 (4), 937–938.
Bradley, R., 1987. Time regained: the creation of continuity.
Journal of the British Archaeological Association 140, 1–17.
Brubaker, R., 2001. Aspects of Mortuary Variability in the
South Indian Iron Age. Bulletin of the Deccan College Post-
Graduate and Research Institute 60–61, 253–302.
Bruce Foote, R., 1979[1916]. Prehistoric and Protohistoric
Antiquities of India. Leeladevi, Delhi.
Caratini, C., Fontugne, M., Pascal, J.P., Tissot, C., Bentaleb, I.,
1991. A major change at ca. 3500 years BP in the vegetation
of the Western Ghats in North Kanara, Karnataka. Current
Science 61 (9 and 10), 669–672.
Connerton, P., 1989. How Societies Remember. Cambridge
University Press, Cambridge.
Devaraj, D.V., Shaffer, J.C.S., Patil, Balasubramanya, 1995.
The Watgal Excavations: an Interim Report. Man and
Environment 20 (2), 57–74.
Fogelin, L., 2003. Ritual and presentation in early buddhist
religious architecture. Asian Perspectives 42 (1), 129–154.
Fuller, D.Q., Korisettar, R., Venkatasubbaiah, P.C., 2001.
Southern neolithic cultivation systems: a reconstruction
based on Archaeo-botanical evidence. South Asian Studies
17, 171–187.
Fuller, D.Q., 2003. Indus and non-indus agricultural traditions:
local developments and crop adoptions on the Indian
Peninsula. In: Weber, S., Belcher, W. (Eds.), Indus Ethno-
biology: New Perspectives from the Field. Lexington Books,
Maryland (Chapter 10).
Graham, M., 1993. Settlement organization and residential
variability among the Raramuri. In: Cameron, C.M.,
Tomka, S.A. (Eds.), Abandonment of Settlements and
Regions: Ethnoarchaeological and Archaeological
Approaches. Cambridge University Press, Cambridge, pp.
138–153.
Harvey, D., 1989. The Condition of Postmodernity. Ch.12–14.
Basil Blackwell, Oxford.
Higuchi, T., 1983. The visual and spatial structure of land-
scapes. MIT Press, Cambridge, MA.
Johansen, P., 2003. Variability in surface distributions of
ceramics from three iron age -early historic sites in the
Tungabhadra Corridor, Northern Karnataka. Paper pre-
sented at the 32nd Annual Meeting on South Asia, October
18th, 2004, Madison, Wisconsin.
Kajale, M.D., 1989. Ancient Grains from Excavations at Hallur
(1976) Season, District Dharwar, Karnataka. Bulletin of the
Deccan College Research Institute 47–48, 123–128.
Korisettar, R., Venkatasubbaiah, P.C., Fuller, D.Q., 2002.
Brahmagiri and Beyond: the Archaeology of the Southern
Neolithic. In: Settar, S., Korisettar, R. (Eds.), In Indian
Archaeology in Retrospect, Volume I. Prehistory, Publica-
tions of the Indian Council for Historical Research,
Manohar, New Dehli, pp. 151–238.
Korisettar, R., Joglekar, P.P., Fuller, D.Q., Venkatasubbaiah,
P.C., 2001. Archaeological Re-investigation and Archae-
ozoology of Seven Southern Neolithic Sites in Karnataka
and Andhra Pradesh. Man and Environment 26 (2), 47–66.
Krishna Sastry, V.V., 1979. The ashmound excavations at
Hulikallu. The Andhra Pradesh Journal of Archaeology 1
(2), 47–55.
Lynch, K., 1960. The Image of the City. MIT Press, Cam-
bridge, MA.
Lawrence, D., Low, S., 1990. The built environment and spatial
form. Annual Reviews in Anthropology 19, 453–505.
Lefebvre, H., 1991. The Production of Space. Basil Blackwell,
Oxford.
Lycett, M., 2001. Transformation of place: occupational history
and differential persistence in seventeenth century New
Mexico. In: Preucell, R.W. (Ed.), Archaeologies of the
Pueblo Revolt: Identity, Meaning, and Renewal in the
Pueblo World. University of New Mexico Press, Albuquer-
que, pp. 61–74.
Meadows Taylor, Captain 1862. Description of Cairns, Crom-
lechs, Kristavaens and Other Celto-Druidical or Scythian
Monuments in the Dekhan. Transactions of the Royal Irish
Academy 24(3), 329–362.
Meadows Taylor, Captain 1853. Notices of Cromlechs, Cairns,
and other Ancient Scytho-Druidical Remains in the Prin-
cipality of Shorapur. Journal of the Bombay Branch of the
Royal Asiatic Society vol. XVII (IV), 380–429.
Mittre, V., Ravi, S., 1990. The environment and economy of the
neolithic cultures of South India: paleobotanical evidence.
In: Sundara, A., Bhatsoori, K.G. (Eds.), Archaeology in
Karnataka. Directorate of Archaeology and Museums,
Mysore, pp. 100–112.
Monahan, B., (in press). South Indian Animal Use: Subsistence
Patterns and Exchange Practices. To appear in Recent
Research on the Archaeology and Paleoanthropology of
South India, edited by V.V. Rami Reddy and Kathleen
Morrison. Munshiram Manoharlal, New Delhi.
Moore, J., 1996. Architecture and Power in the Ancient Andes:
The Archaeology of Public Buildings. Cambridge Univer-
sity Press, Cambridge.
Moorti, U., 1994. Megalithic Culture of South India: Socio-
evolutionary Perspectives. Ganga Kaveri Publishing House,
Varanasi.
330 P.G. Johansen / Journal of Anthropological Archaeology 23 (2004) 309–330
Morrison, K. n.d. Oceans of Dharma: The Political Ecology of
Place. Manuscript in preparation.
Mujumdar, G.G., Rajaguru, S.N., 1966. Ashmound Excava-
tions at Kupgal. Deccan College Post-Graduate and
Research Institute, Poona.
Munn, L., 1921. Ancient mines and megaliths in Hyderabad.
Memoirs of the Manchester Literary and Philosophical
Society 69 (5), 1–11.
Munn, L., 1935. Prehistoric and protohistoric finds in the
Raichur district. Man in India 15, 225–250.
Murty, M.L.K., 1989. Pre-iron age agricultural settlements in
South India: an ecological perspective. Man and Environ-
ment XIV (1), 65–81.
Nagaraja Rao, M.S., 1971. Protohistoric Cultures of the
Tungabhadra Valley. Nagaraja Rao, Dharwar.
Newbold, T.J., 1843. On some Ancient Mounds of Scorious
Ashes in Southern India. The Journal of the Royal Asiatic
Society of Great Britain and Ireland 7, 129–136.
Paddayya, K., 2001. The Problem of Ashmounds of Southern
Deccan in the Light of Budihal Excavations, Karnataka.
Bulletin of the Deccan College Post-graduate Institute
(Diamond Jubilee Volume), Volume 60–61, 189–225.
Paddayya, K., 1999. Excavations at The Neolithic Ashmound
Site Of Budihal, Gulbarga District, Karnataka. Article on
the web @ www.digindia.com.
Paddayya, K., 1998. Evidence of Neolithic Cattle-Penning at
Budihal, Gulbarga District, Karnataka. South Asian Stud-
ies 14, 141–153.
Paddayya, K., 1996. Modern impacts on archaeological sites in
India: a case study from the Shorapur Doab, Karnataka.
Man and Environment XXI (2), 75–88.
Paddayya, K., 1993a. Ashmound Investigations at Budihal,
Gulbarga District, Karnataka. Man and Environment 18
(1), 57–88.
Paddayya, K., 1993b. Further Field Investigations at Budihal,
Gulbarga District, Karnataka. In Bulletin of the Deccan
College Post-Graduate Research Institute 53, 277–322.
Paddayya, K., 1991. The Ashmounds of South India: Fresh
Evidence and Possible Explanation. In Bulletin of the
Deccan College Post-Graduate Research Institute 51-52,
573–626.
Paddayya, K., 1973. Investigations into the neolithic culture of
the shorapur doab. E.J. Brill, Leiden.
Paddayya, K., Thomas, P.K., Joglekar, P.P., 1995. A Neolithic
animal butchering floor from Budihal, Gulbarga District,
Karnataka. Man and Environment 20 (2), 23–31.
Possehl, G., 1989. Radiocarbon Dates from South Asian
Archaeology. The University of Pennsylvania Museum
Occasional Publication of the South Asian Section,
Philadelphia.
Rami Reddy, V.R., 1990. Ashmounds in South India: A
Review. In: Sundara, A., Bhatsoori, K.G. (Eds.), Archae-
ology in Karnataka. Directorate of Archaeology and
Museums, Mysore, pp. 85–99.
Rami Reddy, V.R., 1976. The Prehistoric and Protohistoric
Cultures of Palavoy, South India: With Special Reference to
the Ashmound Problem. Government of Andhra Pradesh,
Hyderabad.
Sastri, T.V.G., Kasturi Bai, M., Vara Prasada Rao, J., 1984.
Veerapuram: A type site for cultural study in the Krishna
Valley. Archaeological and Cultural Research Institute,
Hyderabad.
Sewell, R., 1899. The Cinder-Mounds of Bellary. The Journal
of the Royal Asiatic Society of Great Britain and Ireland 51,
1–16.
Sinopoli, C., Morrison, K., 1992. Archaeological Survey at
Vijayanagara. Research and Exploration 8 (2), 237–239.
Shah, D.R., 1973. Animal Remains from Kodekal Excavation.
In: Paddayya, K. (Ed.), Investigations Into The Neolithic
Culture Of The Shorapur Doab. E.J. Brill, Leiden, pp. 92–
95.
Smith, A.T., 2003. The Political Landscape: Constellations of
Authority in Early Complex Societies. University of Cali-
fornia Press.
Soja, E.W., 1985. The Spatiality of Social Life: Towards a
Transformative Retheorization. In: Gregory, D., Urry, J.
(Eds.), Social Relations and Spatial Structures. St. Martin�sPress, New York, pp. 90–127.
Sundara, A., 1971. New discoveries of ashmounds in North
Karnataka and their implications. In: Ganesan, S., Rajan,
S., Ramaswami, N.S., Sampath, D.S. (Eds.), Professor K.A.
Nilakanta Shastri Felicitation Volume. K.A. Nilakanta
Shastri Felicitation Committee, Madras, pp. 308–314.
Turner, V., 1967. The Forest of Symbols: Aspects of Ndembu
Ritual. Cornell University Press, London and Ithica.
Venkatasubbaiah, P.C., Kajale, M., 1991. Biological remains
from neolithic and early historic sites in Cuddapah district,
Andhra Pradesh. Man and Environment 15, 85–97.
Venkatasubbaiah, P.C., Pavankar, S.J., Joglekar, P.P., 1992.
Neolithic faunal remains from the central Pennar Basin,
Cuddapah District, Andhra Pradesh. Man and Environ-
ment XII (1), 55–59.
Wandsnider, L., 1998. Landscape element configuration, life-
space, and occupation history. In: Sullivan, A.P. (Ed.), Time
and Archaeological Landscapes. University of New Mexico
Press, Albuquerque, pp. 21–42.
Wandsnider, L., 1992. The spatial dimension of time. In:
Rossignol, J., Wandsnider, L. (Eds.), Space, Time and
Archaeological Landscapes. Plenum Press, New York and
London, pp. 257–284.
Woolley, L., 1940. Some aspects of problems in Indian
archaeology. Journal of the Royal Society of the Arts, 87.
Yazdani, G.L., 1936. Notes on a survey of Neolithic sites.
Annual Report of the Hyderabad Archaeological Depart-
ment, 1935–1936.
Zeuner, F.E., 1960. On the origin of the cinder mounds of the
Bellary district, India. Bulletin of the London Institute of
Archaeology 2, 37–44.