Chitradurga: Spatial Patterns of a Nayaka Period

Successor State in South India

BARRY LEWIS AND C. S. PATIL

The Nayaka period in South Indian history spans the centuries between
the collapse of the Vijayanagara empire in the late sixteenth century and the
consolidation of British power in the south at the beginning of the nineteenth
century. It was an era of dynamic social change, during which petty rulers and
subsidiary chiefs, known variously in the literature as little kings, nayakas,1 and
poligars (hereinafter referred to collectively as poligars for the sake of conve-
nience), acquired and expanded territories, claimed increased status and legiti-
macy, and often asserted their independence of higher ruling authority (Chitnis
1974, 2000; Dirks 1994; Dua 1996; Rajayyan 1974; Stein 1985). A few of these
kingdoms, notably Mysore, Keladi, Madurai, Tanjore, and Gingee, were emi-
nently successful and their histories dominate our understanding of the period
(e.g., Aiyar 1991; Brittlebank 1997; Chitnis 1974; Deloche 2000; Dirks 1982,
1994; Michell 1995 : 13; Narayana Rao et al. 1992; Swaminathan 1957; Vrid-
dhagirisan 1995). Signicantly less is known about the scores of smaller kingdoms
and chiefdoms that populated the Nayaka period landscape ( Ramachandra Rao
1943).
Our objective is to describe an archaeological and ethnohistorical case study of
Chitradurga, which was one of the important Nayaka polities of central Karna-
taka (Fig. 1). Between a.d. 1500 and 1800, Chitradurga was at various times a
Vijayanagara province, an independent kingdom, a tributary of the Marathas, a
tributary of the Mughals, a tributary of Mysore, and, nally, a Mysore province. It
also was often at war with most of its neighbors. In spite of its tumultuous recent
past, Nayaka period Chitradurga is so little known that it warrants only a modest
footnote in Karnataka archaeology.
The specic goal of this case study is to identify and interpret major spatial
patterns of the Chitradurga cultural landscape as they existed in the late eigh-
teenth and early nineteenth centuries. The data are primarily drawn from our
recent archaeological surveys in the Chitradurga region (Lewis 1997; Patil 1999)
and from the highly detailed manuscript maps of Mysore made by Colin Mack-

Barry Lewis is a professor in the Department of Anthropology, University of Illinois, Urbana. The
late C. S. Patil was the deputy director of Karnataka Directorate of Archaeology and Museums, The
Palace Complex, Mysore, India, and research associate in the Department of Anthropology, Uni-
versity of Illinois, Urbana.
Asian Perspectives, Vol. 42, No. 2 ( 2003 by the University of Hawaii Press.
268 asian perspectives . 42(2) . fall 2003

Fig. 1. Chitradurga province, showing parganas and the major towns.

enzie and his surveyors in the rst decade of the 1800s (Edney 1997 : 175179;
Phillimore 19451958 : 2 : 91121; Robb 1998). The results provide a distinctively
archaeological perspective of this province, emphasize the important role that
such polities played in the historical developments of the Nayaka period, and
 lewis and patil . spatial patterns of a nayaka successor state 269

illustrate the signicant interpretative value of early East India Company (EIC)
manuscript maps for archaeological site survey and spatial analyses of early mod-
ern India. These results complement the perspective that can be gained from
analyses of contemporary historical and literary texts and contribute signicantly
to a richer understanding of such polities than would be possible otherwise.
We begin with an overview of the recent history of Chitradurga and the data
used in this study. Next, the major cultural features of the region are described as
they were at the beginning of the nineteenth century. The nal section identies
important spatial patterns of these features and interprets their signicance in the
cultural context of that time.

chitradurga in historical perspective

Chitradurga was one of several central Karnataka regions governed by local
chieftains well before the rise of Vijayanagara, which inherited control of the
region from the Hoysalas in the early fourteenth century. By the early sixteenth
century, it was controlled largely by a Bedar or Boya caste family who traced their
origins to southern Andhra Pradesh from which they had emigrated with their
herds (Hayavadana Rao 1930 : 5 : 13801387; Rice 1897 : 1 : 500505). Like most
of the central Karnataka families who rose to power in the sixteenth century (e.g.,
Chitnis 1974 : 914; Ota 1999), the Chitradurga poligars based the legitimacy of
their status as an elite lineage on their relationship with the Vijayanagara kings,
who, it is claimed, appointed several of their line as nayakas after they attracted
the attention of the Vijayanagara rulers through acts of daring and bravery (Put-
tanna 1924; Srikantaya 1941).
After the fall of the city of Vijayanagara in 1565, the Chitradurga family and
most other central Karnataka poligars soon declared their independence of the
remnants of that empire. They also were in the thick of the chronic regional
warfare that became one of the hallmarks of the Nayaka period. They warred
repeatedly with the Basavapatna or Tarikere poligars whose country lay to the
west, with the Harapanahalli poligars to the north at Uchchangidurga; with the
Harati or Nidugal family at Dodderi and later at Nidugal; with the Hatti family,
whose headquarters town was Nayakanhatti to the northeast; and with the Raya-
durga poligars, among others (Hayavadana Rao 1930 : 5 : 13801387; Ota 1999;
Sathyan 1967 : 5361). They enjoyed moderate success in regional warfare until
the mid-seventeenth century, when the Mughals invaded central Karnataka and
established the suba (major province) of Sira. Although the territories of many
former poligars in the region became parganas (lesser provinces) under the Mugh-
als, Chitradurga was one of the fortunate few to enjoy limited independence as a
tributary of Sira (Hayavadana Rao 1930 : 5 : 536537).
These events did not lessen the pace of internecine warfare, which continued
on the borders of Chitradurga and was increasingly directed at the Marathas, who
were pressing down from the north, and, when opportunity presented itself,
against the Mughals. Chitradurgas political position became untenable in the late
eighteenth century after it was caught between the Marathas and Haidar Ali of
Mysore in their struggle for control over central Karnataka. Haidar Ali nally
succeeded in taking Chitradurga Fort2 in 1779 and the region became a Mysore
province ( Rice 1897 : 1 : 504; Saletore 1940; Srikanta Sastri 1928). Madakari
270 asian perspectives . 42(2) . fall 2003

Nayaka the Last, the nal direct ruler of Chitradurga, and his family were impris-
oned at Shrirangapattana, where they died. To break the power of the Bedar caste
in the region, whose members had loyally supported the Chitradurga poligars,
Haidar Ali also is said to have moved more than 20,000 Bedars from Chitradurga
to Shrirangapattana, the young men of which were pressed into service in his chela
battalions3 (Bowring 1899 : 7275; Wilks 1989 : 743).
After the fall of Shrirangapattana in early May of 1799, Chitradurga remained a
province of Mysore under the Wodeyars, whom the British returned to power.
Today it is one of the 27 districts of the State of Karnataka. Until the late 1990s,
its poligar heritage was reected in the boundaries of Chitradurga District, which
encompassed most of the territory held by Madakari Nayaka the Last when Chi-
tradurga Fort fell to Haidar Ali.

data sources
Chitradurgas rich past rst captured international attention in the late nineteenth
century with the discovery of three Asokan edicts at Brahmagiri, Siddapura, and
Jatinga-Ramesvara in the far northeastern corner of the district ( Rice 1903 : 91,
93, 96). Excavations at Brahmagiri and Chandravalli, the latter of which lies
immediately west of Chitradurga, demonstrated early in the twentieth century
that the archaeological record of this region is temporally deep, signicant, and
little known (Krishna 1993a : 2327, 1993b : 2333; Nagarajan and Sundara 1985;
Settar 1976 : 99103; Wheeler 1948 : 180310). Krishna (1993a : 1314, 1993b :
2733, 199) published brief descriptive accounts of Chitradurga Fort and other
archaeological sites in this district in the early 1930s, and local historians, such as
M. S. Puttanna (1924) and H. S. Jois, devoted considerable attention to Chi-
tradurga and its poligars in the rst half of the twentieth century (Srinivas 1991;
Sujatha 1991).
However, since Puttanna and Jois published mostly in Kannada, their research
has had little impact outside of Karnataka. Sri H. S. Jois warrants additional men-
tion because he single-handedly created the archaeological museum housed in
Rangayyana Bagilu, one of Chitradurga Forts gateways, where it is still main-
tained by the Karnataka Directorate of Archaeology and Museums (KDAM).
Three main factors have inhibited archaeological research on the Nayaka
period in Karnataka. First, the Nayaka period is essentially modern times in India,
and, until recently, many archaeologists viewed it as lying outside the domain of
archaeological enquiry. Second, archaeological reconnaissance survey coverage in
South India is spotty and no agencies act as central repositories for site location
information. Third, access to aerial photos and high-resolution digital imagery is
constrained greatly, if not eectively prohibited, by national security concerns.
The immediate consequence is that, not only do archaeologists know little about
the period, they cannot readily turn to many of the kinds of data that are viewed
as essential to archaeological spatial studies elsewhere.
To mitigate the eects of these problems, we began archaeological reconnais-
sance surveys in Chitradurga District in 1996, with emphasis on Nayaka period
fortied sites (Lewis 1997; Patil 1999). The eldwork has been augmented by
archival research on eighteenth- and early nineteenth-century EIC and Indian
 lewis and patil . spatial patterns of a nayaka successor state 271

records (Lewis 2000, 2001). These complementary lines of research helped to

enhance our understanding of the central Karnataka poligars in at least two ways.
First, it enabled us to begin constructing a highly detailed geographic informa-
tion system (GIS) model of the cultural landscape of the Mysore kingdom as it
existed between 1799 and 1808, immediately after the death of Tipu Sultan. The
model combines the results of modern reconnaissance surveys with information
gleaned from large-scale (typically 2 mi to the inch, or 1.27 km to the centimeter)
unpublished EIC maps of the late eighteenth and early nineteenth centuries
(Fig. 2). Analyses of this model, which is complete for Chitradurga but still under
development for the rest of Mysore, facilitate the delineation of important spa-
tial relationships and provide something that is currently unattainable by other
meansa virtual archaeological site le of thousands of contemporaneous Nayaka
cultural features, including towns, villages, roads, reservoirs, and their place names.
Second, it permits us to reconstruct the major features of individual forts, based
on the integration of modern archaeological survey data and EIC fort inventories
that report building-level construction details. The meticulous detail of the EIC
fort inventories enable a comparative analysis of their overall designs, major spatial
divisions, siege supply storage systems, and even identify the names and intended
uses of buildings that are recognizable today only as wall stubs or archaeological
features.4
The following sections report on the preliminary analysis of the major spatial
patterns of the Chitradurga region. The inferences we derive from these analyses
are properly viewed as working hypotheses to be examined in the eld and
against the results of our continuing archival research. Given the nature of the
historical cartographic data that informs most of this analysis, our focus is primar-
ily on economic and political spatial patterns.

chitradurga province
The former Chitradurga province (Fig. 1) lies in Karnatakas southern maidan
region, most of which is a high plateau of gently rolling scrub jungle. A belt of
hills ranges across the western half of the region in a general south to southwest
trend; another major hill range crosses the northeastern part of the province. The
climate is semi-arid due to the rain shadow eect of the Western Ghats, which
lie between Chitradurga and the Arabian Sea. The average annual precipitation is
only about 58 cm, half of which falls during the southwestern monsoon between
June and September (Sathyan 1967 : 2223). Drainage is provided by the Vedavati
and Tungabhadra rivers, which form part of the Krishna Basin and empty into the
Bay of Bengal.

Macrospatial Patterns
To delineate the major spatial patterns of the regions cultural landscape, we rst
describe Chitradurgas major administrative divisions and then examine its ele-
ments. In 1800 Chitradurga was divided into 12 parganas, each of which had a
centrally located major town (Fig. 1). The parganas were roughly of the size of
a modern taluk (a subdivision of a district), but of somewhat greater adminis-
Fig. 2. Example of Mysore GIS details, showing towns, villages, roads, tanks, and major streams in
the eastern part of Holalkere pargana as it existed in 18001801; digitized from unpublished Mysore
Survey maps. However, a GIS is far more than a mere map. What this gure cannot show is that for
each town, village, tank, river, road, and other feature depicted on the map, there is a corresponding
database entry that contains the values for such diverse variables as annual revenue yield, the pres-
ence or absence of a defensive ditch, and stream rank order.
 lewis and patil . spatial patterns of a nayaka successor state 273

trative importance. Although pargana is a Mughal administrative term (Fox 1971;

Richards 1993), comparable local divisions of the landscape existed before Vijaya-
nagara times.
Our analysis suggests that pargana boundaries were drawn on the basis of a wide
range of economic and social concerns that centered on the major towns of the
province. This inference is based on two lines of evidence, (1) the historical con-
tinuity of pargana boundaries, and (2) the congruence of these boundaries with
Thiessen polygons5 centered on the major towns.
While there is little reason a priori to anticipate that local administrative boun-
daries should endure over the centuries, roughly one-third of the modern taluks
of Chitradurga District follow much the same lines as the late eighteenth-century
provincial parganas. The survival of these boundaries after more than 200 years
of profound social, economic, and political change in South India, supports the
view, for which there is considerable external evidence (e.g., Fox 1971; Gordon
1994), that changes of government have often had little impact on low-level
administrative structures in India.
Why should local boundaries endure? Surprisingly, the explanation may rest
more with geography than human agency. Inspection of a network of Thiessen
polygons centered on the provinces major towns, as they existed in 1800, shows
that many, but by no means all, pargana boundaries lay roughly halfway between
a given town and its neighbors (Fig. 3). The pattern is consistent with boundaries
that dene service areas of major towns rather than the limits of territories carved
out by local elites or the organization of revenue collection (i.e., on interaction
patterns based more on simple physical proximity than on at). If true, it further
implies that there was relatively little local-level administrative authority or power
concentrated in the parganas. If the boundaries of these units were primarily terri-
torial or based on revenue yield, it is reasonable to expect that their extent would
clearly reect the relative political power of local elites and local economic fac-
tors, and not be explicable as a function of the distance that separates each town.
That the latter is sucient to explain the patterns based on the data at hand
encourages us to explore this further in our archival research and archaeological
surveys.
Within the province as a whole, the general settlement pattern delineated by
the Mysore Survey was minimally a four-tiered hierarchy of 722 villages, 65
minor towns, 11 major towns, and the provincial capital of Chitradurga Fort. The
capital and its immediate hinterland formed one of the provinces 12 parganas.
The other 11 major towns, all of which were well fortied and built to withstand
sustained sieges, were the economic, social, and religious centers of their respec-
tive parganas. A rough grid of unpaved roads provided the primary means of
communication between the 12 major towns. Seven of the provinces major
towns occupied nodal positions in this grid, and Chitradurga Fort lay at the
intersection of no less than nine roads. Prior to 1779, some of the parganas were
governed by administrators appointed by the Chitradurga nayaka; in a few cases,
however, lesser nayakas of local lineages were appointed to these posts. Under
Haidar Ali and Tipu Sultan, all of the major towns were placed under the control
of commanders appointed from Shrirangapattana.
Within each pargana were 39 minor towns, and all but three of the 65 identi-
ed minor towns were fortied. In marked contrast to the pattern described
Fig. 3. Thiessen polygons based on the major towns, showing the general congruence with pargana
boundaries.
 lewis and patil . spatial patterns of a nayaka successor state 275

above for the major towns, Thiessen polygons centered on the minor towns show
no relationship with pargana boundaries or with the boundaries of modern taluks.
In other words, the distribution of minor towns does not vary with simple fac-
tors of geography. It does, however, respond to the proximity of major towns.
Viewed simply as areas of inuence, the polygons dened on the minor towns
also are notable in that none contain more than one major town. The pattern
suggests the spatial distribution of minor towns is at least partly a function of
the distribution of the major towns they served, or vice-versa. Support for this
inference comes from the subsequent history of the relationship between individ-
ual pairs of major and minor towns. For at least three of the 12 pairs, the late
eighteenth-century minor town now over-shadows, if not completely replaces, its
former nearby major town.6 No single factor accounts for the rise and fall of the
regional importance of these locations.
The 18001801 Mysore Survey map of Chitradurga province also records
722 named villages, all of which appear to have been nucleated settlements, a
typical maidan village pattern that persisted into the twentieth century (Buchanan
1807 : 32; Hayavadana Rao 1930 : 1 : 367). When viewed across time, it is evident
that Nayaka villages diered from towns in several striking ways. First, many vil-
lages did not show the same enduring continuity of settlement that characterizes
most of the towns. With few exceptions, the major and minor towns that were
inhabited between 1799 and 1808 are still inhabited today. However, the villages
have proved to be more ephemeral; many of the 722 villages that we have digi-
tized from the Mysore Survey maps cannot be cross-referenced to village loca-
tions on modern Survey of India topographical maps. Second, while historical
dierences in town locations appear to reect the force of political and economic
changes over the intervening centuries, dierences at the village level are more
easily explained as a result of such basic factors as the removal of any need for vil-
lage defense, new reservoirs (e.g., the 8700 ha Vanivilas Sagar Reservoir and the
1200 ha Gayatri Reservoir, both of which are in the southern half of the district),
the proliferation of tubewells, the increased availability of all-weather roads, and
changes in the regions agriculture (Green 1846; Hayavadana Rao 1930 : 3 : 156
174; Sathyan 1967 : 119120).
Roads and reservoirs exerted greater pull on village location choices than
the small stream valleys that drain the region, largely, we suspect, because most
streams were dry for much of the year, just as they are today.7 Roads provided
the major communication lines and reservoirs oered consistent water supplies.
Of the 65 minor towns in the province, 35 (54 percent) were located within
3 km of a reservoir in 1801, and 26 of these 35 minor towns were less than 1 km
from a reservoir.
Another consistent pattern of village site distribution was to avoid all but the
fringes of the hill ranges that dot the province. The notable exceptions to this
generalization are isolated hills, the tops of many of which were crowned by forts,
temples, or both, and were typically attended by small nearby villages.
In summary, the pargana boundaries of late eighteenth-century Chitradurga
province appear to reect the enduring nature of towns in the region and a cer-
tain level of concern for minimizing travel and transportation costs rather than
power struggles between local elites or an organizational structure imposed from
above. Relatively little administrative authority or power appears to have been
276 asian perspectives . 42(2) . fall 2003

held by pargana-level institutions. The provincial settlement pattern was hierar-

chical and suciently stable to be recognizable today in the modern Chitradurga
landscape. Evidence of the hierarchical integration of major and minor towns can
be seen in the replacement of at least three major towns by their closest minor
town during the past two centuries.

Microspatial Patterns
Chitradurga Fort Even if archaeologists lacked historical context for the region,
site survey would readily demonstrate that Chitradurga Fort is a place apart (Fig.
4). Evidence of its occupation dates back for several thousand years and, before
the advent of the railways in the late nineteenth century, it commanded the single
most important transportation node in the region. This fortied town and hill
fort was the capital of a Nayaka kingdom and, later, a favored provincial head-
quarters town of the Mysore sultanate. It also was an important material symbol
of legitimacy in the region. Whoever held Chitradurga Fort, held the key to the
province.
Among the characteristics that set it apart from the other major towns of the
province were large palaces, courtly architecture, swings and other architectural
trappings of kingship, the latter of which are set in areas dedicated to ritual and
royal display (Fig. 5; Wagoner 2001). The combination of diering architectural
traditions is particularly evident in the multiple fortication lines, where one nds
French- and English-designed ramparts and parapets punctuated by prominent
gateways that combine South Indian and Bahmani-derived elements. Chitradurga
Fort also contains the largest and most well-endowed temples of the province, a
connection that once enhanced the status and legitimacy of this town. Its status
remained unchallenged from the Nayaka period until in the late 1990s, when the
former minor town of Davangere had grown so much larger in population and
economic inuence that it became the headquarters town of a new district carved
out of Chitradurga.
Although it is one of the best preserved Nayaka period forts in South India,
Chitradurga Fort is archaeologically unknown except for Krishnas (1993a : 14
22, 5065, 199) brief description of its major monuments in the late 1920s and
Wagoners (2001) study of its entrance pavilions. C. S. Patil recently completed
a detailed archaeological survey of this site, which will be an integral part of
the Chitradurga monograph that we are now preparing for publication. Our
research on Chitradurga Fort has been aided signicantly by the discovery of
a 240-page manuscript (Barclay 1802) that gives a wall-by-wall, building-by-
building description of the construction, condition, and current and former use of
every building, wall, gate, and tower in the fort, as well as an assessment of what
steps must be taken to make the fort defensible against assault. For example, the
Chitradurga Fort inventory describes the mud brick circular buildings depicted
in Figure 6 as follows: No. 31, 32, 33, 34, 35, granaries. Circular buildings, 12
feet diameter, walls mud, roof mud terrace thatched; Present State: unservice-
able; Where Situated: near the upper palace (Barclay 1802 : 246247). The fort
inventory manuscript, Patils survey, and Colin Mackenzies 1800 plan of Chi-
tradurga Fort, enable us to reconstruct a highly detailed and accurate picture of
Fig. 4. Detail from a plan of Chitradurga Fort drawn by Colin Mackenzie, dated June 1800. (By permission of The British
Library, IOR prints WD 2634.)
278 asian perspectives . 42(2) . fall 2003

Fig. 5. Royal display area of the inner fort at Chitradurga Fort. The gateway in the right foreground
leads to Hidimbeswara temple, one of the oldest temples on the hill. A monolithic pillar and two
swing frames lie between the entrance to this gateway and the Sampige Siddheshvara temple, which
rests at the foot of the hill in the left background. View is to the south.

this poligar headquarters town as it existed at the beginning of the nineteenth

century.
Major Towns Site surveys of the 11 other major towns identied on the 1800
1801 Mysore Survey map of Chitradurga province suggest that, while they are
large and complex sites, they diered in importance, population, area, fortica-
tion complexity, ease of communication, public architecture, siege storage capac-
ity, and other dimensions from Chitradurga Fort. Although the fortied areas of
such sites as Hosadurga and Kanakuppa (Fig. 1), are nearly as large as that of
Chitradurga Fort, one would search in vain within their walls for the remains
of anything that resembles the palaces, royal display areas, and combination of
extra-regional architectural traditions that are distinctive of Chitradurga Fort. The
absence of the material trappings of kingship is consistent with the subsidiary
role ascribed to the major towns by inscriptions and other contemporaneous
documents.
Nevertheless, the major towns were more similar to Chitradurga Fort and to
each other than any of them were to the minor towns. Archaeological site surveys
demonstrate that the major towns typically possessed as many as two to three for-
tication lines and a considerable investment in siege store facilities. While they
lacked the well-endowed major temples that were one measure of the high status
of Chitradurga Fort, they were rich and populous enough to support temples. At
 lewis and patil . spatial patterns of a nayaka successor state 279

Fig. 6. Nayaka period granaries and storehouses in the Upper Palace area of Chitradurga Fort. View
is to the northwest.

Kanakuppa and Hosadurga, both of which were abandoned more than a century
ago, some temples continue to be used for worship today.
One striking fact revealed by our site survey is that roughly one-half of the
major towns have not survived the past two centuries well. Colin Mackenzie
(1803), for example, mapped substantial forts with well-built walls, bastions, and
ditches at Tallak and Mayaconda in 1801. These defensive works have been so
thoroughly recycled into modern buildings and shops that no visible evidence of
their former existence can now be seen on the ground except for the ditches that
formerly encircled the town walls.
Minor Towns and Villages Like the major towns, all but three of the 65 minor
towns identied on the Mysore Survey maps in Chitradurga province were for-
tied, but how they were fortied diered considerably from town to town.
Simply put, the fortications of minor towns share many stylistic elements with
the major towns, but the combination of these elements and their expression
in the design of minor town defenses were diverse. At Aimangala (Fig. 7c), for
example, which is located on a gently rolling plain about halfway between Hir-
iyur and Chitradurga, the town walls dene a 1.7-km perimeter, which would
have required a much larger garrison to defend than a small town could have
280 asian perspectives . 42(2) . fall 2003

Fig. 7. Diversity of minor town defenses: A: locations of Aimangala and Bharmagiri in Chitradurga
province; B: Bharmagiri plan (contour lines are approximate and intended only to give an impres-
sion of the topographical relationship between the parts of the town walls); C: Aimangala plan.

supported. The three bent entrances to Aimangala resemble hundreds of other

South Indian town gateways of this period, but the battered curtain walls and
bastions, which are trapezoidal in plan, are unique among the forts of the prov-
ince. About 35 km to the south-southwest of Aimangala, the remains of Bharma-
giri, another Nayaka period minor town, show a completely dierent expression,
 lewis and patil . spatial patterns of a nayaka successor state 281

in which the builders crowned the crests of two adjacent hills with square towers
and then built the town between walls on the saddle between the towers (Fig.
7b). While some of the diversity of these town defenses can undoubtedly be
accounted for by their having been built at dierent times during the past few
hundred years, much of it appears to reect local decisions and designs. As was
noted earlier in the discussion of the spatial patterning of pargana boundaries,
lower level administrative institutions appear to have enjoyed considerable auton-
omy from the provincial ruling elites.
The 722 settlements that were identied as villages on the 18001801 Mysore
Survey map also were quite variable. Although few villages are delimited by fort
symbols on this map, it is clear from site surveys, other Mysore Survey documents
and maps, and from contemporary accounts (e.g., Buchanan 1807 : 304, 310) that
most villages made at least some attempt to protect their families and property
from marauders. The larger villages were typically surrounded by walls with
towers in the angles and an outer perimeter of dense hedges of thorns; smaller
communities often based their defense on thorn hedges and a single stone tower,
or hude, in the middle of the village. The remains of mud and rubble village walls
punctuated by relatively low and narrow gateways are the most enduring and
obtrusive archaeological evidence of Nayaka habitation site locations identied in
reconnaissance surveys.
In summary, Chitradurga Fort possesses many features that set it apart from the
other towns of the region, and it would easily be identiable as a principal place
even if historical documentation were absent. Only the most important towns of
the province were prepared to resist assault by an army, but, among these sites,
only Chitradurga Fort shows the necessary investment of resources to withstand
attack by a disciplined army with siege guns. The other major towns are less well
fortied, their temples are less prosperous, and they lack architecture associated
with royal display. All of these towns, however, are more similar to each other
than they are to the minor towns of the region. The latter show highly variable
plans, diverse defensive works, and locations that are often far removed from
major communication routes. Villages were just as diverse as the minor towns, if
not more so. Like contemporaneous maidan settlements elsewhere in Karnataka,
villages were nucleated and designed to protect the persons and, if possible, the
property of the inhabitants. The major factors that determined Nayaka habitation
site locations appear to have been history, road proximity, water sources, arable
land, and defense.

conclusion
As the Chitradurga example demonstrates, Nayaka archaeology can be as much
about archival research as about archaeology. Both are essential to the under-
standing of these turbulent centuries of early modern South Indian history. The
Mysore poligars are interesting because they represent historically documented
instances of the emergence of small states that strived for independence in a
region that had been thrown into anarchy by the collapse of the larger state
( Vijayanagara in this case), of which they had formerly been a part. A broad out-
line of the many political, economic, and social factors that played important roles
in the rise and fall of the Mysore poligars is given by the historical record. How-
282 asian perspectives . 42(2) . fall 2003

ever, even after the documentary record has been thoroughly examined (and
much remains to be done in this area, e.g., Desikachar 1977), important pieces of
the picture would be missing if archaeological research were not available to pro-
vide the spatial and material contexts that are essential to a complete interpreta-
tion of these data.
By digitizing 200-year-old, large-scale maps of the Mysore kingdom, we have
assembled a virtual archaeological site le of nearly 1000 Nayaka period site loca-
tions in Chitradurga province. Roughly 30,000 more locations in the Mysore
kingdom have yet to be digitized. This represents a ood of fresh information, the
accuracy of which we have only begun to conrm through eld survey.
The Chitradurga data enable us to delineate several of the major spatial patterns
of one Nayaka polity at the beginning of the nineteenth century. We emphasize
gross economic and political patterns in this article because these aspects are
readily identied through the integrated analysis of historical cartography and
archaeological reconnaissance survey, which are the primary data we use here.
Nevertheless, it is a work in progress. Many other aspects of the Chitradurga cul-
tural landscape (e.g., climatic, geographical, and ritual landscapes) have yet to be
examined.
The important themes that emerge from this analysis are the resilience of
lower-level administrative structures in the parganas, and such enduring concerns
as defense and communication. The distribution of major towns appears to reect
their role as trade and service centers for nearby small towns and villages, with
Chitradurga Fort occupying the central nodal position for the entire province.
This primarily economic interpretation of regional patterning is also supported by
the tendency for minor towns to occasionally overtake and replace their closest
major town. A further implication of the pattern is that the internal boundaries of
the province may not delineate territories carved out of the region by the political
and military might of competing poligar lineages. Rather, they may simply reect
the limits of contiguous, but otherwise largely self-contained service areas that
were nominally controlled by local elites.

acknowledgments
This article is based on a paper presented at the 66 th Annual Meeting of the Society
for American Archaeology, New Orleans, Louisiana, 1822 April 2001. We grate-
fully acknowledge the generous research support of the American Institute of Indian
Studies, the UIUC Research Board, and Intel Corporation. Jagadeesh C. and Shan-
kar Athani, both of Chitradurga, and Venkata Naik of Mysore, made many con-
tributions to every stage of the eld research. The data used in this article are on
le with the Karnataka Directorate of Archaeology and Museums and the UIUC
Department of Anthropology.

notes
1. Nayaka, with the rst letter capitalized, refers to the early modern period of that name in
South Indian history. As used here, nayaka, in lower case, can mean a local military governor
appointed by a higher king or a little king who traces the legitimacy of his rule to an ancestor
who held such an oce in the same headquarters town or general region. As Wilson (1968 : 372)
explains, the meaning varied considerably across India. In the study region, it is sometimes used
as a synonym for poligar.
 lewis and patil . spatial patterns of a nayaka successor state 283

2. The name Chitradurga refers both to the province and its main city. To minimize possible con-
fusion, our use of the name applies only to the province; by the term Chitradurga Fort we mean
the city and fort that form the headquarters town of that province.
3. In Anglo-Indian usage, a chela battalion was a military unit comprised of slaves, prisoners, and
converts ( Yule and Burnell 1903 : 190). James Scurrys (1831) account of his experiences as a
slave-soldier in one of Mysores chela battalions gives a rst-hand account of the brutal treatment
these soldiers received. Scurrys book also includes several chapters on the years his battalion was
assigned to Chitradurga Fort.
4. Our monograph on Chitradurga Fort, which is in preparation, will be the rst major intrasite
spatial study to incorporate these data.
5. Thiessen polygons are dened around points in a plane by connecting lines drawn so that they lie
halfway between each point and its neighbors. The resulting polygons delineate service areas
or areas of inuence for each point (DeMers 2000 : 305307) based simply on the implicit
assumption that interaction with a given point increases with ones proximity to that point. The
use of Thiessen polygons in archaeology has had a somewhat speckled past. First applied in
archaeology in the 1960s and 1970s, critics took the concept to task for not considering signi-
cant social and geographical factors known to be at work in human interactions (e.g., Haselgrove
1986). The recent availability of powerful desktop GIS software applications, which typically
include tools to generate Thiessen polygons, encourage archaeologists to reconsider the analytical
value of this and similar tools, such as cost surface analysis, as estimates of spatial interaction
(Wheatley and Gillings 2002 : 151159). The data needed for a cost surface analysis of the spatial
relationships between parganas are unavailable.
6. The town pairs include Jagalur, which replaced Kanakuppa; Challakere, which replaced Dodderi;
and Davangere, which replaced Harihar.
7. It is also possible that the 17991808 Mysore Survey maps are biased against sites that were lo-
cated far o the roads. Future eld reconnaissance survey in Chitradurga District will check
this possibility.

references cited
Aiyar, R. S.
1991 History of the Nayaks of Madura. Reprint of 1924 edition. New Delhi: Asian Educational
Services.
Barclay, R.
1802 Report on the fort of Chittledroog. British Library, Oriental and India Oce Collections,
Board of Control, IOR/F/4/201/4575.
Bowring, L. B.
1899 Haidar Ali and Tipu Sultan and the Struggle with the Musalman Powers of the South. Oxford:
Clarendon Press.
Brittlebank, K.
1997 Tipu Sultans Search for Legitimacy: Islam and Kingship in a Hindu Domain. Delhi: Oxford
University Press.
Buchanan, F.
1807 A Journey from Madras through the Countries of Mysore, Canara, and Malabar. 3 vols. London.
Chitnis, K. N.
1974 Keladi Polity. Dharwar: Karnatak University.
2000 The Nawabs of Savanur. New Delhi: Atlantic.
Deloche, J.
2000 Senji Ville Fortiee du Pays Tamoul. Paris: Ecole Francaise dExtreme-Orient.
DeMers, M. N.
2000 Fundamentals of Geographic Information Systems. 2nd ed. New York: John Wiley.
Desikachar, S. V.
1977 Archival wealth and post-Vijayanagara Karnataka history. The Indian Archives 26(12) :
118.
Dirks, N. B.
1982 The pasts of a Palaiyakarar: The ethnohistory of a South Indian little king. Journal of Asian
Studies 41 : 655683.
284 asian perspectives . 42(2) . fall 2003

1994 The Hollow Crown: Ethnohistory of an Indian Kingdom. 2nd ed. Ann Arbor: University of
Michigan Press.
Dua, J. C.
1996 Palegars of South India: Forms and Contents of their Resistance in Ceded Districts. New Delhi:
Reliance Publishing House.
Edney, M. H.
1997 Mapping an Empire: The Geographical Construction of British India, 17651843. Chicago:
University of Chicago Press.
Fox, R. G.
1971 Kin, Clan, Raja, and Rule: State-Hinterland Relations in Preindustrial India. Berkeley: Uni-
versity of California Press.
Gordon, S.
1994 Kinship and pargana in eighteenth-century Khandesh, in Marathas, Marauders, and State
Formation in Eighteenth-Century India: 122150, by S. Gordon. Delhi: Oxford University
Press.
Green, C.
1846 Memorandum on public works in Mysore, in Selections from the Records of the Mysore Com-
missioners Oce: 118 (paginated separately), by Oce of the Commissioner. Bangalore:
Mysore Government Press.
Haselgrove, C. C.
1986 Central places in British Iron Age studies: A review and some problems, in Central Places,
Archaeology and History: 312, ed. E. Grant. Sheeld, UK: University of Sheeld.
Hayavadana Rao, C.
1930 Mysore Gazetteer. New ed. 5 vols. Bangalore: Government Press.
Krishna, M. H.
1993a Annual Report of the Directorate of Archaeology and Museums in Karnataka for the Year 1929.
Mysore: Karnataka Directorate of Archaeology and Museums.
1993b Annual Report of the Directorate of Archaeology and Museums in Karnataka for the Year 1930.
Mysore: Karnataka Directorate of Archaeology and Museums.
Lewis, R. B.
1997 Forts of the Poligar chiefdoms, Chitradurga District, Karnataka, India. Paper presented at
the 26th Annual Conference on South Asia, 1619 October 1997, Madison, WI.
2000 The Mysore parganas as expressions of regional identities, a.d. 18001808. Paper pre-
sented at the 29th Annual Conference on South Asia, Madison, WI.
2001 The Mysore kingdom at a.d. 1800: An archaeological analysis of the Mysore Survey of
Colin Mackenzie. Paper presented at the 16th International Conference on South Asian
Archaeology, 26 July 2001, Paris, France.
Mackenzie, C.
1803 Remarks on the forts on the northern frontier of Mysore in 1800 and 1801 (extracted
from the papers of the Mysore Survey), prepared by Capt. Colin Mackenzie of the Engi-
neers and Superintendent Mysore, Fort St. George, 12 January 1803. University of
Southampton, Hartley Library, Wellington Papers 3/3/100.
Michell, G.
1995 Architecture and Art of Southern India: Vijayanagara and the Successor States. The New Cam-
bridge History of India 1 : 6. Cambridge: Cambridge University Press.
Nagarajan, K. G., and A. Sundara
1985 M. H. Krishna: His Life, Work and Contribution to Karnataka Archaeology. Mysore: Karna-
taka Directorate of Archaeology and Museums.
Narayana Rao, V., D. Shulman, and S. Subrahmanyam
1992 Symbols of Substance: Court and State in Nayaka Period Tamil Nadu. Delhi: Oxford Univer-
sity Press.
Ota, N.
1999 Beda nayaka and their historical narratives in Karnataka during the post-Vijayanagara
period, in Kingship in Indian History: 163194, ed. N. Karashima. Japanese Studies on
South Asia No. 2. New Delhi: Manohar.
 lewis and patil . spatial patterns of a nayaka successor state 285

Patil, C. S.
1999 Karnataka Kotegalu [Forts of Karnataka]. Hampi: Kannada University.
Phillimore, R. H.
1945 Historical Records of the Survey of India. 3 vols. Dehra Dun: Survey of India.
1958
Puttanna, M. S.
1924 Chitradurgada Paleyagararu [Chitradurga Poligars]. Bangalore.
Rajayyan, K.
1974 Rise and Fall of the Poligars of Tamilnadu. Madras: University of Madras.
Ramachandra Rao, P. B.
1943 The Poligars of Mysore and Their Civilization. Teppakulam: Palaniappa Bros.
Richards, J. F.
1993 The Mughal Empire. The New Cambridge History of India, vol. 1, pt. 5. Cambridge:
Cambridge University Press.
Rice, B. L.
1897 Mysore: A Gazetteer Compiled for Government. Revised ed. 2 vols. London: Archibald
Constable and Company.
1903 Inscriptions in the Chitaldroog District. Epigraphia Carnatica, vol. 11. Bangalore: Mysore
Government Central Press.
Robb, P.
1998 Completing our stock of geography, or an object still more sublime: Colin
Mackenzies survey of Mysore, 17991810. Journal of the Royal Asiatic Society, series 3,
8 : 181206.
Saletore, R. N.
1940 The Conquest of Citradurga by Hyder Ali. Quarterly Journal of the Mythic Society
29 : 171188.
Sathyan, B. N., ed.
1967 Mysore State Gazetteer: Chitradurga District. Bangalore: Government Press.
Scurry, J.
1831 The Captivity, Suerings, and Escape of James Scurry, Who was Detained a Prisoner during Ten
Years, in the Dominions of Hyder Ali and Tippoo Saib. London: Henry Fisher.
Settar, S., ed.
1976 Archaeological Survey of Mysore, Annual Reports: 19061909, A Study. Aids to Research
Series No. 1, vol. 2. Dharwar: Karnatak University.
Srikanta Sastri, S.
1928 Capitulation of Chitradurga. Quarterly Journal of the Mythic Society 18 : 145154.
Srikantaya, S.
1941 Chitaldrug. Quarterly Journal of the Mythic Society 31 : 339356.
Srinivas, M. V.
1991 Hullur Srinivasa Jois, in Historiography of Karnataka: 231235, ed. S. U. Kamath. Banga-
lore: The Mythic Society.
Stein, B.
1985 State formation and economy reconsidered. Modern Asian Studies 19 : 387413.
Sujatha, H. S.
1991 M. S. Puttanna and his contributions to the history and culture of Karnataka, in Histori-
ography of Karnataka: 235242, ed S. U. Kamath. Bangalore: The Mythic Society.
Swaminathan, K. D.
1957 The Nayakas of Ikkeri. Madras: P. Varadachary.
Vriddhagirisan, V.
1995 The Nayaks of Tanjore. Reprint of 1942 ed. New Delhi: Asian Educational Services.
Wagoner, P. B.
2001 Kannada Kalla Upparige (Stone Palace): Multistoried entrance pavilions in pre- and early
Vijayangara architecture. Ars Orientalis 31 : 167183.
286 asian perspectives . 42(2) . fall 2003

Wheatley, D., and M. Gillings

2002 Spatial Technology and Archaeology: The Archaeological Applications of GIS. London: Taylor
and Francis.
Wheeler, R.E.M.
1948 Brahmagiri and Chandravalli 1947: Megalithic and other cultures in Mysore State. Ancient
India 4 : 180310.
Wilks, M.
1989 Historical Sketches of the South of India in an Attempt to Trace the History of Mysore. 2 vols.
Reprint of 1930 ed.; originally published in 1810. New Delhi: Asian Educational Serv-
ices.
Wilson, H. H.
1968 A Glossary of Judicial and Revenue Terms, and of Useful Words occurring in Ocial Documents
relating to the Administration of the Government of British India. 2nd ed. Delhi: Munshiram
Manoharlal.
Yule, H., and A. C. Burnell
1903 Hobson-Jobson: A Glossary of Colloquial Anglo-Indian Words and Phrases, and of Kindred
Terms, Etymological, Historical, Geographical and Discursive. London: John Murray.

abstract
South Indian historical accounts, inscriptions, and literature describe many attempts
at state formation and political independence by local elites between the collapse of
Vijayanagara control of its heartland in the sixteenth century and the eighteenth-
century rise of British hegemony in the south. Most of these elites, known variously
as little kings, nayakas, and poligars, did not survive long in the political turmoil of
the era. Those that did endure into the early nineteenth century were soon margi-
nalized or removed from power by East India Company policies and direct military
intervention. Archaeological site surveys, guided by contemporaneous East India
Company manuscript maps, fort inventories, building plans, and other records,
enable researchers to reconstruct and interpret major spatial patterns of the cultural
landscapes of these small polities. This approach, which the authors are currently
applying in their investigations of the Mysore kingdom, yields a fresh perspective of
South Indian little kings and chiefs that complements the work of historians and
contributes signicantly to the understanding of the nature of these polities. This
article describes a case study of the major spatial patterns of the Chitradurga poligars
of central Karnataka, as they were in the late eighteenth and early nineteenth cen-
turies. The results provide a fresh perspective on this poligar province and illustrate
the signicant interpretative value of contemporaneous colonial documents for
archaeological site survey and spatial analyses in India. Keywords: India, Mysore,
Nayaka period, poligars, spatial analysis, colonial archives, archaeological site survey.