ISBN 978-4-902325-61-4

Occasional Paper 10

Linguistics, Archaeology
and
the Human Past

Edited by
Toshiki OSADA and Akinori UESUGI

Indus Project
Research Institute for Humanity and Nature
Kyoto, Japan
2011

Occasional Paper 10

Linguistics, Archaeology
and
the Human Past

Edited by
Toshiki OSADA and Akinori UESUGI

Indus Project
Research Institute For Humanity And Nature
Kyoto, Japan
2011

Occasional Paper 10: Linguistics, Archaeology and the Human Past

Editor: Toshiki OSADA and Akinori UESUGI
Copyright Indus Project, Research Institute for Humanity and Nature 2011
ISBN 978-4-902325-61-4

Indus Project, Research Institute for Humanity and Nature (RIHN)

457-4 Motoyama, Kamigamo, Kita-ku, Kyoto 603-8047 Japan

Tel: +81-75-707-2371

Fax: +81-75-707-2508

E-mail: osada@chikyu.ac.jp

Printed by Nakanishi Printing Co. Ltd., Kyoto, Japan

CONTENTS

Preface
Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

Kuldeep K. Bhan

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

27

P. Ajithprasad, SV Rajesh and Bhanu Sharma

Annual report of excavation at Kanmer 2007-08 and 2008-09

71

J. S. Kharakwal, Y.S. Rawat and Toshiki Osada

Report on the faunal remains recovered from Kanmer, Gujarat,

during the third field season (2007-08)

105
Pankaj Goyal and P.P. Joglekar

Preliminary anthracological analysis from Harappan Kanmer:

Human-environment interactions as seen through fuel resources exploitation and use

129

Carla Lancelotti and Marco Madella

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

(Field seasons 2006-2009)

143
Shahida Ansari

Memorial stones at Kanmer, Gujarat, India

175

Tilok Thakuria, J.S. Kharakwal, Toshiki Osada and Y.S. Rawat

Preface
I am very glad to announce the publication of the tenth volume of Occasional Paper (henceforth,
OP). I sincerely thank for all the contributors of this volume.
Our excavations at Kanmer (Kachchh, Gujarat) and Farmana (Haryana), both in India, were
completed by March, 2009. The first report of the Kanmer excavation was published in the second
volume of OP, and was reprinted in the second volume of Indus Civilization: text and context
(Manohar). The second report was published in the fifth volume of OP and was reprinted in the
third volume of RIHN-Manohar Indus project series titled Current Studies on the Indus Civilization
(Manohar). This OP includes the third and fourth report of the Kanmer excavations during 20072009.
This OP consists of seven papers: the first two papers are written by Professors Kuldeep Bhan and P.
Ajithprasad, both from Department of Archaeology and Ancient History, the Maharaja Sayajirao
University of Baroda, India. Professor Bhan was a visiting professor at the International Research
Institute for Japanese Studies (IRIJS) in Kyoto from July 2008 to March 2009. During this period
he collaborated with Professor Uno who is one of our active core members based at IRIJS. Professor
Bhan contributes a paper on ethnoarchaeological approach toward pastoralism in the Late Harappan
Gujarat. Subsistence systems of the Indus Civilization are one of the key topics we have been
pursuing. To fully understand them, it is very important to know about pastoralism in addition to
different agricultural systems.
Professor P. Ajithprasad was a visiting fellow at our Institute for three months in 2008. He writes
a paper titled Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat.
This paper is based on his own research with his students before our Indus project started. Since
then, Professor Ajithprasad and our project members have conducted field research exploring
archaeological sites, including Harappan sites in India, in order to collect more extensive data. We
plan to publish a handbook on the excavated Harappan sites, based on our collaborative research,
next year.
The remaining five papers deal with the outcome of excavations at Kanmer. Dr. J.S. Kharakwal of
Rajasthan Vidyapeeth, Dr. Y.S. Rawat of Archaeological Department of Gujarat State Government
and myself act as director and co-directors of Kanmer Archaeological Research Project. We
collectively overview the main results of excavations during the two seasons (2007-2008 and
2008-2009). The other four papers deal with specific topics. Dr. P.P. Joglekar and Mr. P. Goyal

of Deccan College give us a report on the faunal remains. Dr. C. Lancelotti and Dr. M. Madella
of Department of Archaeology and Anthropology, Instituci Mil i Fontanals, Spain, contribute
a paper on preliminary anthracological analysis. The paper by Dr. Shahida Ansari focuses on
ethnoarchaeological study at Kanmer. Mr. T. Thakuria and others write on memorial stones which
they recorded at the village of Kanmer.
In this OP we happen to have three papers on ethnoarchaeological study one by Professor Bhan
and the second by Dr. Ansari and the third by Mr. Thakuria. India has a very long tradition in this
approach, and I am very impressed that Indian scholars still maintain this important tradition.
Anthracological analysis by Drs. Lancelotti and Madella, on the other hand, presents quite a new
approach to the investigation of the palaeo-environment of the Indus Civilization. According to
them, (t)he anthracological evidence also suggests --- that the water regime of the Rann might
have been different from today, with higher water levels or longer periods of inundation. Their new
approach thus helps us better understand the palaeo-environment of Kanmer, and of Kachchh in
general, which is a very important topic for our project.
Lastly, besides the contributors of this volume, I would like to express my gratitude to Dr. Uesugi
who did fine editorial work on the whole volume.

Toshiki OSADA

Project leader and Professor

Research Institute for Humanity and Nature

Kyoto, Japan

osada@chikyu.ac.jp

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

Pastoralism in Late Harappan Gujarat, western India:

an ethnoarchaeological approach
Kuldeep K. Bhan
Department of Archaeology and Ancient History
Maharaja Sayajirao University of Baroda

Abstract
The modern state of Gujarat has emerged as a very important area for understanding the Harappan Tradition and especially the later
phase of Harappan Period. The paper reviews the archaeological data of Late Harappan Gujarat in the light of ethnographic data of
pastoral communities of Gujarat. It has been argued in this paper, that shift in land use that is reflected in the proliferation of sites
of this phase in arid locale of north Gujarat that provided nutritious grasslands rich in minerals and salts, adequate water supply and
ideal landscape led to the development of seasonal short - lived camp settlements. During Late Harappan Phase there seems to be
more emphasis and on pastoral economy than agriculture.

INTRODUCTION

which will serve as a background in understanding

and appreciating archaeological and ethnographic

This paper is concerned with the relationship

data.

1)

between pastoralism and settlements and the

manner in which this relationship is expressed in

PREHISTORIC CLIMATE

the archaeological record. An ethnoarchaeological

investigation of pastoralism carried in Gujarat will
be applied to body of archaeological data of western

The climate and environment of Harappan period

India from a time period of second millennium B.C.,

has long been debated in South Asian archaeology.

identified with Late Harappan Phase of the Harappan

The disagreement comes concerning the amounts of

cultural tradition. This will allow more reliable

rainfall and interpretations have ranged from greater

reconstruction of pastoral economy as well as better

rainfall, to lesser or the same rainfall as today. The

understanding of the settlement pattern associated

earliest scholars including Marshall (1931), Dikshit

with this economy.

(1938), Mackay (1943) and Piggott (1962) argued that

Understanding the ecology is necessary first step

prehistoric climate and environment were distinct

in the analysis of pastoralism. The interaction among

from today. Most of them argued for a wetter climate

the animals, land and people produces a complex web

on the basis of variety of archaeological evidence. This

of ecological relationships. Besides, in any attempt

included the presence of laboriously constructed dams

to understand the settlement pattern, subsistence

in south Baluchistan called Gabar-bands, use of burnt

strategies and cultural change the climate and

bricks at Mohenjo-daro and the depiction of animal

environment are prerequisite in any such study. In the

which are commonly found in damp jungle country.

following pages we will first discuss prehistoric and

present-day geographical and environmental setting

lakes Sambhar, Didwana and Pushkar in Rajasthan

-1-

Later Singhs (1977, 1971) pollen sequence from salt

Kuldeep K. Bhan

also suggested greater rainfall during Harappan

suggestion of marshy habitat is not a precondition

Period, which was later supported by Bryson and

to either rhinos existence, nor should it be treated as

Swain (1981: 135-45) on the basis of climate data from

indicator of any particular climate. Rhinos in Africa

other areas. The palaeoenvironmental reconstructions

live in areas, which are semi arid with xerophytic

indicated there was increased rainfall (as compared

vegetations. Besides, the repeated presence of draught

today) in northwest South Asia during Mature

resistant crops millet in the Harappan phase of

Harappan phase. Singh and others used this sequence

Gujarat suggests that the climate might not have been

as an explanation for the decline of the Indus Valley

very different from than today. This argument finds

tradition (Agrawal and Sood 1982; Allchin and

additional support from the pollen analysis from the

Allchin 1997: 208). Vishnu-Mittre (1974a) expressed

Nal sarovar in north Gujarat. On the basis of this

doubts about the pollen indicators used by Singh

pollen analysis Vishnu-Mittre and Sharma (1979: 99)

for suggesting wet climate. Misra (1984: 461-89)

has indicated Savannah grassland types in the region

also argued the testing of this climatic theory and

of north Gujarat during second millennium B.C.

the Harappan collapse. Shaffer and Lichtenstein on

For Gujarat, the available localized data is generally

the basis of recalibrated radiocarbon determinations

interpreted as indicating no major difference between

refuted Singhs climatic data. They suggested that sub-

third/second millennium B.C and today. (Meadow

phase IVa (between 3000-1800 B.C.) should fall in

1989; Rissman and Chitalwala 1990). Therefore it

sub-phase IVb period of increased aridity (Shaffer

appears that there is some sort of consensus among

and Lichtenstein 1989: 120-21).

scholars that there were no major climatic fluctuations

This hypothesis of the increased aridity finds

in Gujarat during Harappan period and the climate

additional support in the recent study carried by

was not very much different from today. To appreciate

Enzel and his colleagues on the lacustrine record

palaeoenvironment and the influence that climate

of Lunkaransar Lake in the Thar Desert. Their

had on human affairs it is very useful to briefly outline

record suggests that the Lunkaransar Lake rose

below the present environment and some current

abruptly around 5000 B.C. and persisted with minor

adaptations to it.

fluctuations for the following 1000 years and then fell

abruptly to range of 10 to 40 cm of water by 4200

PRESENT DAY GEOGRAPHICAL

AND ENVIRONMENTAL SETTING

B.C. and the final drying occurred between 2894 to

2643 B.C. (Enzel et al. 1999: 127). However, there are
a host of other scholars who argued that climate was
similar to present times (Fairservis 1956; Hegde 1977;

The region of Gujarat is made up of two major

Raikes and Dyson 1961; Meadow 1989; Rissman and

geological zones. The first is the Deccan trap, which

Chitalwala 1990; Posshel and Rawal 1989). Kenoyer

was formed as a result of huge flows of lava during

(1991) notes that even if there were changes in rainfall,

Mesozoic. The trap formation dominated the

the area is so vast that it would not have affected all

lithology of peninsular Gujarat, known as Saurashtra,

regions uniformly.

the main parts of the southern mainland and Kutch.

In Gujarat, large numbers of Harappan and other

The second feature is the recent alluvium brought

contemporary settlements has revealed bones of

from the adjacent highlands by the rivers of Gujarat

rhinoceros, which is now confined to Assam in the

plain. The Mahi River congenitally divides this into

east and some parts of Himalayan Terai in India.

northern and southern segments, although it is an

Chitalwala (1990: 79-82) informs that Zeuners (1963)

arbitrary boundary.
-2-

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

Figure 1 Map showing the various regions of Gujarat, shading indicate concentration of pastoral communities

S o uth Gu j a r at c a n b e d i v i d e d i nt o f o ur

some parts of the bhal tract wheat can be grown,

geographical regions for analytical purposes (Figures

when water withdraws around October. The residual

1 and 2). First and second are the northern and

humidity is sufficient to permit the maturation of

southern halves of the Gujarat a well-defined

wheat without any further moisture input. In the

corridor of alluvium bordered by salt waste and sea

central zone soils are high quality but are saline to

on the west and hills that extend up to the Ghats. The

the west and thin on the western slopes. Sorghum is

third region is the Kutch, a rocky and arid collection

the most popular grain during rabi season as well. It

of outcrops which surrounded on three sides by

is also the most common fodder crop for the cattle

seasonally flooded salt flats called Rann. The last,

(Patel 1977: 44).

to the south of Kutch is the peninsula Saurashtra or

Kathiawad, which is surrounded by the sea on all but

150 cm (Figure 2) and increases sharply on the inner

its northern border.

eastern hills, which have tropical dry deciduous forest

South of Gujarat extends towards Bombay

vegetation. Anogeissus latifolia, Terminalia tomeentora,

(Mumbai) and is bounded in the east by the granite

Terminalia tomeentora and Tactone granlis (teak) are

hills of Sahyadri and Satpura ranges reaching up to

the dominant trees (Gaussen et al. 1968: 38). Grasses

1000 m ASML (Patel 1977: 17). From these hills

may occur but the canopy is more open. Andropogan

the Mahi, Dhadrar, Narmada, Kim and Tapti flow

pumilus, Ischoemum angustifolium (Champion 1935:

to the Gulf of Khambhat. On the west, the Gulf of

85), Iseilma laxum, Apluda artistata, Heteropgon

Khambhat and Arabian Sea bound it. The boundaries

contortus and Cymbopogon martini (Whyte 1964: 99)

of Gulf of Khambhat are locally known as bhal and

are characteristic grasses of this area.

bhalbaru area. Here soils are sandy and salty and

the ground water brackish. The modern cultivators

It is again bounded in the east by Satpura hills and

avoid this area due to heavy flooding. However, in

on the northwest by the Aravallis, which rise up to

-3-

The rainfall in South Gujarat ranges from 60 to

North Gujarat is a sandier and more arid plain.

Kuldeep K. Bhan

900 m. To the northwest the discontinuous hills

is locally known as Nani-Banni and has some of the

reach up to 100 m AMSL (Patel 1977: 17-18). North

best grassland types in India, having very high protein

Gujarat is bordered on the west by a salt desert and

content (refer Mann 1916 and Burns et al. 1916 for the

low-lying neck of the Saurashtra peninsula. The rivers

analyses of grasses). This area is the breeding center

flowing through this region includes the Rupen,

of the worlds best tropical breed known as Kankrej

Banas, Sabrmati and Mahi. They are fed by monsoon

or Wadhial (George 1985) and Murrah or Mehsana

rains and are ill suited for irrigation (Leshnik 1968:

buffaloes (Whyte 1964: 98). Besides, it also supports

298). The northern plain is drier, receiving 80 cm

large population of sheep and goat and Asiatic wild

rain annually in the east grading down to 40 cm in

ass.

the west. Soils are alluvium and have been derived

from wind and water erosion of the Aravalli ranges.

trap is patchy, while older facies date to Jurassic tomes

The percentage of silt and sand are higher, reaching

and comprise uplifted marine sediments, sandstone,

up to 90% in the west. Much of the area is saline and

limestone and shales (Wadia 1957: 264; Spate and

alkaline treeless steppe that merges with the desert

Learmonth 1967: 98). The soils of Kutch are mixed

thus is not encouraging for agricultural activities.

as a result of its parent rock. The outcrops of Kutch

Besides, the landscape of north Gujarat is relieved

face the sea on the southwest, while on all other

by frequent occurrences of relict sand dunes, giving

sides desolate salt flats, the Ranns, bond them. The

rise to numerous small lakes, near the inter dune

Ranns of Kutch are alluvial deposit of clay and sand

depressions. These lakes retain monsoon run off for

brought down by minor rivers of Rajasthan as well as

two to six months after the monsoon. The soils of

the eastern Nara of Sindh. The Ranns of Kutch were

north Gujarat are rich in chlorides, sulphates and

originally shallow bays connecting with sea (Bombay

carbonates of sodium. This is especially true with the

Presidency 1884: 3; Patel 1977: 20; Gupta 1977: 205).

areas, which are frequently flooded and have poor

During monsoon season the Ranns get flooded with

drainage systems.

salt water to a depth of 1 m, but when water recedes

The extensive alkaline areas in this zone are known

with the seasonal change a thick deposit of salt

as usar lands and have characteristic grass vegetation.

remains. This Rann supports a large population of

This area is dominated by Acacia, Capparis, Euphorbia,

flamingoes and wild ass.

Zizphus (Gaussen et al. 1968: 33) and Prosopis spieigera

(Whyte 1964: 98). The grasslands occur as both

ranges from 30 to 40 cm per year. The forest regime

undergrowth and open patches. The dominant grasses

is of Acacia, Capparis, Zizphus, Salvadora oleoides

are Erichloa ramosa, Chloris virgata, Eragrostis ciliaris,

and Prosopis specigera. Like the western section of the

Dactyoctenium aegyptium, Sporobolus cormandelianus,

north Gujarat the southern section of Greater Rann

Echinochloa colonum, Eleusine compsessa, Eleusine

of Kutch is vast grassland locally known as Moti Banni

brevifolia, Cynodon dactylon, and Aristida sp. Legumes

and is much favored by the breeders (Bose 1975: 6).

found growing in this area are Heylandia latibrosa and

The dominant grasses are various Andropogan species,

Indiggogera sp. In the same area but adjacent small

Anthistiria imberdis, Aristida adesencionis, Aristida

salt-water streams, river banks and the marshy area,

adesenscionis, Aristida hystricula, Sporobolus indicus,

dominant vegetation are Dichanthium annulatum,

Cynodon dactylon, Chloris barbata, Eleusine flagellifera,

Sporobolus marginatus and Eriochloa ramosa.

Eleusine aristata, Phrcigmites karke, Eragrostis ciliaris,

Sporobolus marginatus and Cressa cretica, etc. are

Eragrostis cynosuroides, Eragrostis amabilis, Aeuropus

observed growing even on the pure salt crust. The tract

vilosus, Halopyrum musronatum, Dendroclamas

-4-

In the third region of Kutch, the flow of Deccan

The climate of Kutch is most arid and rainfall

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

Figure 2 Map showing the elevation and rainfall of Gujarat region.

strictus and Adiantum lunulatum.

vegetation of Dichanthium and Cenchrus grasslands

The fourth region is the peninsula of Saurashtra.

growing in weathered patches. The rocky hills of the

The central part of Saurashtra is made of an

interior, the Gir and Barda hills and the northern

irregular plateau, bisected by peaks near Chotila in

coast of Saurashtra are studded with areas of pastures

Surendernagar District, Girnar in Junagadh district

and are scenes of much pastoral activity (Government

and Alech and Barda in Jamnagar district. Soils

of Gujarat 1969: 20). The pastoral animal stock of this

of Saurashtra are clayey black cotton soil derived

area are a breed of medium sized, heavy cattle, useful

form from weathering of bedrock. The rainfall

for agriculture work and the cows give good milk.

over Saurashtra ranges from 40 - 80 cm. Thus the

This breed of cattle is locally known as Gir, Surati or

vegetation of the peninsula is the same that of Acacia-

Sorahti. Besides, they also breed sheep, goat, buffalo

Capparis series as found in north Gujarat plain. Along

and camel.

the northwestern coast fronting the Gulf of Kutch is

stretch of Mangrove swamp (Government of Gujarat

mainland of Gujarat at several times in the past. The

1970: 32). Meanwhile the Barda hills to the west, Mt.

land bridge connecting it to Gujarat plain appears

Girnar and Gir hills to the south, and the Plitana hills

to have been flooded up to six months of the year

to southwest support some anogeissus terminalia

until 1813, while the travelers report of 16th century

tectona and acacia angogeissus woodland (Gauseen

state that a branch of Indus ran by the Khambhat

et al. 1968: 20). Many bedrock hillocks are also found

town (Government of Gujarat 1969: 170). Silting

there, some of them cover only a hectare, while

by the rivers of the mainland and the peninsula has

some of them are considerably larger. The tips of

connected the two wide swat of alluvial deposit

these hillocks are always barren with tufts of grassy

known as the Nal depression (Rissman 1985: 57). The

-5-

Saurashtra may have been cut off from the

Kuldeep K. Bhan

rivers originate in the central highland region and

frequency of crop failure in the state. Among the local

flow in all directions in radial pattern. The large rivers

administrative units called talukas, of Gujarat, 22%

Bhadar, flowing westwards and the Shetrunji flowing

experience total crop failure at least once every six

eastwards are seasonally dry and turn into sporadic

years on average (Rissman 1985: 351). Another aspect

pools during hot summer months. The soils of the

of the precipitation regime is the localized nature of

area clayey black cotton soil, derived from the residue

cloud burst. While one area may experience plentiful

of weathered basalt. Sandy alluvium and marine mud

rainfall an adjoining area may be passing through

characterize the southwestern coastal strips.

draught.

A distinctive feature of Gujarat today and perhaps

Another important part of the subsistence

in Harappan times as well, is the inherent seasonality

is animal husbandr y, by pastoral transhumant

in the ecological aspects of the region. It lies in the

communities like Rabaris and Bharvads, who do

zone of the monsoon marked by an annual variation

not traditionally own any agricultural land and

in rainfall of greater than 30% of the mean (Rissman

derive income and subsistence from breeding or /

1985: 351). A great deal of fluctuation from year

and herding cattle, sheep, goat, camel and buffaloes.

to year is evident; often an extremely dry one will

The profound nature of the seasonality with dry

follow a very rainy year. These shifts do not occur in

and wet periods in Gujarat results in movement of

a predicable sequence and thus make rainfall a risky

large population of these pastorals and their animals.

resource.

Though these groups are distributed throughout

Generally speaking, the agriculture of Gujarat

Gujarat, large numbers are present in the bordering

is characterized by dry farming. Eighty five percent

villages of Nani and Moti Banni, Barda, Gir hill and in

of agriculture is rain fed. The greater majority of

northern and southern fringes of the coastal Gujarat

crop production occurs in monsoon season and is

(Figure 1). It is in this environmental backdrop

harvested in autumn (kharif). The black cotton soil

that we will be examining the archaeological and

is labor reducing and during monsoon swells with

ethnoarchaeological data.

water and later develops cracks as the dry farming

season wears on (Rissman 1985: 164). In much of

ARCHEOLOGICAL BACKGROUND

Gujarat, hoeing with blade harrow is practiced as

a substitute to ploughing. Annual ploughing to
destroy weeds is not necessary because only one crop

During Late Harappan phase, the previously

is grown a year (Patel 1977: 39). Millets dominate

integrated region of the Indus Civilization broke

subsistence crops of Gujarat bajra (Pennisetum

into three localized cultures defined by Shaffer

typhoides) and jowar (Sorghum bicolor). The village-

(1991) and Kenoyer (1991) as Localization Era. These

based agriculturists maintain up to five animals as

cultures named as Jhukar Phase, Cemetery H (Punjab

an incidental adjustment or an internal part of the

Phase) and Rangpur Phase, and were named after

farming enterprise, with the production of food and

the important sites where specific pottery styles were

cash crops as their preliminary activity.

first discovered or the geographical regions in such

settlements were found (Kenoyer 2005 and Bhan

Rainfall is the major source of moisture in Gujarat

and subsistence depends upon it. The productivity

2001).

of the subsistence depends upon the year of good

monsoon. The dependence of subsistence production

Gujarat. Here also, a high degree of continuity from

upon rainfall is reflected in statistics on the relative

Mature to Late Harappan Phase has been indicated,

-6-

Rangpur phase refers to the entire region of

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

Figure 3 Map of Gujarat showing major archaeological sites mentioned in the text.

though it is most evident in ceramics. Both regional

the introduction of Lustrous Red Ware. This Phase

pottery that are specific to Gujarat and classical Indus

was defined as the Initial Phase of Late Harappan.

forms undergo a period of change as the Mature

The increase in frequency of Lustrous Red Ware,

Harappan ends. Alternation of certain ceramic forms

painted Black and Red Ware and Coarse Red Wares

and the introduction of the Lustrous Red Ware

signaled the beginning of Rangpur III, the final

a novel method of surface treatment of burnishing

period at Rangpur and represents the final phase of

the vessels to high luster (Rao 1963: 23) are thought

Late Harappan in Gujarat (Bhan 1989). However,

to represent the Late Harappan Phase in Gujarat.

recent radiocarbon determinations from Rojdi have

Nevertheless, a clarification about Rangpur sequence

completely altered the earlier periodization and

that was used for more than three decades to define

interpretation of Late Harappans in Gujarat.

the periodization of Harappan Gujarat is needed,

since the interplay between settlement patterns is a

Possehl and Rawal have suggested that all the

crucial variable in the interpretation of past economy

settlements with pottery comparable to Rojdi should

and culture change.

be dated to urban Mature phase and not to the

In terms of Rangpur sequence, Period IIA was

Late Harappan Phase (1989). Now the emerging

thought to be contemporary with Mature Harappan,

immerging picture from their research is that there

Period IIB signaled the onset of Harappan period.

are two categories of settlements during Mature

While certain ceramic forms like Indus goblet and

Harappan phase in Gujarat. The first category of

beakers disappear, other forms seem to have simply

settlements, which Possehl and Rawal designated

altered shape and decoration (Rao 1963: 22, 59-65).

as Sindhi Harappan, share material inventory

The alteration of form continued in IIC along with

of Mature Harappan sites of Indus (1989) and

-7-

On the basis of radiocarbon determinations,

Kuldeep K. Bhan

Figure 4 Jaidak - 1 (Pithad): Plan of entrance gate in the southern fortification wall.

the regional style ceramics (Bhan 1994). Large

island and the Ranns of Kutch banks (with the

concentration of such settlements has been reported

Lothal in Bhal region) belonged to the urbanized

from Kutch and is represented by excavated sites

Harappan core area of the Indus/Ghaggar-Hakra-

of Desalpur, Surkotada and perhaps Dholavira in

Nara Valley. While the Sorath Harappan, Prabhas

Kutch, Gola Dhoro (Bagasra), Lothal and Nageshwar

Patan and Padri tradition in Saurashtra and the

in Saurashtra, and Nagwada and Jekhada in north

Anarta tradition in north Gujarat plains formed

Gujarat represent this group (Figure 3). Besides

the agro-pastoralists periphery, while the Ranns of

sharing the classical Harappan material inventory

Kutch banks and the coastal borders of Saurashtra

most of the settlements are found associated with

as narrow zones of interaction (1997:84). However,

some craft activity like shell bangle and ladle, stone

on the basis of ceramics and stratigraphic analysis of

bead, or faience production. The other category

a Mature Harappan site at Gola Dhoro (Bagasara)

designated as Sorath Harappan are represented by

in Saurashtra, Ajithprasad and his colleagues have

small rural agriculture or pastoral camp settlements

indicated that though Rangpur sequence is not

and has more or less completely lack of evidence for

completely vertical, but the pottery sequence and the

substantial complex craft activity. These settlements

stratigraphy at Gola Dhoro is fully in agreement with

were earlier thought to represent IIB-C of Rangpur

the Rangpur ceramic sequence. Like Rangpur, at Gola

sequence or Initial phase of Late Harappan or Post

Dhoro also the Mature Harappan material inventory

Urban Phase in Gujarat. Herman (1997a : 187-

is followed by Rangpur II A, II B-C pottery forms

198) questioned the Rangpur sequence on the basis

(Ajithprasad et al. 1999). It is interesting to note here,

of stratigraphy and radiocarbon determinations

that the excavations at Mature Harappan (Sindhi)

from Rojdi, relying heavily on the remarks made by

sites in Gujarat have always revealed IIB-C ceramic

Misra (1965) on the Rangpur sequence. Herman

types above the Mature Harappan deposits. However,

puts forward core/periphery model to explain the

it is not clear as yet if it is because that the Harappan

phenomenon.

people of IIB-C phase (Sorath) occupied these

Herman suggests that during the Harappan period

settlements after the Mature Harappan or it really has

(second half of the third millennium B.C), the Kutch

some chronological significance. The argument seems

-8-

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

Figure 5 Circular hut structures at Jekhada (after Momin 1983)

to have come about a full circle and what we need at

been recovered so far. Rice grains (Oryza sativa) have

present are set of consistent radiocarbon dates form

been identified form the impressions in potsherds at

Gola Dhoro (Bagasra) itself and many more sites so as

Lothal and Rangpur IIA (Ghosh and Lal 1963: 161-

to fix its proper chronological position in Harappan

175; Rao and Lal 1985: 667-684). Rice grows wild

Tradition of Gujarat.

today in the marshes of Rajasthan, Madhya Pradesh

Nearly 94% of these settlements of this phase are

(Vishnu-Mittre 1961: 18) and Gujarat (Vishnu-Mittre

located in Saurashtra, while 6% of the settlements

and Savithri 1982: 207), it cannot be concluded

are located on the eastern border of Little Rann

that it was cultivated for subsistence. Besides, the

of Kutch in north Gujarat. The average size of the

study of charcoal from Lothal (Rao and Lal 1985:

settlements in Saurashtra is 5.3 hectares (Possehl

667-684) and Rangpur (Ghosh and Lal 1963: 161-

1980:65). However, settlements at Lolai, Jaidak, and

175) have indicated the presence of trees like Acacia,

Rojdi approximately range between 4 to 7 hectares

Albizbia, Andina cordifolia and Tectone gradis (teak).

having structures with stone foundations and stone

The presence of teak at Lothal is intriguing. Today

walls circling them (Figure 4). Another category of

it grows in eastern Gujarat in the areas with rainfall

settlements of this phase has the remains of round hut

ranging from 75 to 175 cm and falls under the tropical

2)

floors as revealed at Jekhada (Figure 5)(Momin 1983)

deciduous type (Rao and Lal 1985: 680). However,

and Kanewal (Figure 6)(Mehta et al. 1980), in north

it is of interest to note here, that the strategy and

Gujarat, Vagad (Sonawane and Mehta 1985: 38-44)

methods used at both the sites were not intensive

and Nesadi in Saurashtra (Mehta 1984: 227-230). The

enough to take them as representative of Harappan

recently located settlements of this phase in north

sites of Gujarat.

Gujarat are extremely small in size and have very thin

and patchy habitational deposit. They fall into similar

faunal data and the apparent millet cultivation

settlement patterns as well as functional categories of

prompted Meadow (1989) to propose a second

pastoral camp settlements as well as other prehistoric

revolution during the second millennium B.C.

settlements of north Gujarat.

However, Weber has refuted this on the basis of

In the greater Indus subsistence pattern based

evidence of millet and early radiocarbon dates

on domestic cattle and the winter crops of wheat

from Rojdi in Saurashtra, which as mentioned

and barley is established by sixth millennium B.C.

above are proposed to be contemporary to the

(Meadow 1989), though subsequent diversity in

Mature Harappan phase in Gujarat. The subsistence

the Harappan subsistence is a much debated topic.

information of this phase (Sorath Harappan) comes

Little information is available on the subsistence

mostly from the American and the Gujarat State

pattern of Mature Harappan phase settlements in

Archaeology Department excavation program carried

Gujarat, interestingly neither wheat nor barley have

out at Rojdi. The subsistence information available

-9-

The first appearance of horse and donkey in the

Kuldeep K. Bhan

from Rojdi A and B include millets Eleusine

settlements such as Lothal B, Rojdi C, Prabhas Patan

coracana (ragi), Panicum milliare (little millet) and

II, Jaidak, Rangpur IIC and perhaps Babarkot in

Setaria tomesntosa, S. glauca (yellow fox millet), S.

Saurashtra appear to have witnessed a continuation

italica (fox millet). Weber argues that S. italica did not

of a short period of urban organization of the Mature

become a prominent cultigen until 2000 B.C. (1989).

Harappan Phase into the Late Harappan Phase 3).

This has been disputed by Reddy (1994: 28), she is of

At Lothal and Surkotada also there is evidence

the opinion that Webers argument is based on a single

of over-crowding and encroachment rather than

site and this could be result of changing demands by

abandonment and decline. Most of the Sindhi

the community and the subsistence system. There are

settlements except Rangpur were abandoned after

various theories regarding the center of origin of the

this stage. As a matter of fact some settlements Rojdi

different millets and these are not considered native

C show expansion, though Lustrous Red Ware was

of Gujarat and seem to have been brought to here

not reported from any of the above mentioned sites

from elsewhere, however, timing and dynamics of this

except at Rangpur IIC (Rao 1963). Herman (1997a:

process is much debated (Weber 1989). Although, the

196; 1997b: 100) is of the opinion that the presence

lowest level at Rojdi has revealed a few barley grains,

of Lustrous Red Ware in this phase at Rangpur is

Weber (1989) argues that barley could have been

debatable, because massive pits disturbed the layers.

traded instead of being grown locally.

Now the question arises whether Lustrous Red Ware

In Gujarat, both the categ ories of Mature

does really belong to IIC or III of Rangpur sequence.

Ha r a p p a n P h a s e s e t t l e m e n t s i n d i c a t e t h e

This perhaps represents a transitional traditional

predominance of cattle husbandry over the raising

phase from Mature Harappan to Late Harappan.

of sheep, goat, buffalo and pig. However, excavations

Traditionally, the appearance of Lustrous Red Ware

at Nag wada in north Gujarat (Patel 1985) and

and changes in certain ceramic forms have has been

Nageshwar in Saurashtra (Shah and Bhan 1990)

used to distinguish between Mature Harappan and

have yielded large amounts bones of wild animal

the Late Harappan Phases.

like sambar, blackbuck, chital, gazelle and nilgai and

other rare forms like hare, camel and wild ass (Patel

slightly later than the above mentioned sites, are

1985: 47). Most of these animals inhabit grasslands

represented by round hut floors with post holes

interspersed with scrub jungle.

along the periphery such as at Ratanpura (Figure 7)

Another categ or y of settlements, perhaps

(Bhan 1989), Kanewal (Figure 6)(Mehta et al. 1980)

in north Gujarat, Vagad (Sonawane and Mehta

LATE HARAPPAN PHASE

1985) and Nesadi (Mehta 1984: 27-30) in Saurashtra.

Except Kanewal none of the excavation reports are

As already mentioned above certain changes were

fully published as yet. However, reexamination of the

taking place around the turn of second millennium.

pottery from Nesadi and Jekhada has indicated that

Researchers have seen these changes as a result of

some of the hut floors might belong to this phase.

failing or at least changes in international exchange

A preliminary analysis of the published pottery

networks ( Jarrige 1995: 5-30) and / or a change in river

from Kanewal trench V laid at Kesarisinghs Khetra

system (Misra 1984). In Gujarat also such changes

represents two phases of circular hut floors positively

have been observed and three-tired settlement pattern

belonging to this phase. A successive occupation

seems to have been established.

from Sorath Harappan Phase to Late Harappan

Phase is represented at this site. At Oriyo Timbo in

At Desalpur IB and Surkotada IC in Kutch and

- 10 -

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

Figure 6 Circular huts at Kanewal (after Mehta et.al. 1980)

Figure 7 Ratanpura: irregular patchy rammed floor (courtesy Dept. of Archaeology, M. S. University of Baroda)

- 11 -

Kuldeep K. Bhan

Figure 8 Map showing distribution of prehistoric settlements in the western section of north Gujarat

- 12 -

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

Table 1 Radiocarbon Dates for Late Harappan Phase in Gujarat.

Site

Lab #

Date (5730 life, B. C.)

Mid range for Corrected

Dates (after Ralph 1973)

Surkotada, IC

TF-1294

1780 100

2170, 2110, 2050

TF- 1297

1795 100

RRL- 91

2020 170

PRL - 19

1245 165

PRL - 20

1485 110

Rojdi C

BETA-61767
BETA- 61768
PRL-1084
PRL-1081

1730 60
1675 60
1860150
108115

Vagad
Ratanpura
Malvan

BM 2612
BM 2615
TF-1081

181050
180050
80095

Oriyo Timbo

PRL-1427
PRL-1424
TF-31

1385115
1910135
1275110

TF-32

1550100

P-204
P-472
P-473
P-474
TF-42

1600130
130070
1330 70
1240190
1175120

TF-43

1040105

Prabhas Patan,
Period III

Ahar

Navdatoli
Chandoli

Reference

Agrawal and Kusumgar

1973
2170, 2110-2130, 2050
Agrawal and Kusumgar
1973
2620, 2340-2460, 2190-2290 Kusumgar, Lal and Sarna
1963
16001640, 1460, 1220-1260 Kusumgar, Lal and Sarna
1963
1780 -1910, 1690, 1660, 1650 Kusumgar, Lal and Sarna
1963
21102130, 2080, 2040
Herman 1997b
2080,2040,1920-1950
Herman 1997b
2340-2410, 2160, 2070
Herman 1997b
800,400-570,210-360
Herman 1997b
Possehl 1992
2160, 2140
M. S. Uni.
2160, 2120-2140, 2110
M. S. Uni.
940,900,800-900, 800
Herman 1996
Possehl 1992
1690-1730,1570-1600
2420-2480, 2180, 2110
1570-1600,1490, 1310-1380

Possehl 1992
Possehl 1992
Kusumgar, Lal and Sarna
1963
2020-2040, 1770-1870,1660 Kusumagar, Lal and Sarna
1963
2080,1800,1690
Stuckenrath 1963
1570-1600,1510-1400-1450
Stuckenrath 1963
1600-1640, 1510, 1400-1450 Stuckenrath 1963
1650,1460,1170 -1210
Stuckenrath 1963
1600-1630,1320-1370,1220- Kusumgar, Lal and Sarna
1240
1963
1300,1170,1020
Kusumgar, Lal and Sarna
1963

Saurashtra, Ratanpura (Figure 7) and Datrana in

into consideration. The radiocarbon dates from Oriyo

Gujarat settlements of this phase are without any

Timbo (PRL-1427) and Prabhas Patan (PRL-91)

features. At times they only have irregular patches of

calibrated to the beginning of third millennium B.C

rammed floor.

are too early. Although greater precision is desired,

we will have to wait further excavations and more

In Gujarat radiocarbon dates are inconsistent.

Ho w e v e r, o n t h e b a s i s o f t h e r a d i o c a r b o n

carbon-14 determinations.

determinations ( Table 1) available from form

Surkotada IC, Vagad IC, Rojdi C, Prabhas Patan III

settlement count and average site size. Majority of the

and Ratanpura in Gujarat and Ahar in Rajasthan,

settlements are situated near river banks on alluvial,

Chandali in Maharashtra and Navdatoli in Madhya

residual black cotton or black cotton soil famous for

Pradesh outside Gujarat provides a time bracket of

its moisture retaining capacity, the prime agricultural

C. 2000-1500 B.C., which can be proposed for this

soil of Saurashtra. Until now, approximately, 136

phase in Gujarat, if the late date of Malvan Phase I

settlements of this phase have been reported from

(TF-1084) and Rojdi C (PRL-1081) are not taken

Gujarat, out of which nearly 31% (51) are located in

- 13 -

Saurashtra witnessed dramatic decrease in the

Kuldeep K. Bhan

Table 2 Archeobotanical remains from Late Harappan Sites of Gujarat

Surkotada, Period IC

Millet:

Eleusine coracanna, Setaria italica, Setaria virids L. (70% of the total assemblage)

Others:

Grasses:
Andropogon sp., Arundinella metzii, Brachiaria repatans, Dichanthium sp., Echinochloa stangina
Sedges:
Cyperus micheleanus, Eriophorum sp., Fimbristylis ovata, Amaranthus sp., (Vishnu-Mittre and Savithri
1982:214)

Rojdi C

Millet:

Eleusine Coracona, Seteria italica, Panicum miliare, Sorghum bicolor (high percentage of Steria sp.)

Pulses:

Phaseolus, Pisum, Lens

Others:

Abelmoschus, Amarthanthus, Chenopodium album, Dactylotenium, Digitatia, Euphorbia postrata (Weber 1991;
Weber 1989: 179)

Lothal

Millet:

Present

Others:

Acacia sps. (Rao and Lal 1985: 679)

Rangpur

Millet:

Pennistrtum typhoides

Others:

Acacia sp., Albizzia sp., (Ghosh and Lal 1963: 163, 169, 171)

Oriyo Timbo and Babar kot

Millets:

Eleusine Coracana, Setaria sp. and Panicum sp.

Others:

Aizoon sp., Carex sp., Chenopodium/Amranthus sp., Cyperus sp., Digitatia sp., Zizyphus sp. (Reddy 1994, 1991:
80-81; Rissman 1985)

Saurashtra, conversely a large number of settlements

other Chalcolithic settlements reported from form

63% (87) have been located in north Gujarat (Figure

this region show close similarities in their location

8).

as well as in the nature of the settlement (Bhan

In north Gujarat, settlements of this phase are

1994: 2001). Most of the settlements are less than

situated bordering the Little Rann of Kutch and

0.5 hectares. Some of the settlements like Chagda,

in the wastelands close to Rupen River and its

Gudel, Khaksar and Popatpura in Bhalbaru region

tributaries. Most of the sites are associated with relict

of north Gujarat and Dhama, Mujpur, Nayka, Vanod

sand dunes associated with inter dune depression,

and Valewada in north Gujarat are so small that

which retain water for three to four months after

only a few potsherds and bones are recovered from

monsoon. Settlements are small, having thin and

these sites. Locating such sites is extremely difficult

patchy habitational deposits, with high density of

in any type of exploration and we suspect that some

animal bones and at times out numbering the other

number of settlements would have been missed in our

material relics of the site. As a matter of fact, all

explorations. This perhaps explains why Misra and

- 14 -

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

Leshnik (Leshnik 1968) could not locate Sujinipur

intensity thus larger settlements with greater potential

only a decade after Raos survey (Rao 1963: 177).

and complex infrastructure (Desai 1985: 103). This

Information on agriculture of this phase comes

intensive explorations carried out in eastern section

mostly from Saurashtra and to some extent from

of north Gujarat by Foote (1912), Sankalia and Karvel

Kutch, is indicative of summer crop cultivation,

(1949) and Misra and Leshnik (Leshnik 1968: 259-

mainly millets (Table 2). The archaeobotanical

309) has revealed large number of microlithic using

remains from Rangpur (Ghosh and Lal 1963), and

comminutes communities and absence of Bronze

Surkotada (Vishnu-Mittre and Savitri 1982) are only

Age settlements, except one isolated settlement

from one or two samples and therefore results are

at Sujinipur. Although some interaction with the

limited. However, rigorous methods of recovery

chalcolithic communities of Gujarat is reflected in

were employed at Rojdi (Weber 1989), Oriyo Timbo

the presence of some fragmentary pottery and copper

(Wagner 1982; Reddy 1994) and Babarkot (Reddy

tools recovered at the site of Langhnaj (Sankalia

1994). The millet crops recovered from this phase

1974). The absence of Bronze Age settlements in the

include Panicum miliare, Eleucine coracana, Seteria

western section of the north Gujarat and nature of

italica, Sorgum and Echinochloa sps. Besides, the

the settlements indicate that the settlements were not

examination of charcoal samples from Rangpur have

located for harnessing the agricultural potential of

indicated tree types like Acacia, Albizzia sps., Soymida

the region as thought earlier (Hegde and Sonawane

febrifuga, Pterocarpus santalinus (Ghosh and Lal 1963:

1986), but for some other reasons. The presence of

161-175).

excellent nutritious grassland in the western section,

The analysis of archaeozoological remains from

that are supposed to be best grasslands types of India,

the majority of the excavated sites of this phase show

and availability of water in the inter dune depressions,

that the predominance of cattle over sheep, goat and

that retain water on average for five to six months

buffalo continues. The faunal analysis carried out

after the monsoon season, perhaps were the primary

at Ratanpura revealed 67.68% of cattle, 13% sheep/

determinants for the development of settlements

goat and 3.06% buffalo, while the wild animals are as

in this region. Thus on the basis of location, and

high as 16.04% that and include chital 7.58%, sambar

their ephemeral nature, we can suggest that the

2.97%, nilgai 1.44%, pig , blackbuck and gazelle

settlements were temporary camps of the people

together is represented by 1.35%. Besides, other rare

involved in pastoral activities. The overall increase in

forms include hare, dog, fowl and wild ass (Bhan and

the number of the settlements during this phase in

Shah 1990). The faunal analysis from north Gujarat is

north Gujarat perhaps suggests that the pastoralism

indicative of higher exploitation of wild animal forms

increased significantly during this phase, although the

as compared to the settlements from Saurashtra.

area seems to have also attracted the Early, Mature

The western section of north Gujarat, in which

and Sorath Harappan for similar reasons (Bhan

settlements are clustered together, has low rainfall

1994: 2001). A similar situation has been noticed by

(44-55 cm), alluvial soil with high salinity and very

Mughal in Cholistan desert where a 26% increase in

high percentage of silt and sand reaching up to 90%

pastoral camp settlements have been recorded in Late

and the rate of crop failure is also comparatively very

Harappan Phase (1997: 56)

high. Therefore, this area has a low cropping intensity,

smaller population potential and low-level of

on the Late Harappan settlements at Oriyo Timbo

infrastructure. Conversely, the eastern section of the

and Babarkot in Saurashtra by Reddy (1994) have

region with higher soil fertility had higher cropping

further substantiatesed and strengths strengthened

- 15 -

Recent palaeoethnobotanical studies carried out

Kuldeep K. Bhan

this view. Reddys investigations have suggested that

north Gujarat. This argument, though forwarded as

Babarkot cultivation of millet crops by the occupants

early as 1986 (Bhan 1989), has been unable to gain

was primarily as food grains. The practice of sedentary

confirmation either for against (Possehl 1997: 456;

agriculture was supplemented by sedentary animal

Mughal 1992: 217; refer Sonawane and Ajithprasad

husbandry. The greater emphasis on the cultivation

1994: 136 for a divergent view). This has highlighted

of Eleucine coracona (which has a limited use as a

the problem of accurately differentiating pastoral

human food) compared to other millets Seteria

sites from hunter-gatherer and/or agriculturists.

italica and Panicum miliare at Babarkot was perhaps

Therefore an ethnoarchaeological study of pastoralism

because it was cultivated as green fodder for animals,

in Gujarat was undertaken and the preliminary

She further suggests that the cultivation of millets was

results of this study are briefly presented below.

for use both as animal fodder as well as for human

Ethnographic observations of the contemporary

consumption at Babarkot. It is suggestive of an

pastoral comminutes of Gujarat provides insight into

economy more akin to sedentary agriculturists, with a

pastoral economy and helps to further strengthen the

minor pastoral component, that of sedentary animals

proposed argument.

(1994: 381-382).

Reddys palaeobotancial analysis at Oriyo

ETHNOARCHAEOLOGICAL
STUDIES

Timbo has indicated that millet was not cultivated

at the site but instead was brought into the site as
highly processed grain. She further suggests that
these grains were either brought to this settlement

Livestock raising in Gujarat is almost exclusively

through trade or exchange with farming groups in

carried out by professional breeders. A distinctive

the area, or cultivated elsewhere (outside the range

feature of the pastoral communities like Rabari,

of settlement) by the occupants and brought along

Bharvad, Jat and to some extent Charan is that they

with them on their seasonal migration to the site

make living primarily by breeding and/or herding

(1994: 382). Reedys observation finds additional

animals. They are collectively called Maldharis. At

support from Rissmans (1985) seasonality data from

present there is no restriction of principle which

the annuli growth ring of cattle teeth that places the

governs the type of livestock they raise. Their livestock

occupation of Oriyo Timbo from March through

are the type best suited to local ecological situations.

July. Reddy (1994: 383) also stresses that the economy

However, the meaning of Maldhari perhaps denotes

of Oriyo Timbo was not exclusively based on animal

those who emphasizes cattle-breeding rather than

husbandry, since wild plants and animal exploitation

cattle-tending (Hellbusch 1975:123). The environment

was significant component of their subsistence

imposes conditions upon animal breeders that

strategies.

dictate the type of animals which can be produced

The natural incentives of the environs of north

successfully, tradition not with standing. An area,

Gujarat seems to have attracted various pastoral/

once suitable only for cattle may be transformed by

hunter-gatherers and hunter-gatherers to north

extended over-uses to one suitable for smaller animals

Gujarat as early as third millennium B.C (Bhan 2001).

like sheep and goat. Environmental factors influence

The pastoral activities also seem to have increased

the choice of animal but cultural traditions connect

substantially during Late Harappan phase (Bhan

these communities with specific animals. In general

1986: 1994), which is reflected in an increase in the

Rabaris are camel-breeders, the Charan and Jats

number of settlements of this phase in arid locale of

breed cattle along with buffalo, and the Bharvads are

- 16 -

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

sheep and goat breeders. It seems the tradition was

more strictly observed in the past than at the present

considerable skill and knowledge. They are most

(Hellbusch 1975: 124).

careful about mating practices, early castration and

These breeders pursue their business with

Jats are the only Muslim breeders of Gujarat and

herd their animals separately, take them to best grazing

inhabit the Moti-Banni area of Kutch. The Hindu

grounds at the best season and produce excellent

breeder groups are distributed all over Gujarat, but

animals with the expenditure that could hardily

large number inhabit the bordering villages of Banni

lowered (Keatinge 1917: 17 and 31). The breeding

tract in Kutch, eastern border of Little Rann of

communities of north Gujarat are responsible for the

Kutch in north Gujarat (Nani Banni), Barda and Gir

development of world famous tropical breed of cattle

hills, and the northern fringes of Coastal Gujarat.

like Kankrej or Wadhial which are similar to the Zebu

Certain cultural traits of Jats in Kutch are shared

cattles characteristic hump and dewlap, which is very

with other animal breeding communities of Gujarat.

proudly displayed as a symbol on Harappan seals.

Among these, of ethnographic interest are caste

In the very recent past they were exported to north

panchayat, caste assemblies, and veneration of goddess

and South America for grading up indigenous cattle

(Hellbusch 1975: 122). Unfortunately, statistical data

(Imperial Gazetteer 1909: 11). Besides, breeders of this

regarding these communities does not exist.

region are also responsible for the development of

Figure 9 Plan of the monsoon camp at Masali, north Gujarat

- 17 -

Kuldeep K. Bhan

Figure 10 Monson camp site at Masali, north Gujarat (observe marshy wasteland in the background).

Murrah and Mehsana buffaloes and are equally apt at

along with other produces (Huntingford 1980: 81).

raising a local breed of sheep, which yields fine quality

Abul Fzual describes Kutch in the Ain-i-Akbari (1583-

wool, most of which is exported. It must have taken

1590) as a territory with an excellent breed of horses,

thousands of years for these breeders to evolve the

good camels and goats. The reference to Rabaris

best breed of animal suited to tropics, which would

appears in Rasmala and other chronicles of Gujarat,

stand fly and mosquito nuisance and tropical diseases

which mentions of Lakho Fulani, who was a son of a

and could live grazing monsoonic grass and roughage.

Rabari girl and ruled Kutch during 10th century AD.

Such breeds can only be produced by a long

(Williams 1958: 76) Later during the early twentieth

and uninterrupted period of breeding on sound

century we find frequent flattering mention of these

principles, which are thoroughly understood by

breeders in British policy discussions.

these professional breeders. Animals are not allowed

to cover their own progeny. They select their stud

out in Gujarat over a period of six years (1989-1995).

animal with great care. The best bull-calves of a

A number of variables regarded as possible factors in

particular year are chosen. The second or third calves

the choice of site of location were recording during

of exceptionally well shaped, big, young bull of good

survey. These include site maps, location, distance

pedigree are chosen.

and gradient and known sources of water. A variety

There is scanty, though useful, mention of

of cultural remains left on these camps were also

these breeders in ancient literature. The Periplus of

recorded and mapped. Besides, a special attempt was

Erythraean Sea mentions the great herds of cattle

made to record the type of vessels these pastoralists

in Saurashtra and export of clarified butter (ghee)

carry with them.

- 18 -

Ethnoarchaeological studies have been carried

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

Figure 11 Mosoon camp at Daro-no-Nes (complex 2 and 3), Barda hills

The study has revealed that these breeders

source is at least 100 m or more from the camping

aggregate during monsoon season all along the river

area perhaps to avoid being the victim of mosquitoes.

banks of north Gujarat specially Rupen, bordering

areas of the Ranns, and also around the salt marshy

archaeologists who think of pastoral camps as

wastelands known as padthar or kahrphat. In

ephemeral sites without substantial traces that

Saurashtra they occupy the foothills of forested areas.

would enter the archaeological record, the monsoon

In north Gujarat and Kutch the presence of nutritious

campsites we studied in the Barda hills boast of

grasses, water in inter dune depressions and the

a wealth of architectural features. As mentioned

availability of various salts and minerals in this area

above the slopes of foothills are usually preferred for

attracts even those pastoralist groups who may not

locating the camp settlements. The complexes are

undertake long summer migrations to south or east

provided with one or two circular huts along with

Gujarat later. Pastoralists here construct huts of thorn

animal enclosures. Usually stone are used to construct

bushes over hastily made circular plaster floors and are

the animal pens as well as the living area, which are

often linked together to animal enclosures (Figures

provided with hearths and wattle and daub super

9 and 10). Usually, the camps are set on a dune top

structures. At regular intervals the walls of the animal

or its slope to avoid rainfall accumulation. One of

enclosure have been provided with outlets to drain

the obvious restrictions is on the location of camps

out rainwater. Some of the complexes have raised

near or on the black soil and other soils because the

stone platforms to hold water vessels (Figure 11).

animals get infected by muddy hooves and are also

ideal for transmitting a variety of diseases. The water

Diwali festival, which falls in later part of October or

- 19 -

In c ontra st to th e e xp e c tati ons o f many

The year of these breeders usually commences with

Kuldeep K. Bhan

Figure 12 Maldaris on their annual migration from north Gujarat

early November (Figure 12). At this time each breeder

make them enduring from an archaeological point of

and their family leave for their monsoon camp and

view.

at the onset of summer they disperse into south and

eastern Gujarat. Of specific interest to archaeologists

pastoral camps have indicated that the vessels with

are archaeological material consequences and the

constricted necks and narrow orifice have a greater

location of such camps. During summer season

frequency than the vessels with short necks and wide

they move with their herds, halting 1-2 days on the

orifice given their importance in transporting water

harvested fields or urban vacant lands, leaving behind

and dairy products during migration in pastoral

ephemeral hearth scars and little debris. At times they

economies. This has been further substantiated by

may also spend 4-20 days in summer camp usually

ethnographic studies carried out in two pottery

after negotiating with landowner. Occupational debris

manufacturing centers in Kutch (Choksi 1995: 87-

is ploughed back into the field leaving little evidence

104). This is one of the important indicators that

of the summer encampment, while the rainy season

could be used to delineate pastoral assemblages. It

camps usually have substantial stone or thorn animal

is interesting to note here that the ceramics analysis

enclosures and plastered platforms or living space.

of one of the Late Harappan settlements in north

These features are of special interest because some

Gujarat has revealed interesting results. The analysis

camp sites are repeatedly used. While evidence for

carried out by one of our graduate students from the

the individual monsoon season occupation in certain

kiln area at Datrana has indicated that more than

cases perhaps may not preserve very well, the fact

sixty percent of vessels recovered from the area have

that these features are at times refurbished and reused

comparatively long necks and narrow orifices (Devi

- 20 -

Morphological studies of the vessels from five

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

2000: 51-56). Similar studies from other sites of north

was perhaps the result of increased environmental

Gujarat perhaps will also be rewarding.

degradation combined with a break down of the

Harappan integration sphere around 2000 B.C.
It can be further proposed that the settlements

CONCLUSIONS

close to R anns or Khar phats having g reater

archaeological debris as monsoon season camps,

The Harappans of Gujarat were perhaps sure of two

while settlements like Dhama, Mujpur, Nayaka,

things: First, that in the not too distant future the

Vagel, Vanod, Valewada, Papotpura, etc. (Figure 8)

monsoon would be bad enough to ruin his subsistence

have comparatively very little habitational deposit as

production; second that things would be better

summer encampments. Dispersion and aggregation

somewhere else. This type of monsoon places a heavy

of prehistoric population in north Gujarat is seen

premium on subsistence production. Pastoralism and

as result of seasonal availability of animal food

agriculture play complementary roles, keeping large

resources. However, an interdisciplinary approach is

numbers of animals acts as a food reserve for those

warranted to study the seasonality of these sites. Of

times when crops fail as a result of unpredictable,

particular importance in such studies would be the

but regular, disasters such as draught, and perdition

use of ethnoarchaeobotanical, (like the one developed

by insects and birds. Therefore in such a situation a

by Reddy (1994), for Saurashtra), zooarchaeological

mixed pastoral-agriculture background strategy may

data, ethnographic models and improved excavation

be the most secure way to ensure long survival. It is

methods like open area excavation and single layer

in this environmental background the proliferation

plan as discussed by Harris (1979). Whether this

of pastoral encampment in the arid locale of north

hypothesis presented here is proven correct or not,

Gujarat should be viewed. In summary, it can be

further investigation in north Gujarat based on

proposed that the proliferation of settlements in the

the model presented here will contribute to better

arid locale of north Gujarat, that provided nutritious

understanding of the Harappan tradition in South

grasslands rich in minerals and salts, water in the

Asia.

inter dune depressions and the presence of sand

dunes provided an excellent breeding ground for the

Acknowledgments

animals. The decentralization of Harappan state in the

T h i s p ap er c o m e s f r o m th e a r c ha e o l o g i c a l

early second millennium B.C. as suggested by Reddy

research work of last more than two decades and

also (1994) and perhaps accompanied by degradation

ethnoarchaeological research undertaken carried out

in environment allowed for wider range of choices

during 1989-1995 in Gujarat. The paper was written

of economic adaptation and particularly facilitated

in calm and beautiful surrounding of International

the integration of two distinct but contemporary

Center for Japanese Studies, Kyoto (Nichibunken)

economies agriculture and pastoralism in the semi-

and I would like to acknowledge the support I

arid environment of Gujarat.

received from the entire faculty and particularly

Taking the environmental and locational criteria

my counter part, Prof. Takao Uno. I would like to

of the present day pastoral camps into consideration

thank Dr. Seetha Narhari Reddy for accompanying

it seems that the area has played a very crucial role in

me in the initial stages of the fieldwork. I was highly

the development of settlements in north Gujarat from

benefited by her critical comments and later her work

very early times. Increase in the number of pastoral

on palaeoethnobotanical study in Saurashtra. I would

camp settlements during Late Harappan phase

also like to acknowledge Dr. Paul Rissmans work on

- 21 -

Kuldeep K. Bhan

the excavation at Oriyo Timbo that influenced my

New Data from Bagasra, Gujarat. Paper presented at the

thought process. Professor and J. M. Kenoyer and Dr.

annual meeting of ISPQS, Pune.

Allchin, B and F.R. Allchin (1997) Origin of Civilization.

Massimo Vidale also share my acknowledgements for

Viking-Penguin India, New Delhi.

their encouragement and many discussions I had with

Bhan, K.K. (1989) Late Harappan Settlements of Western

them. I am indebted to Mr. Gregg Jamison for going

India, with Specific Reference to Gujarat, in J.M. Kenoyer

through the first draft of this paper and suggesting

(ed.) Old Problems and New Perspective in the Archaeology

many changes. Although, the above mentioned

of South Asia. Wisconsin Archaeological Report,

individuals, and many others have contributed

Madison. pp.61-74.

directly or indirectly to contents of this paper,

Bhan, K.K. (1994) Cultural Development of the Prehistoric

the author accepts the sole responsibility for the

Period in North Gujarat with reference to Western India.

interpretations presented here. At the end, I would

South Asian Archaeology 10: 71-90.

like to acknowledge the Ford Foundation Research

Bhan, K.K. (2001) In the Sand Dunes of North Gujarat,

Grant for carrying out the ethnographic work in

in Dilip K. Chakrabarti (ed.) Indus Civilzation Sites In

India: New Discoveries. Marg, Mumbai. pp.96-105.

Gujarat.

Bhan, K.K. and D.R. Shah (1990) Pastoral Adaptation in

the Late Harappan Tradition of Gujarat. Paper presented
in the seminar on Rising Tends in Palaeoanthropology:

Notes

Environmental Changes and Human Responses (Last

1) To answer the basic question, what is pastoralism, is not as

Two Million Years), Pune.

simple as it might see. In general, societies specializing

Bombay Presidency (1880) Gazetteer of the Bombay Presidency,

in animal husbandry are called pastoralist and those

vol. 5: Cutch, Palanpur, and Mahi Kanta. Bombay.

required periodic movement to maintain their animals

Bombay Presidency (1884) Gazetteer of the Bombay Presidency,

are called pastoral nomads. The term nomads are pastoral

vol. 7: Kathiawar. Bombay

nomads. The term nomads and pastoralists are generally

Bose, A.S. (1975) Pastoralism nomadism in India, in L.S.

used interchangeably, at basic level they are analytical

Leshnik and G.D. Sonhteimer (eds.) Pastoralists and

distinct, the former referring to movement and the later to

Nomads of South Asia. O. Harrasowitz, Wiesbaden. pp.1-

a subsistence.

15.

2) Jekhada is also spelt as Zekhada. We prefer to maintain

Bryson R.A. and A.M. Swain (1981) Holocene Variation of

Jekhada in our text the name referred in the Survey of

Monsoon and rainfall in Rajasthan. Quaternary Research

India map.

16: 135-45.

3) A strong argument for a late date of Surkotada IC has

Champion, H.G. (1935) A Preliminary Survey of the Forest

been put forward recently on the basis of recalibrated

types of India. Indian Forest Records 1: 1-204.

radiocarbon dates and the ceramic of this phase. Refer

Chitalwala, Y.M. (1990) The Disappearance of Rhino from

Herman 1997: 99.

Saurashtra: A study of Paleoecology. Bulletin of Deccan

College, Postgraduate and Research Institute 49: 79-82.

References

Choksi, A. (1995) Ceramic Vessels: Their Role in Illuminating

Agrawal, D.P. and R . Kusumgar (1973) Tata Institute

Past and Present Social and Economic Relationship. Man

Radiocarbon Date List X. Radiocarbon 15(3): 547-85.

and Environment XX(1): 87-104.

Agrawal, D.P. and R.K. Sood (1982) Ecological Factors

Desai, A. (1985) Spatial Aspects of Settlement Pattern, A Study

and the Harappan Civilization, in G.L. Possehl (ed.)

of the Narmada Command Area of Mehsana District,

Harappan Civilization. American Institute of India

Gujarat. New Delhi.

Studies, New Delhi. pp.223-252.

Devi, Y. Krishnabarnati (2000) Metrical Analysis of Sorath

Ajithprasad, P., V.H. Sonawane, A. Mujamdar and K. Dimri

Harappan Pottery from Datrana. Unpublished M.A.

(1999) The Harappan Cultural Sequence at Rangpur and

Dissertation, Department of Archaeology, M.S. University

- 22 -

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

Gujarat (India): a reassessment, in B. Allchin and F. R.

of Baroda.

Allchin (eds.) South Asia Archaeology 1995. Oxford and

Dikshit, R.B.K.N. (1938) Prehistoric Civilization of the Indus

IBH Publishing co. Pvt. Ltd., New Delhi. pp.187-198.

Valley. University of Madras Press, Madras.

Enzel, Y., L.L. Ely, S. Misra, R. Ramesh, R. Amit, B. Lazer,

He rma n , C . F. ( 199 7 b ) Ha r a p p a n G u j a r a t : T h e

S.N. Rajaguru, V.R. Baker, and A Sandler (1999) High-

Archaeological-Chronological Connections. Palorient

Resolution Holocene Environment Changes in Thar

22(2): 77-112.
Huntingford, G.W.B (ed.)(1980) The Periplus of the Erythrean

Desert, North Western India. Science 284: 125-128.

Fairsevis, W.A. (1956) Excavation in the Quetta Valley, West

Sea. The Hckluyt Society, London.

Pakistan. Anthropological Papers of the American

Imperial Gazetteer of India 1909, vol. II.

Museum and Natural History 45(2). American Museum

Jarriage, J.-F. (1995) Du neolithique a la civilization de I Inde

ancienne: Contribution des recherches archeologiques

of Natural History, New York.

dans le nord-ouset du souscontinent into Pakistanais. Ars

Foote, R .B. (1912) Indian Prehistoric and Proto-historic

Asiatiques L: 5-30.

Antiquities in the Indian Museum: Note on Age and

Keatinge, G.F. (1917) Note on Cattle of Bombay Residency.

Distribution. Government of Madras, Madras.

Department of Agriculture, Bombay.

Gaussen H., P. Legris, F. Blasco, V.M. Meher-Homji and J.P.

Troy (1968) Notice de la feuille Kathiawar. Extrait des

Kenoyer, J.M. (1991) The Indus Valley Tradition of Pakistan

Travaux de la section scientifique et technique de L Institute

and Western India. Journal of World Prehistory 5(4): 331-

Francais de Pondichery. Hors. Serie 9.

385.
Kenoyer, J.M. (2005) Culture Change during Late Harappan

George, S. (1985) Operation Flood: an appraisal of current

Period at Harappa: New Insights on Vedic Aryan Issue

Indian dairy Policy. Oxford University Press.

, in E.F. Bryant and L.L. Patton (eds.) The Indo-Aryan

Ghosh, S.S. and K.R. Lal (1963) Plant Remains from Rangpur.

Controversy: Evidence and Inference in Indian History.

Ancient India 18-19: 161-77.

Routledge, London/New York.

Government of Gujarat (1970) Gujarat State Gazetteers:

Kusumgar, R., D. Lal and R.P Saran (1963) Tata Institute of

Jamnagar District. Ahmedabad.

Radiocarbon Dates List I. Radiocarbon 5: 273-82.

Government of Gujarat (1969) Gujarat State Gazetteers:

Leshnik, L.S. (1968) Prehistoric Exploration in northern

Bhavnagar District. Ahmedabad.

Gujarat and parts of Rajasthan. East and West 18 (3-4):

Gupta, S.K. (1977) Holocene silting in the Little Rann of

295-310.

Kutch, in D.P. Agrawal and B.M. Pande (eds.) Ecology

Mackay, E.J.H. (1943) Chanhu-daro Excavations 1935-36.

and Archaeology of Western India. Concept Publishers,

American Oriental Series Vol. 20. Boston Museum of Fine

Delhi. pp.181-93.

Arts, New Haven.

Harris, E.C. (1979) Principles of Archaeological Stratigraphy.

Mann, H. (1916) Fodder Crop of Western India. Department

Academic Press Inc., New York.

of Agriculture No. 77. Bombay.

Hedge, K.T.M. (1977) Late Quaternary Environment in

Marshall, J. (1931) Mohenjo-daro and the Indus Valley

Gujarat and Rajasthan, in D.P. Agrawal and B.M. Pande

Civilization. A Probsthain, London.

(eds.) Ecology and Archaeology of Western India. Concept

Meadow, R . (1989) Continuity and Chang e in the

Publishers, Delhi. pp.181-93.

Hegde, K.T.M. and V.H. Sonawane (1986) Landscape

Agriculture of the Indus Valley: The Paleobotanical and

and Settlement Pattern in Rupen Estuary. Man and

Zooarcheological Evidence, in J.M. Kenoyer (ed.) Old

Environment X: 23-30.

Problems and the New Perspective in the Archaeology of

South Asia. Wisconsin Archaeological Report, Madison.

Hellbusch, S.W. (1975) Changing in meaning of ethnic names

pp.61-74.

as exemplified by the Jats, Rabari, Bharvad and Charan in

Northwestern India, in L.S. Leshnik and Gunther-Dietz

Mehta, R.N. (1984) Valabhi: A Station of Harappan Cattle

Sontheimer (eds.) Pastoralists Nomads in South Asia. Otto

Breeders, in B.B. Lal and S.P. Gupta (eds.) Frontiers of the

Harrassowitz, Wiesbeden.

Indus Civilization. Books & Books, New Delhi. pp.227230.

Herman, C.F. (1997a) The Rangpur Sequence of Harappan

- 23 -

Kuldeep K. Bhan

Mehta, R.N., K.N. Momin and D.R. Shah (1980) Excavations

Rao, R.K and K. Lal (1985) Plant Remains from Lothal,

at Kanewal. Archaeological Series No. 17. M. S. University

in S.R. Rao (ed.) Lothal: A Port Town (1955-62), vol

of Baroda.

II. Memoirs of the Archaeological Survey of India 78.

Archaeological Survey of India, New Delhi. pp.667-684.

Misra, V.N. (1965) Book Review: Ancient India, Bulletin of

Archeological Survey of India nos. 18-19 (1962-63). The

Rao, S.R . (1963) Excavations at Rang pur and other

Explorations in Gujarat. Ancient India 18, New Delhi.

Eastern Anthropologist 18(1): 44-52.

Misra, V.N. (1984) Climate, a factor in the rise and fall of the

Rao, S.R. (1978) Lothal: A Port Town (1955-62), vol.I. Memoirs

Indus Civilization: Evidence from Rajasthan and Beyond,

of the Archaeological Survey of India 78. Archaeological

in B.B. Lal and S.P. Gupta (eds.) Frontiers of the Indus

Survey of India, New Delhi.

Rao, S.R . (1985) Lothal: A Port Town (1955-62), vol.II.

Civilization. Books & Books, New Delhi. pp.461-489.

Momin, K.N. (1983) Excavations at Zekhda. Puratattva 12:

Memoirs of the Archaeological Survey of India 78.

Archaeological Survey of India, New Delhi.

120-125.
Mughal, M.R. (1997) Ancient Cholistan: Archaeology and

Reddy, S.N. (1991) Archaeological Investigation at Oriyo

Timbo (1989-90): A Post Harappan Site in Gujarat. Man

Architecture. Ferozsons, Lahore.

and Environment XVI(1): 79-81.

Mughal, M.R . (1992) Jhukhar and the late Harappan

Cultural Mosaic of the Greater Indus Valley, in C.

Reddy, S.N. (1994) Plant Usage and Subsistence Modeling:

Jarriage (ed.) South Asian Archaeology 1989. Prehistoric

An Ethnoarchaeological Approach to the Late Harappan

Press, Wisconsin Madison. pp.213-221.

of Northwest India. Ph.D Dissertation, the University of

Wisconsin, Madison.

Mughal, M.R . (1982) Recent Archaeological Research

in Cholistan Desert, in G.L. Possehl (ed.) Harappan

Rissman, P.C. (1985) Migratory Pastoralism in Western India

Civilization. American Institute of India Studies, New

in the second Millennium B.C., the evidence from Oriyo

Delhi. pp.85-96

Timbo (Chiroda). Ph.D Dissertation, the University of

Patel, A. (1985) Vertebrate Archaeofauna from Nagwada: A

Pennsylvania, Philadelphia.
Rissman, P.C and Y.M. Chitalwala (1990) Harappan

Preliminary Study. M.A. Dissertation, Department of

Civiliz ation and Oriyo Timbo. Oxford and IBH

Archaeology, M.S. University of Baroda.

Patel, G. (1977) Gujarats Agriculture. Ahmedabad.

Publishing Co./American Institute of Indian Studies,

Piggott, S. (1962) Prehistoric India. Second Edition. Cassell,

New Delhi.
Sankalia, H.D. (1974) Prehistory and Protohistory of India and

London.

Pakistan. Deccan College, Pune.

Possehl, G.L. (1980) Indus Civilization in Saurashtra. B.R.

Sankalia, H.D. and I. Karve (1949) Primitive Microlithic

Publications, New Delhi.

Cultures and People of Gujarat. South Western Journal of

Possehl, G.L. (1992) The Harappan Civilization in Gujarat:

Anthropology LI (I): 28-34.

The Sorath Harappan and Sindhi Harappan. The Eastern

Anthropologist 45: 115-154.

Shaffer, J.G. (1991) The Indus Valley, Baluchistan and

Helmand Tradition: Neolithic through Bronze Age, in R.

Possehl, G.L (1997) The Transformation of the Indus

Enrich (ed.) Chronologies in Old World Archaeology. Third

Civilization. Journal of the World Prehistory 2(4): 425-72.

Edition, vol. I. University of Chicago Press, Chicago.

Possehl, G.L. and P.C. Rissman (1991) The Chronology of

pp.441-456.

Prehistoric India: From Earliest Times to the Iron Age, in

R. Enrich (ed.) Chronologies in the Old World Archaeology.

Shaffer, J.G. and D.A. Lichenstein (1989) Ethnicity and

Change in the Indus Cultural Tradition, in J.M. Kenoyer

University of Chicago Press, Chicago. pp.465-490.

Possehl, G.L. and M.H. Rawal (1989) Harappan Civilization

(ed.) Old Problems and New Perspective in Archaeology of

and Rojdi. Oxford and IHB Publishing Co./American

South Asia. Wisconsin Archaeology Report, Madison.

Institute of Indian Studies, New Delhi.

pp.117-126.

Raikes, R .L and R .H. Dyson (1961) The Prehistoric

Singh, G. (1971) The Indus Valley Culture, seen in the

Climate of Bulachistan and the Indus Valley. American

Context of Post Glacial Climatic and Ecological Studies

Anthropologist 63: 265-281.

in northwest India. Archaeology and Physical Anthropology

- 24 -

Pastoralism in Late Harappan Gujarat, western India: an ethnoarchaeological approach

A Study in inter-relationship. Geophytology 5(4) 1: 46-53.

of Oceania 6(2):177-89.
Singh, G. (1977) Stratigraphic and Palynological Evidence for

Vishnu-Mittre and R . Savihtri (1982) Economy of the

Desertification in the Great Indian Desert. Annals of the

Harappans, G.L. Possehl (ed.) Harappan Civilization.

arid Zone 16(3): 310-20.

American Institute of Indian Studies, New Delhi. pp.20522.

Sonawane, V.H. and P. Ajithprasad (1994) Harappa Culture

and Gujarat. Man and Environment XIX(1-2): 129-130.

Vishnu-Mittre and C. Sharma (1979) Pollen Analytical

Studies at Nal Lake (Nalsarovar). Palaeobotanist 26(1): 95-

Sonawane, V.H. and R .N. Mehta (1985) Vagad: a rural

Harappan Settlement in Gujarat. Man and Environment

104.
Wadia, D.N. (1957) Geology of India. Macmillan, London.

IX: 38-44.
Spate, O.H.K. and A.T.A. Learmonth (1967) India and

Weber, S. (1991) Plant and Harappan Subsistence. Oxford

Pakistan: A General and Regional Geography, third

and IBH Publishing Co./American Institute of Indian

edition. Methuen, London.

Studies, New Delhi.

Weber, S. (1989) Plant and Harappan Subsistence: An Example

Stuckenrath, R . Jr. (1967) University of Pennsylvania

of Stability and Change from Rojdi. Ph.D. Dissertation, the

Radiocarbon Dates. Radiocarbon V: 82-103.

Suraj Bhan (1975) Excavation at Mitathal (1968) and other

University of Pennsylvania, Philadelphia.

Wagne, G.E. (1982) Floatation for Small-artifact Recovery

Exploration in the Sutlej-Yamuna Divide. Kuruskshetra

at Oriyo Timbo. University Museum, University of

Univeristy, Kuruskshetra.
Vats, M.S. (1940) Excavation at Harappa. Government of

Pennsylvania, Philadelphia.
William, L.F.R. (1958) The Black Hills: Kutch in History and

India Press, New Delhi

Legend. Booksellers Association of Great Britain and

Vishnu-Mittre (1961) Plant economy in Ancient Navdatoli-

Ireland, London.

Maheshwar, in J. Clutton-Brock, Vishnu-Mittre and A.N.

Wheeler, R.E.M. (1958) The Indus Civilization. Cambridge

Gulati (eds.) Technical Report on Archaeological Remains.

University Press, Cambridge.

Deccan College, Pune. pp.13-52.

Whyte, R.O. (1964) The Grassland and Fodder Resources of

Vishnu-Mittre (1974a) Palaeobotanical Evidence in India,

India. Council of Agriculture Research, New Delhi.

in S.J. Hutchension (ed.) Evolutionary Studies in World

Zeuner, F.E. (1963) Environment of early Man with Special

Crops. Cambridge University Press. pp.3-33.

Reference to the Tropical Regions. Baroda.

Vishnu-Mittre (1974b) Plant Remains and Climate from the

Late Harappan and other Chalcolithic Cultures of India:

- 25 -

Kuldeep K. Bhan

- 26 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Archaeological explorations in the Saurashtra coast of

Junagadh District, Gujarat
P. Ajithprasad, SV Rajesh and Bhanu Sharma
Department of Archaeology and Ancient History,
The M.S. University of Baroda, Vadodara India

Abstract
A recent exploratory survey carried out in the Saurashtra coast in Junagadh district of Gujarat in western India revealed several
Harappan/Chalcolithic and Historic period sites. The newly discovered Chalcolithic sites included Prabhas ware sites and a few
sites showing Early Harappan Sindh related ceramic assemblages. This brief report discusses the importance of these new discoveries
in understanding the Chalcolithic cultural development and patterns of settlements that evolved in the region through the Historic
period.

Introduction

habitation in Saurashtra has changed drastically.

Discovery of several new sites and identification

The discovery of Harappan relics from Rangpur

of a few region specific Chalcolithic cultural

on the Bhadar river in Saurashtra in the early 1930s

traditions at Somnath/Prabhas Patan and Padri

invoked considerable interest in the geographical

along the Saurashtra coast (Figure 1) in Junagadh

spread of the civilization beyond the Indus basin.

and Bhavnagar districts respectively played a major

Two decades later, in the 1950s, similar discoveries

role in bringing about this change (Subbarao 1958,

from Lothal and several other sites in the region

Shinde 1992, Shinde and Kar 1992, Sonawane and

have firmly established the southward extension of

Ajithprasad 1994, Ajithprasad 2004). Of the two, the

the Harappa culture into Saurashtra (Figure 1). In

Prabhas remains was the first regional Chalocolithic

fact, the Harappan influence was felt much further

tradition affiliated to the Harappa culture to be

south in the estuaries of the Narmada and the Tapi in

identified after the joint excavation in 1955-56 by the

south Gujarat (Rao 1958). Over the years, more than

M. S. University of Baroda and the then Bombay

700 Harappan and Harappan affiliated Chalcolithic

State Archaeology Department at Prabhas Patan/

sites have been reported from Gujarat (Possehl 1997,

Somnath in the Saurashtra coast of Junagadh district

Rajesh and Patel 2007). A quick glance through the

(Subbarao 1958, IAR 1955 to 57, Nanavati et al. 1971).

distribution of sites will suggest that majority of the

Further excavations at the site in 1972 by the Deccan

above sites are located in Saurashtra. The Saurashtra

College, Pune and the Gujarat State Archaeology

region or Kathiawad as it is traditionally known

Department resulted in the discovery of an even

therefore figures prominently in the Harappan studies

earlier Chalcolithic cultural assemblage (the Pre-

in Gujarat.

Prabhas) stratified below the Prabhas remains (IAR

In t h e l a s t t w o o r t h r e e d e c a d e s , o u r

1971-72, Dhavalikar 1984). The identification of

understanding about the Harappan/Chalcolithic

Pre-Prabhas assemblage from Prabhas Patan as a

- 27 -

P. Ajithprasad et al.

Figure 1 Location of Prabhas Patan/Somnath and other major Harappan Chalcolithic sites mentioned in the paper

regional Chalcolithic culture that existed prior to the

traditions therefore must have played a major role

beginning of the Harappan tradition in Saurashtra

in the Harappan and the Sorath Harappan cultural

presented a picture of Chalcolithic habitation in the

developments in Saurashtra. However, very little is

Saurashtra coast at least 500 years earlier than hitherto

known about these two early Chalcolithic cultural

believed (Dhavalikar and Possehl 1992, Sonawane and

traditions; especially their geographical spread and

Ajithprasad 1994, Sonawane 2000, Ajithprasad 2002).

other settlement features. A systematic exploratory

Meanwhile, excavation at Padri in the Bhavnagar

survey was therefore carried out in and around

coast of Saurashtra in the early 1990s brought to light

Prabhas Patan in the Saurashtra coast with a view

the Padri Chalcolithic assemblage that was very

of assessing the importance of Prabahas and other

different not only from the Pre-Prabhas and Prabhas

chalcolithic assemblages in the region.

but also from the Harappan. Dated to the middle of

fourth millennium BCE by radiocarbon, the regional
Chalcolithic remains reported from Padri are even
earlier than the Pre-Prabhas (Shinde 1992, Shinde
and Kar 1992). These two early regional Chalcolithic
- 28 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Exploration Planning
and Strategy

Environmental Settings
The Saurashtra Peninsula in western India is a

The Prabhas and the Pre-Prabhas assemblages at the

reasonably large landmass jetting in to the Arabian

type-site Somnath/Prabhas Patan are culturally

Sea (Figure 1). It is separated in the east from the

and stratigraphically well documented. Among

mainland by the Gulf of Cambay/Khambhat and in

the two, the Pre-Prabhas is a very rare assemblage

the north by the Gulf of Kachchh. The central part

reported until very recently only from the early levels

of Saurashtra is an elevated and rugged hilly land

at Prabhas Patan. Little is known about its spread

formed of the Deccan lava flow. The rather flat and

into geographically nearby areas. The Prabhas pottery

narrow coastal plain is formed of tidal mudflats,

on the other hand is slightly better known, on being

sandy beaches and even coral reefs along the Jamnagar

reported in small numbers in the excavations from

coast of the Gulf of Kachchh. The narrow coastal

Lothal and Rojdi in association with the Harappan

land in between, drained by several small rivers and

remains. Besides, nine more sites in central Saurashtra

streams, had supported human occupation from the

showing Prabhas type pottery had been reported by

Palaeolithic times onwards. The Chalcolithic site at

P.P Pandya, S.R. Rao and others in the late 1950s

Prabhas Patan in Junagadh district is located on the

(IAR 1958). Barring these reports, the immediate

fringes of this fertile strip in the southern coast of

geographical spread of these two assemblages is rather

Saurashtra.

poorly known (Rao 1963). This has greatly impeded

in assessing the regional character and impact of

deposits are found along the seacoast and in the

this tradition in the overall cultural development

interior Saurashtra. The most predominant basal

of the region. The current exploration therefore

rock in the region is the Deccan Trap, which in the

was undertaken to locate Chalcolithic sites along

central part of Saurashtra is exposed extensively as the

the Saurashtra coast especially in the vicinity of

rugged tableland at a general height varying from 300

Somnath/Prabhas Patan. It is assumed that if there

to 600 m. There however are several relict tableland

are Chalcolithic cultural relics dating back to 3000

features that form high hills. Most prominent among

BCE showing region-specific traits at Somnath there

these is the Girnar hill in Junagadh district which at

is a strong possibility of coming across a few more

1117m is the highest in Saurashtra followed by the

sites contemporary to the Prabhas assemblage in the

Barda hills which is 637m (Merh 1995). In addition

vicinity of Prabhas Patan. As a working strategy we

to this, sedimentary formations of the Tertiary

selected an area falling within 40 to 50 km radius

sandstones and Quaternary miliolites are found as

from Somnath along the coast in the first leg for an

basal rocks especially in the coastal areas. These two

intensive foot-survey for locating sites after assessing

formations today are quarried out in an alarmingly

the geographical and environmental setting of the

large scale as they are used as raw materials in the

region. We used topographic sheets of 1:250,000

flourishing cement and detergent industries. The

and 1:50,000 scale as well as satellite images for this

coastal region also has several littoral dune formations

purpose.

and beach sand deposits. The area along the coast is

Several types of rock formations and Quaternary

mostly covered by a sandy loam and the black cotton

soil whose thickness varied considerably. The mantle
of soil cover has maximum thickness near the coast
and it becomes thin towards north in the interior.
- 29 -

P. Ajithprasad et al.

Figure 2 Location of Chalcolithic and Historic period sites in

the Saurashtra coast of Junagadh and Porbandar districts of Gujarat
Chalcolithic sites: 1. Somnath/Prabhas Patan, 2. Warodra-I, 3. Haveda Bapa (Warodra-II), 4. Bhana bhai no Khetar, 5. Khandini
Timbo, 6. Khandero, 7. Bhavani Timbo, 8. Shapur, 9. Lohij-I (Sanavav Pa-1), 10. Lohij-II, 11. Lohij-III, 12. Kokru, 13. Khaganath
no Timbo, 14. Madhavrai no Timbo.
Historic Period sites: 15. Kalej-I, 16. Kalej-II, 17. Thumliu no Timbo-I, 18. Thumliu no Timbo-II, 19. Humbliu Timbo, 20.
Gelki, 21. Vakodar, 22. Vanankio Timbo, 23. Gayatrimata ni Thumdo-I, 24. Ganeswari ni Timbo, 25. Dhoro Timbo, 26. Madsimar,
27. Panchnath Mahadev, 28. Sonesa ni Timbo.

This tract is extremely fertile and a major part of the

of Junagadh district that receive more than 800 mm

modern agriculture is confined to this narrow strip

annually. The highest rainfall that had been recorded

of land that affords cultivation of food grains/millets

in this region in recent years was 1105 mm in 1998.

such as bajra, jowar; pulses as well as groundnuts,

The area therefore is quite green and supports good

cotton and other cash crops.

cultivation and natural growth of vegetation. The

Saurashtra falls within the semi-arid climatic

annual average winter temperature is 11 Celsius and

zone wherein the rainfall ranges from 400 mm

the maximum temperature in the summer ranges

in the northern region to 800 mm in the south-

from 38 to 41 Celsius. The hilly forested terrain here

eastern Saurashtra coast. Rainfall in the coastal area

is predominated by mixed deciduous type vegetation.

is relatively less as it ranges from 400 to 600 mm. It

The littoral landscape along the coastal line is marked

is a bit better, about 700 mm, further inland. There

by typical coastal topography with lush green tropical

are some pockets of heavy rainfall in the hilly areas

vegetation mixed with scrub-jungles where the soil

- 30 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Table 1-1 Newly discovered/revisited Chalcolithic and the Historic Period sites along the Junagadh coast of Saurashtra.
Chalcolithic sites
#
1

Site
Somnath/Prabhas Patan (Nagara Timbo)

Village,

Approx. size

Cultural affiliation and remarks

Somnath

600 x 150 m

Pre-Prabhas, Prabhas and Early Historic

20 54N, 70 25.3E
2

Warodra-I

(9 htr.)
Warodra

10 x 10 m

(20 48.75N; 70 31.41E)

3

Haveda Bapa (Warodra -II)

(0.01 htr.)
Warodra

200 x 135 m

20 49.07N; 70 31.07E
4

Bhana bhai no Khetar

Early Harappan Sindh related pottery.

Sorath Harappan + Early Historic

(2.7 htr.)
Pasnawara

95 x 50 m

(Gayatrimata ni Thumdo-II)

Sorath Harappan + Early Historic

(0.48 htr.)

20 49.36N; 70 33.33 E
5

Khandini Timbo

Lodhwa

20 49.35N; 70 35.52E
6

Khandero (Bhimbhai no Khetar)

Lodhwa

2049.45N, 70 35.46E
7

Bhavani Timbo

Dhamlej

20 47N, 70 35.9 E
8

Shapur (Puncha Hira ni Khetar)

Shapur

21 5.65N; 70 7.56E
9

Lohij-I (Sanavav Pa-1)

Lohij

21 9.53N, 70 3.45E
10

Lohij-II (Sanavav Pa-II)

Lohij

21 9.49N, 70 3.50E
11

Lohij-III (Devariya Nagari)

Kokru
Khaganath no Timbo
Madhavrai no Timbo

Historic

25 x 20 m

Sorath Harappan (Heavily abraded) +

(0.5 htr.)

Late Medieval

100 x 70 m

Sorath Harappan + Early

(0.7 htr.)

Historic

220 x 140 m

Prabhas + Early Harappan. Shell bangles

(3.08 htr.)

+ Late Medieval

175 x 40 m

Prabhas, Shell objects;

(0.7 htr.)

Stone structures

10 x 10 m

Prabhas/Sorath Harappan.

(0.01 htr.)

Very few descriptive potshreds.

---------Sil

200 x 75 m

Historic and Medieval

Prabhas

(1.75 htr.)
Atroli

2113.14N, 70 0.33E
14

(0.12 htr.)

Early Harappan Sindh pottery + Early

21 10.81N, 70 2.70E
13

Prabhas + Sorath Harappan + Early

Lohij

21 9.66N; 70 3.1E
12

40 x 30 m

Mul-Madhavpur

2115.32N; 69 58.79E

135 x 120 m

Prabhas + Sorath Harappan + Late

(1.6 htr.)

Medieval

65 x 30 m

Sorath Harappan

(0.19 htr.)

cover is thin.

channel and flows into the Arabian Sea near Somnath.

Drainage in Saurashtra follows a radial pattern

The Chalcolithic Prabhas site at Somnath is located

as most of the rivers and streams originate from the

very close to the seashore on the left bank of the river.

elevated plateau in the centre. The present study area

The Saraswati is about 52 km long and joins the Hiren

in the Junagadh coast is drained by four major rivers,

near Kalij village about 2 km east of Somnath. The

the Hiren, the Saraswati, the Megel and the Noli and

third river Megal joins the Arabian Sea about 33 km

their tributaries (Figure 2). As they originate from the

northwest of Somnath between Simar and Chorwad

south of the elevated central Saurashtra plateau, the

villages. It has about 46 km long channel. The Noli is

rivers do not have great channel length. The longest

a small stream with just 36 km long, narrow channel

among these is the Hiren, which has a 64 km long

that joins the Sea near Sariyaj village. Yet another

- 31 -

P. Ajithprasad et al.

Table 1-2 Newly discovered/revisited Chalcolithic and the Historic Period sites along the Junagadh coast of Saurashtra.
HIstoric Period sites
#
15.

Site
Kalej-I ( Jiva Bhai no Khetar)

Village

Approx. size

Cultural affiliation and remarks

Kalej

45 x 30 m

Early Historic (RPW)

20 53.58N; 70 29.95E
16

Kalej-II (Pinch Per)

(0.13 htr.)
Kalej

20 53.45N; 70 30.1E
17

Thumliu no Timbo
Humbliu Timbo
Gelki

Umbri

Vakodar

Sutrapada

Vanankio Timbo

200 x 100 m

Early Historic (RPW)

200 x 200 m

Medieval + Early Historic (RPW)

(4 htr.)
Ghorakhmadi

20 55.36N: 70 32.4E
21

Early Historic (RPW)

(2 htr.)

20 53.61N; 70 31.05E
20

160 x 120 m
(1.92 htr.)

20 54.07N; 70 31.71E
19

Early Historic (RPW)

(0.49 htr.)
Ghorakhmadi

20 54.4N; 70 31.6E
18

70 x 70 m

300 x 150 m

Medieval + Early Historic (RPW)

(4.5 htr.)
Warodra

(Warodra-III)

200 x 200 m

Early Historic

(4 htr.)

20 48.84N; 70 32.04E
22

Gayatrimata ni Thumdo-I

Pasnawara

20 49.3N; 70 33.2E
23

Ganeswari ni Timbo
Dhoro Timbo

Adri

Madsimar

Shidogar

Panchnath Mahadev
Sonesa ni Timbo

200 x 150 m

Early Historic + Late Medieval?

15 x 10 m

Late Medieval

(0.15 htr.)
Chorwad

(21 01.77N 70 12.88E

27

Early Historic + Late Medieval

(3 htr.)
Simar

21 1.04N; 70 16.73E
26

200 x 200 m
(4 htr.)

20 56.72N 70 8.6E
25

Early Historic + Late Medieval

(0.83 htr.)

2057.7N; 7017.13E
24

110 x 75 m

12 x 11 m

Early Historic

(0.13 htr.)
Gotana

21 5.40N, 70 12.84E

145 x 105 m

Early Historic + Late Medieval

(1.5 htr.)

small stream having just about 18 km long channel

the team moved away to adjoining areas after taking

that meets the Sea further northwest near Sil village

into account of the natural depressions and other

is the Nali. It drains the village Lohij and surrounding

geographical features such as the presence of fertile

area which has reported three Chalcolithic sites in the

tracts, marshy areas and rocky/stony waste etc., which

exploration.

are readable in the Survey of India topographic maps

and satellite images of the region.

Assuming that prehistoric or Chalcolithic

settlements would be linked with perennial water

bodies as they were dependent on water availability,

both known and newly located sites and their

we thought it would be a good strategy to use the

geographical setting in detail. Geo-coordinates and

course of these river channels as the main guideline

approximate height from the sea level are recorded

to carry out our foot survey. Ideally speaking after a

using hand held GPS, Garmin eTrex Vista. A total

thorough survey of the areas close to the riverbanks

27 sites were either newly located or revisited during

- 32 -

As a part of the recording strategy we recorded

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

The Chalcolithic Sites

the exploration (Table 1). The sites belonged to

different cultural periods starting from the Harappan
Chalcolithic to the Late Medieval period through the

The total number of Chalcolithic sites found in this

Early Historic. Further divisions within these major

region is 14. Three among these are found along the

cultural periods of occupation have been made on

course of a small stream channel that drains through

the basis of the predominance of regional cultural

Lodhwa and Dhamlej villages about 28 km east of

traits within the artefact assemblages and taking into

Somnath. Rest of the sites are all located just on

account their chronological context. The Harappan

the seacoast; some of them, in fact, on the seashore

Chalcolithic sites have therefore been divided into:

washed by the high tide today. A short description of

(a) The Early Harappan Sindh related assemblage, (b)

each site is given below.

the Pre-Prabhas, (c) the Prabhas and (d) the Sorath

Harappan. A cursory look at the distribution of sites

1. Prabhas Patan/Somnath

in the explored area will reveal that there is a good

(20 54N, 70 25.3E)

concentration of sites in the coastal region during the

Locally known by Nagara Timbo, this ancient site is

Early Historic and Medieval periods (Figure 2). Sites

located on the right bank of the Hiren river. This is

belonging to the Chalcolithic period are also equally

a large and impressive site spread about 600 x 150 m

abundant along the coast.

along the right bank of the river and has an over all

The list includes 14 sites belonging to the

7.0 m thick cultural deposit (Figure 3). This is the

Harappan Chalcolithic and 13sites belonging to the

type-site for the Prabhas and the Pre-Prabhas culture

Historic period (Table 1). A large number of sites, 9 of

remains, as the excavations in the 1950s (Subbarao

them, showed both Chalcolithic and Historic period

1958, Nanavati et al. 1971) and in the 1970s (IAR

occupation. Seven among the fourteen Chalcolithic

1955 to 1977, Dhavalikar 1984) had provided enough

period sites are predominated by the Prabhas pottery

evidence of these two Chalcolithic cultural traditions

and three sites show evidence of Early Harappan

in the earliest levels at the site.

Sindh related assemblage. Other Chalcolithic sites

show primarily the Sorath Harappan affiliation. The

was occupied during the Early Historic period, which

historic period occupation along the coast starts from

is represented by the Red Polished Ware (RPW),

the Early Historic period as indicated by the presence

red ware, grey ware and black-and-red ware. The

of Red Polished Ware (RPW) and sherds of torpedo

south-western side of the mound has now partially

jars or the so called Roman Amphorae. Among the

been encroached by agricultural activities causing

18 sites belonging to this period, 6 are reasonably large

enormous destruction. There is little evidence of

and rich, measuring between 2 and 4 hectare. Some of

Harappan or Chalcolithic occupation in the artefact

the sites show continued occupation in the Medieval

assemblages found on the surface as they are buried

period too. Most of the medieval sites in the area are

deep at the bottom of the mound/site. However, rain

of Late Medieval period. A brief description of the

gullies and the sections exposed by the construction of

sites given below provides geomorphic context and

a recent road cutting across the site revealed isolated

general archaeological content of the sites.

potsherds and remains of stone-built structures of the

Most of the upper part of the mound actually

Chalcolithic period.

- 33 -

P. Ajithprasad et al.

Figure 3 The Chalcolithic mound at Prabhas Patan

2. Warodra-I

cluster of pottery found at Warodra-I and is also just

(20 48.75N; 70 31.41E)

on the seashore. The site probably on a stabilized

This site is a small cluster of potsherds that are found

coastal sand dune is now being systematically

exposed on the sea shore in between two small

destroyed by the sand mining activity going on in the

elevations known as Vanankio Timbo and Haveda

area. More than half of the site is completely razed

Timbo in the south of Warodra village (Figure 4).

down by this activity. Besides, the site is bisected by a

The fact that the cluster is exposed due to erosion and

road track. Sections exposed by these activities clearly

high tide activity on the shore is a clear indication

show cultural deposits belonging to two periods: the

of the extent of damage these natural agents may

Harappan and the Historic

have caused to the site. What is more interesting

is that the pottery collected from this site showed

a few isolated potsherds and shell bangle fragments

close resemblance with the pottery reported from

which probably have come from the early levels

Early Harappan Sindh related pottery from north

(Figure 7). The pottery shows Sorath Harappan

Gujarat (Figures 5 and 6). The assemblage included

features. Since the basal part of the sections cut in the

some coarse pottery, which appeared hand-made.

deposit was all found covered by a pile of thalus, it was

Unfortunately, in both the categories, there are no real

not easy to ascertain the actual context of this pottery.

rim-sherds in the collection, except one, that would

The bulk of the subsequent deposit measuring 1.50

have helped the identification more precise in terms

m belongs the Historic period. Apart from pottery,

of its cultural affiliation. In addition to pottery there

remains of several stone structures are also reported

are a few T. pyrum shell columella with saw-cut marks

from this deposit.

that are found at the site.

The Harappan relics at the site are restricted to

The Historic period occupation belonged to the

medieval period after which the site was abandoned.

3. Haveda Bapa (Warodra-II)

What is important as far as the abandonment of

(20 49.07N; 70 31.07E)

the Medieval occupation is concerned is the 70 to

This site is located about 200 m northwest of the

80 cm thick sand deposit that buries the site almost

- 34 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 4 Warodra-I, The Early Harappan Sindh pottery site at Warodra-I washed away by high tides

Figure 5 Close view of the in situ Early Harappan Sindh type pottery at Warodra-I

- 35 -

P. Ajithprasad et al.

Figure 6 Early Harappan Sindh related pottery from Warodra-I

Figure 7 Harappan pottery from Haveda Bapa, Warodra-II

- 36 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

completely. As this deposit appears to be a local

incised decoration (Figures 12 and 13). Some of the

phenomenon, it is possible that the deposit was

pot-sherds show bichrome paintings. This site has a

formed by a localized high-energy event that hit the

few shell columella and small pieces of shell bangles as

coast in the Medieval times. Further investigation is

well as a few mammal bones.

needed to ascertain the veracity of this event.

The Early Historic occupation at the top of

the mound in the northwest of Prabhas occupation

4. Bhanabhai no Khetar (Gayatrimata ni

measures 100 x 90 m and has a thickness about 1.5 m.

Thumdo-II)

Pottery recovered from this site is predominated by

(2049.36N; 7033.33E)

RPW and associated wares.

The site is about 500 m east of the Gayatri Mandir on

the way to the temple from Pasnawara village. Some

6. Khandero (Bhimbhai no Khetar)

part of the mound is destroyed by recent draught

(2049.45N, 7035.46E)

relief digging. The soil on this mound is a bit stony as

This site is located in the field about 100 m north of

at many places the basal rock is found exposed on the

the site Khandini Timbo across the Vekri stream. It

surface. The Harappan site is identified on the basis

seems the large slightly elevated field in fact is cut

of a few potsherds primarily that of basins and pots

into two by the stream. The entire field is now under

(Figure 8), spread to an area at present covering 95 x

heavy cultivation enhanced by the new irrigation

50 m. In addition to the Harappan pottery there are

techniques. This site also has a prominent cultural

also Early Historic pottery represented by the RPW

deposit of the Early Historic period whose remains

and other types as well as Medieval period sherds

almost obliterate the Harappan artefact. It is only

all on the surface with no stratigraphic distinction.

after the recent intense agricultural activity the

There are a few cut-columella of shell and shell

underlying Harappan artefacts have started getting

working wastes collected from the site (Figure 9). It

exposed on the surface. The Harappan occupation

is not very easy to suggest to which cultural context

at the site is represented by a cluster of badly abraded

they may belong to, as they are all collected from the

pottery spread to an area roughly covering 10 x 10 m.

surface.

This could be either part of the Khandini Timbo site

or a seasonal occupation near to the stream by the

5. Khandini Timbo

Harappans.

(2049.35N; 7035.52E)
The site is located about 3 km southeast of Lodhwa

7. Bhavani Timbo

village on the bank of a Vokri/small stream. The

( 2047N, 7035.9E)

Harappan assemblage is found clustered on the south-

The site is located about 2 km south west of the

eastern edge of the large mound, which also has an

Dhamlej village. It is a reasonably large mound

Early Historic period deposit on the top.

measuring 100 x 70 m in size and has a height close

The Harappan site measures 40 x 30 m and is

to 1.0 m from the surroundings. Bulk of the artefacts

exposed due to intense agricultural activity. The total

found at the site is of Early Historic period including

deposit may not be more than 60 to 70 cm thick

a terracotta animal figurine (Figure 43). However

at this site. The Prabhas and the Sorath Harappan

there are a few Harappan potsherds, mostly non-

potteries (Figures 10 and 11) collected from the site are

descriptive ones (Figure 14), found within the

predominated by convex sided bowls of the Prabhas

Historic assemblage on the surface. It is therefore

type, pots with incipient rim and coarse wares with

quite possible that the Harappan occupation of the

- 37 -

P. Ajithprasad et al.

Figure 8 Harappan potsherds from Bhanabhai no Kheter, Pasnawara

Figure 9 Shell bangles and craft waste from Bhanabhai no Khetar (a few of the shell bangles are of the Early Historic period)

- 38 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 10 Prabhas and Sorath Harappan pottery from Khandini Timbo, Lodhwa

Figure 11 Prabhas and Sorath Harappan pottery from Khandini Timbo, Lodhwa

- 39 -

P. Ajithprasad et al.

Figure 12 Prabhas and Sorath Harappan Pottery from Khandini Timbo, Lodhwa:
1, 2, 4, 6 Prabhas bowls and dish; 5, 7-9 Sorath Harappan Red ware; 10 Coarse balck and red ware with incised decoration
(1, 6 reddish brown slip and dark pinkish brown painting, 2 Buff slip, 6 white and dark pinkish brown painted lines)

- 40 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 13 Prabhas and Sorath Harappan Potteries from Khandini Timbo, Lodhwa:
1-12 Red ware (1, 10 Light reddish brown slip) Large open mouthed vessels, 5-8 small/medium size pots/jars

- 41 -

P. Ajithprasad et al.

Figure 14 Harappan pottery from Bhavani Timbo, Dhamlej

site is partly concealed by the later, Early Historic

IIB and IIC type. A few Early Harappan Sindh type

deposits.

potteries, especially medium size pots/jars with out

turned rim and shallow dishes are also found in the

8. Shapur-I (Puncha Hira Chavda no Khetar)

collection (Figures 22 and 23). A few pottery discs

(2105.65N; 7007.56E)

and a small shell bangle piece are the other antiquities

The site is located about 2 km southwest of Shapur

found at the site (Figure 24). The collection includes

village on the Mangrol - Bara road and is about 500

a few mammalian bones and teeth too (Figure 25). A

m away from the seashore. The site is a large oblong

few isolated sherds of the Late Medieval pottery are

mound measuring 200 x 150 m (Figure 15). A recently

also reported from the site. Overall thickness of the

constructed partition wall divides the mound into

deposit is about 1.50 m.

eastern and western half and the eastern half is badly

destroyed by relief digging work. This has exposed

9. Lohij-I (Sanavav Pa-I)

a number of rubble stones on the surface at the site

(2109.53N; 70 8.45E)

suggesting the presence of buried stone structures.

The site is located about 2 km south of the village

The western half is less disturbed as it is only used for

and is at present an agriculture field. It measures

cultivation.

approximately 0.70 hectare (175 x 40 m) and more

The artefacts recovered from the site includes the

than half of the site is under intense cultivation and

Harappan/Chalcolithic pottery especially Prabhas

therefore quite badly disturbed (Figure 26). The

type bowls, pots, dishes and dish-on-stand (Figures

relatively undisturbed part shows remains of rubble

16 - 21). The dish-on-stand resembles Rangpur Period

stone structures on the surface. At some places,

- 42 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 15 Shapur-I, General view of the Prabhas site at Shapur

patches of stone robbing are also visible on the surface.

of Sanavav Pa site within the forestland that has been

The site appears to have a total deposit of 1 to 1.5 m.

recently developed by acacia plantation. Since there is

The artefacts collected from the site incorporated

a good growth of vegetation all around, it is difficult

Prabhas type pottery and pottery resembling the

to locate actual features of the site. However it is

Rangpur IIB and IIC type. There are Prabhas bowls,

possible to trace remains of extensive rubble stone

Prabhas pots, Harappan dish-on-stand, basins, dishes

structures exposed on the surface within this area.

and stud handled bowls (Figures 27 - 30). This site

Remains of about 65 to 70 cm thick rubble stonewalls

appears to be a better preserved Prabhas site than the

of a large structure measuring 60 x 20 m are found

other small sites reported in the current exploration.

exposed in one of the areas where the vegetation cover

is relatively less. The other end of this structure has

10. Lohij-II (Sanavav Pa-II)

been partially dug up exposing an ashy underlying

(2109.49N, 703.50E)

deposit. A few Medieval potteries have been found

A small 10 x 10 m area with isolated sherds of the

from this cutting indicating perhaps the structures

same type of pottery is found about 60 m east of the

also may belong to that period.

above site. This could be an extension of Lohij-I or a

pastoral site formed slightly later in time.

potteries were reported from the site by Ms. Kalini

However, a few Early Harappan Sindh related

Khandwal in 2001 (Personal communication)

11. Lohij-III (Devariya Nagari)

while she was exploring the site at a time the Forest

(2109.66N; 70 03.1E)

Department was carrying out some digging at the

The site is located about half a kilometre further west

site. From one of the cuttings she collected the

- 43 -

P. Ajithprasad et al.

Figure 16 Prabhas bowls and dishes from Shapur-I

Figure 17 Prabhas pottery from Shapur-I

- 44 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 18 Prabhas pottery from Shapur-I:

1-6, 8 Prabhas bowls with pinkish brown paintings; 7, 11 Sorath Harappan (2 black core, buff slip); 9, 10 Prabhas basin and dish

Early Harappan pottery (Figure 31). Now, with the

Sindh related pottery at Datrana in north Gujarat

discovery of a few more shreds of similar pottery

(Ajithprasad 2002).

from the coastal site at Warodra-I, there appears

to have some evidence to suggest the contact of

coast, the entire landscape is predominated by high

the Early Harappans in the Saurashtra coast dating

coastal sand dunes that bury even a medieval burial

back to the first half of the third millennium BCE.

ground in the village. The sand deposit here is at

This argument is supported by the presence of both

present 4m high. Such a thick deposition of sand

the Pre-Prabhas pottery and the Early Harappan

indirectly suggests involvement of a major Tsunami

- 45 -

About half a kilometre further west along the

P. Ajithprasad et al.

Figure 19 Prabhas and Sorath Harappan dishes and dish-on-stand from Shapur-I:
1, 2, 5, 6 Prabhas ware; 3,4 Sorath Harappan
(5, 6 light greyish brown slip and dark brown painting; 5 painting on a white background).

type single high-energy event affecting the coastal

and it appears the site has about 1 to 1.25 m thick

area.

cultural deposit. The artefacts collected from this site

included Prabhas type convex sided bowls, Harappan

12. Kokru

pots, basins, dish on stand and dishes, etc (Figure 33).

(2110.81N, 7002.7E)

A small broken copper knife also is found from the

The site is located very close to the Khari stream

site (Figure 34).

(Figure 32) at about 1 km south of the village Sil and

about 1.50 km northeast from the seashore. It is badly

13. Khaganath ni timbo

disturbed by recent intense agriculture activities

(2113.14N, 7000.33E)

that have brought out on the surface a few rubble

It is about 1 km southeast of the Atroli village in

stone slabs from some of the buried structures. The

between the Khaganath temple and the Hanuman

artefacts at the site are found spread in a rectangular

temple. On the west, the site is protected by rocky

field over an area measuring 200 x 75 m (1.5 hectare)

ridges of very low height, that act as an effective

- 46 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 20 Prabhas and Sorath Harappan pottery from Shapur-I:

1-9 Red ware (1,7,8 dark brown slip)

windshield. Besides, it is located close to a shallow

farming activity. The rubbles that are exposed due to

natural depression, which at present retains water till

ploughing are collected by the farmers for building

the beginning of summer. The artefacts on the surface

compound walls of low height around their fields.

at the site are found spread over an area of 135 x 120 m

(Figure 35). In fact, the site is now cut up into small,

are of the Prabhas along with a few Sorath Harappan

cultivated fields and is also dissected by a narrow

pottery, which are badly abraded and often the slip

pathway. The site has good collection of Prabhas

completely pealed off the surface. The vessels include

pottery along with the Sorath Harappan. It also has

dishes, dish on stand, large storage jars, pots, Prabhas

a patch of Medieval artefacts on the top, especially at

bowls and Prabhas type pots (Figures 36 and 37). In

the highest part of the site. Wherever the medieval

addition to these there are a few stud-handled bowls

deposit is disturbed and partially removed it is easy to

as well as convex sided hemispherical bowls in the

locate the Prabhas pottery mixed with the Medieval

collection.

The artefacts collected from the site primarily

remains. The overall thickness of the Prabhas deposit

at the site may be around 1.5 to 2.0 m.

14. Madhavrai no Timbo

(2115.32N; 6958.79E)

The site appears to have several stone structures

belonging to the Prabhas occupation. These

The site is a cultivated field just behind the Madhavrai

structures are disturbed partially at present due to

temple in the heart of the Mul-Madhavpur village

- 47 -

P. Ajithprasad et al.

Figure 21 Prabhas and Sorath Harappan large size vessels from Shapur-I:
1, 5 Prabhas; 2-4 Sorath Harappan

Historic Period Sites

and is very close to the coastal State Highway. A

few Chalcolithic / Harappan potsherds are found
in a cultivated field at the edge of the present village

15. Kalej-1 ( Jiva Bhai no Khetar)

settlement, which in fact, is recently cut to expand the

(2053.58N; 7029.95E)

field. It is in this recently cut part of the settlement

The site is located on the left bank of the Saraswati

the Harappan pottery are found. As the entire village

river and hardly 1 km south of the Vikrameshwar

sits over the ancient mound, it is difficult to assess the

Temple on the river bank. The large mound is under

actual size of the original Chalcolithic occupation.

heavy cultivation and difficult to explore. A few Early

It appears the major part of the site is lying below

Historic pottery including sherds of the RPW, fine

the present day settlement at the site. The potteries

red ware and grey ware are collected from the edge of

collected are mostly non-descriptive sherds and are

this site. The site measures approximately 120 x 95 m

abraded due to continuous cultivations in the field.

and has a cultural deposit of about 1.0 to 1.50 m.

- 48 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 22 Early Harappan Sindh type pottery from Shapur-I

Figure 23 The Early Harappan Sindh type pottery from Shapur-I:

1-6 fine red ware (2 light greyish brown slip, 5 and 6 dark reddish brown slip)

- 49 -

P. Ajithprasad et al.

Figure 24 Shell bangle piece and pottery discs from Shapur-I

Graffiti on the right bottom sherd

Figure 25 Animal bones from Shapur-I

- 50 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 26 Lohij-I, General view of the Prabhas site in Lohij village

Figure 27 Prabhas pottery from Lohij-I

- 51 -

P. Ajithprasad et al.

Figure 28 Prabhas and Sorath Harappan pottery from Lohij-I

16. Kalej-II (Panch Pir)

in addition to a possible perimeter wall that protected

(2053.45N; 7030.1E)

it from the encroachment of the two rivers. Remains

This site is located about 300 m further south of

of a temple including the image of a Nandi and a few

Kalej-I. The artefacts mostly belonging to the Early

sculptures are also found at the site.

Historic period is spread to an overall 120 x 70 m

area. The maximum cultural deposit is approximately

predominated by the RPW. This includes several

1 to 1.25 m from the sections exposed in the field due

distinct shapes of the RPW pots, bowls, basins and

to agriculture activity. Red Polished Ware (RPW)

sprinklers (Figures 38). There is also a torpedo jar

and associated pottery are the main artefacts collected

sherd with black resin stained interior (Figure 41) and

from the site.

other associated red ware and grey ware of the Early

The pottery found on the surface at the site is

Historic period in the collection. A few decorated

17. Thumliu no Timbo

shell bangles and typical areca-nut shaped terracotta

(2054.4N; 7031.6E)

beads are also found at the site (Figure 42).

This is located in Ghorakhmadi village, just on the

road from Ajota to Ghorakhmadi at the meeting

18. Humbliu Timbo

point of the Saraswati and the Bhundio streams. The

(2054.07N; 7031.71E)

site measuring approximately 2 hectare (180 x 120 m)

This Early Historic site is located on the opposite

is partly disturbed due to cultivation. It has an overall

of Thumlio ni Timbo across the river Saraswati on

deposit above 1.5 m. The site appears to have several

it left bank in the Umbri village. The site in fact is

rubble stone structures buried underneath the deposit

a large settlement which is spread along the bank of

- 52 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 29 Prabhas and Sorath Harappan potteries from Lohij-I:

1-5 Prabhas bowls, 6-9 Sorath Harappan bowls
(9 Buff slip and reddish brown paintings, 10-14 Sorath Harappan dishes and dish-on-stand

the Saraswati up to almost 700 m, though the main

artefacts found on the surface are predominated by

concentration is found within an area covered by 300

the Early Historic RPW and associated coarse red

x 90 m. The site has a culture deposit close to 2.0 m

ware and grey ware (Figure 39). A stone rotary quern

thick and has buried rubble stone structures. The

is also found at the site.

- 53 -

P. Ajithprasad et al.

Figure 30 Prabhas and Sorath Harappan Potteries from Lohij-I:

1 Red ware with bichrome dark reddish brown and red slip with a faint cream patch in between,
2-3 Grey ware with dark reddish brown slip at the rim; 4-9, 11-12 Prabhas ware
(4-7, 9 a light buff/cream wash of slip, dark brown painted bands at the rim); 10 Soarth Harappan red ware

- 54 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 31 Early Harappan Sindh type pottery from Lohij (reported in 2001 by Kalini Kandwala)

19. Gelki

2.00 m. As the top part of the mound is levelled for

(2053.61N; 7031.05E )

cultivation the entire mound is quite ashy and loose.

The Gelki mound is located about 2 km southwest

Artefacts collected from the site mainly belong to

of Umbri village and is close to Sutrapada in the

Late Medieval period. However, a few sherds of the

southeast. This is a large ashy mound belonging to

RPW (Figure 40) are also found at the site indicating

the Late Medieval period. However, a few sherds

possible Early Historic occupation at the site.

of the RPW are also found at the site suggesting

there is some evidence of early occupation prior to

21. Vanankio Timbo (Warodra- III)

the Medieval period. This is now almost completely

(2048.84N; 70 32.04E)

concealed underneath the Medieval deposit. The site

This Early Historic period site is located 1.5 km

measures 200 x 200 m and has an habitation deposit

southeast of the village Warodra on the seashore. It is

close to 2.0 m.

a large site measuring approximately 4 hectare (200 x

200 m) and has 1.5 to 2.0 m thick maximum deposits.

20. Vakodar

The site is disturbed badly by cultivation. A good

(2055.36N; 7032.4E)

collection of RPW, red ware and grey ware are the

This Medieval site in Ghorakhmadi village is located

main pottery reported from the site. Shell bangles,

about 2 km north of the village on the right bank

terracotta discs/objects and animal bones are the

of the Bhundio stream that joins the Saraswati near

other artefacts and animal remains collected from the

Umbri. The site is reasonably large measuring 300

site.

x 150 m having an overall cultural deposit of about

- 55 -

P. Ajithprasad et al.

Figure 32 The Kokru mound across the Khari stream where it joins the sea

Figure 33 Prabhas pottery from Kokru

- 56 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 34 A broken copper/bronze knife blade from Kokru

Figure 35 General view of Khaganath Timbo, the Prabhas site in Atroli

- 57 -

P. Ajithprasad et al.

Figure 36 Prabhas and Sorath Harappan pottery from Khaganath Timbo

Figure 37 Prabhas pottery from Khaganath Timbo

- 58 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 38 RPW from Thumliu Timbo, Ghorakhmadi

Figure 39 Early Historic RPW and associated pottery from Humbliu Timbo, Ghorakhmadi

- 59 -

P. Ajithprasad et al.

Figure 40 Early Historic pottery, Red Polished ware from Vakodar and Ganeswari ni Timbo

Figure 41 Torpedo jar sherd from Thumliu no timbo, Ghorakhmadi

- 60 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Figure 42 Shell bangles, TC beads and perforated disc, Early Historic from Thumliu no Timbo, Ghorakhmadi

Figure 43 Historic period, Terracotta animal figurine from Bhavani Timbo, Dhamlej

- 61 -

P. Ajithprasad et al.

Figure 44 RPW sprinkler with incised decoration from Ganeswari ni Timbo

22. Gayatrimata ni Thumdo-I

Artefacts at the site are found spread over an area

(2049.3N; 7033.2E)

roughly measuring 200 x 200 m. The site appears

This site is located on the way to Gayatri Temple

to have a cultural deposit of 1.0 to 1.5 m lying over

in the village Pansawara. The site in fact is a large

the basalt bedrock, which is found exposed on the

mound of low elevation which today is under

surface in the vicinity of the site. The artefacts are

cultivation. As the entire site has good growth of

predominantly of the Early Historic period, including

standing cotton crops the actual extend of the site

the RPW and associated wares (Figures 40 and 44)

could not be ascertained from the surface spread. It

and a large rotary quern and a few shell bangles. A

appears that the artefacts are spread to an overall area

few remains of the Late Medieval period pottery are

of about 100 x 75 m. The pottery collected from this

also found on the surface indicating a two period

site is predominantly of Medieval period. However, a

occupation at the site.

few sherds of RPW and associated red and grey ware

are also found at the site. These may suggest an Early

24. Dhoro Timbo

Historic habitation at the site.

(2056.72N; 7008.6E)
The site is located half a kilometre southeast of

23. Ganeswari ni Timbo

Shidogar village. It measures 200 x 150 m in overall

(2057.7N; 7017.13E )

size and has a habitation deposit close to 1.0 m. The

This is a reasonably large mound located about half

artefacts collected from the site primarily include

a kilometre north of the Adri village on the left bank

RPW and other associated Early Historic pottery and

of a small stream, the Sankotri, that drains the region.

a few sherds of the Late Medieval period.

- 62 -

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

25. Madsimar

sites reported in the exploration are located close to

(2101.4N; 7016.73E)

present seashore, within a distance of 10 km. Besides,

The site is located 200 m north of the Simar village

many of them are very close to drainage channels

on the left bank of the Magel river. This is a small

be it tributary streams as in the case of sites on the

site measuring approximately 65 x 30 m with a low

Saraswati, Noli, Megal etc., or else close to a natural

elevation. The site belongs to the Late Medieval

depression as in the case of the site at Khaganath

period as the pottery from the site indicates. Some

temple. At least three small sites are found just a few

religious structures of recent period are found at the

hundred metres away from the edge of the seashore.

site.

As one moved interior about 20 to 25 km away from

the coastal line, the landscape changed into rocky

26. Panchnath Mahadev

and forested. There are no Chalcolithic/Harappan

(2101.77N; 7012.88E)

sites in this rocky terrain. It seems the Harappan/

Located about 3 km southwest of Chorwad village,

Chalcolithic communities have avoided settling down

this site has a very small and isolated spread of pottery

in this rocky terrain primarily due to the absence of

that includes a few RPW sherds. There is also a small

arable land for their agricultural activities. The fertile

rock cut well in the middle of the ruins at the site.

alluvial soil close to the coast offered a more conducive

The thickness of habitation is less than 40 cm. This

environment for the Chalcolithic pastoralists and

appears to be a sort of pastoral campsite of the Early

they selected pockets along the coastal belt based on

Historic period.

the availability of water for their settlements. It will

not be difficult to visualize this scenario in operation

27. Sonesa ni Timbo

as early as the Pre-Prabhas occupation at Somnath

(2105.40N, 7012.84E)

dating back to 3000 BCE. Our exploration so far

The site is located about 2 km east of Gothana village

has not revealed sites having Pre-Prabhas pottery.

and is close to the Nali stream. The village road in

This does not necessarily suggest complete absence

fact cuts through the site. This is an ashy mound

of the Pre-Prabhas related assemblage in the region.

incorporating predominantly Medieval pottery and

It is possible that this early Chalcolithic deposit is

a few Early Historic vessels especially the RPW. The

not exposed on the surface due to the accumulation

site is approximately 150 x 100 m in size and has about

of later period deposits. This is the situation even at

1.5 m thick cultural deposits

Somnath where the surface artefacts do not reflect

the Pre-Prabhas assemblage at all. It is only in the
excavation the Pre-Prabhas pottery was identified in

Site Distribution and

Settlement Features

the earliest habitation layers. Therefore, it will not be

all together right to rule out the possibility of a PrePrabhas substratum in some of the newly discovered

Geographical features of Saurashtra, especially

sites in the region.

availability of water and soils suitable for cultivation

and vast pastoral lands conducive for stock-raising,

Early Harappan Sindh Related

played a decisive role in the selection of sites for

Assemblages

occupation both in the Chalcolithic and the Historic

This specifically refers to a cultural assemblage

periods. A cursory look at the overall pattern of

primarily reported from Moti Pipli and several

distribution of sites suggests that almost all the

other sites such as Nag wada, Santhli, Panchasar

- 63 -

P. Ajithprasad et al.

Table 2 Distribution of sites based on their approximate size in different cultural periods
Cultural Periods

>0 1
htr.

>12
htr.

>2 3
htr.

>3 4
htr.

>4 5
htr.

>5 6
htr.

>6 7
htr.

>8 9
htr.

Total # of Sites

Chalcolithic
Early Harappan (Sindh type)

c. 3000-2500 BCE
Pre-Prabhas

1?

18

c. 3000-2500 BCE
Prabhas

c. 2400-1900 BCE
Sorath Harappan
c. 2400-1700 BCE
Historic Period
Early Historic
c. 1st century ACE
Medieval

11

c. 1300 ACE

etc. in north Gujarat (Sonawane and Ajithprasad

Researcher, the M. S. University of Baroda, while

1994, Majumdar and Sonawane 1996-97, Majumdar

exploring the Saurashtra coast of Junagadh had

2001, Ajithprasad 2002). In fact, there are nine sites

reported Early Harappan Sindh related pottery

in north Gujarat, apart from Moti Pipli, which

(Figure 31) from Lohij-III, a site near Mangrol

incorporate this early assemblage. At Nagwada and

(Majumdar 2001). The site falls within a patch of

Santhli, the two excavated sites of this category, the

recently reclaimed forestland and is covered with

Early Harappan pottery showing close similarity with

good vegetation. As a result, the visibility of cultural

Sindh and Baluchistan region are found associated

relics at present on the surface at this site is limited.

with burials. The pottery is found as burial goods in

While Ms. Kandwala was exploring the region in

both inhumation burials and in symbolic pot-burials

2001, there was some digging activity going on at

at these sites. Interestingly, the early Harappan pottery

the site and she could collect a few Early Harappan

at Nagwada is confined to the burials alone and is not

pottery from the pits dug out for plantation purpose.

reported from the regular habitation deposit. Santhli

We however, could not find any Sindh related

on the other hand has a very thin habitation layer that

pottery at this site in our survey. Most of the isolated

incorporated the Early Harappan pottery. Our studies

potsherds we collected from the surface belonged to

have suggested that the Early Harappan pottery of the

the Medieval period. Despite the thick growth of

early third millennium BCE is generally spread along

vegetation it is possible to figure out the plan of a very

the eastern margins of the Little Rann of Kachchh

large multi-roomed stone structure with thick outer

in north Gujarat and adjoining regions of southern

walls. The antiquity of this structure could not be

eastern Kachchh.

ascertained from the surface finds.

A few years back, in 2001, Ms. Kalini Kandwala,

- 64 -

Yet another site that has the Early Harappan

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Sindh related pottery is Warodra-I, which is located

Prabhas Sites

a few metres away from the seashore. In fact, the site

Reported first in the mid 1950s from Prabhas

is represented by just a few potsherds exposed on the

Patan/Somnath in the Saurashtra coast, the Prabhas

sandy shore (Figures 4 and 5). The potsherds are

assemblage included a group of pottery very different

too fragmentary, because of the constant inundation

from the so called Late Harappan pottery with

during the high tides and we could collect only one

which it was found associated with (IAR 1957,

diagnostic sherd. This and most of the small un-

Subbarao 1958, Dhavalikar 1984). It is generally

diagnostic potsherds that we have collected from the

represented by hemispherical bowls with a thin

site looked very similar to the Early Harappan Sindh

tapering rim and medium size pots and jars with

related pottery reported from north Gujarat sites

a short rim and an incipient neck. The painted

(Figure 6). A third site that has incorporated a few

decorations on the vessels are also distinct from the

sherds of the Early Harappan Sindh related pottery

contemporary Harappan style. The Prabhas bowls are

(Figures 22 and 23) is Punja Hira no Khetar in Shapur

generally decorated with geometric patterns within a

village, situated close to the seacoast in between Lohij

panel at the rim. The colour used in the paintings is

and Warodra. The potsherds a few rims and body

either purple or violet. Painted decorations similar

and base parts of pots and dish-on-stand resemble

to the above are found at the shoulder of Prabhas

those reported from Moti-Pipli in north Gujarat

pots too. These features make it different from

(Sonawane and Ajithprasad 1994, Ajithprasad 2002,

the Harappan pottery. The Prabhas assemblage at

Majumdar and Sonawane 1996-97). At the Shapur

Somnath is associated with rubble stone structures

site, the Sindh related pottery is found along with

and showed the use of stone blades of locally available

the Prabhas pottery. A few potsherds, especially the

chert, chalcedony and jasper in addition to copper

nos.2 to 3 illustrated in Figure 30 from Lohij-I, are

implements and lapidary beads of semiprecious

different from the usual Prabhas pottery and are

stones, steatite and faience.

closer to the Early Harappan Sindh type pottery

(Figure 30). These discoveries from the sites along

other Harappan sites such as Lothal (Rao 1985:432)

the Saurashtra coast are particularly significant as

and Rojdi (IAR 1957-58) in the excavation. It has

the Pre-Prabhas pottery is reported from Datrana in

also been reported in the surface survey from seven

north Gujarat in association with the Sindh related

more sites in central Saurashtra. These discoveries

Early Harappan pottery (Ajithprasad 2002). This,

certainly suggests the interactive sphere of the

therefore, suggests that the cultural contact and the

Prabhhas cultural domain extending into the interior

movement of people between Saurashtra coast and

Saurashtra even up to north Gujarat. Most of the

north Gujarat and probably Sindh may have begun

newly located sites in the current survey, in fact, have

around 3000BCE itself in the Early Harappan times.

a mixture of the Prabhas pottery and the Sorath

The discovery therefore has far reaching consequence

Harappan pottery which certainly is an indication

in understanding the beginning and development of

that Prabhas assemblage had a reasonably large spread

Chalcolithic way of life in Gujarat. The earlier idea of

around Somnath, along the coast. Seven sites among

Urban phase Harappans introducing the farming and

the 13 newly discovered Chalcolithic/Harappan sites

pastoral way of life in Gujarat seems to be at peril in

in the region show the predominance of the Prabhas

the light of the new discoveries.

pottery. Both typical hemispherical bowls and

The Prabhas pottery has been reported from

medium size pots with incipient neck are found in all

these sites. Besides, many of them have characteristic
- 65 -

P. Ajithprasad et al.

panels of geometric patterns painted in light pinkish

or red colour on them. These are small sites, often

deposit and the quality of structural constructions

less than one hectare in size and approximately just

are often a measure of the importance of the site in a

a metre or a little more than a metre thick cultural

systemic perspective wherein all the sites in a region

deposit (Table 2). There however are three large sites

are functionally inter-related. Each site plays a specific

- Shapur-I, Khaganath Timbo in Sil and Sanavav Pa

role in the development and proper operation of

in Lohij measuring between 1.5 and 3.00 hectare

cultural systems. Variation in the economic activities

affording more elaborate rubble stone structures. If

of the society therefore is an expression of the varying

all these sites are contemporary, hat would suggest

environmental setting of the sites and different

a functionally integrated network of settlements

strategies adopted by the society for the exploitation

belonging to the Prabhas culture, which has hitherto

of resources available in the environment. Hence,

not been reported. Besides, the graded size of

the effect of geographical features and environmental

settlements within the network is an indication of the

factors are immense in the development of different

degree of evolution of this regional cultural tradition

type sites in a region. It is obvious that these factors

in the Saurashtra coast.

are crucial in the selection and in the development

The over all size and thickness of occupation

of sites in the Saurashtra coast too from the fourth

The Sorath Harappan Sites

millennium BCE onwards.

As has been mentioned earlier the Prabhas pottery at

most of the above sites is found in association with

both the Chalcolithic and the later, Historic period

the Sorath Harappan pottery that can be compared

may reveal that there is no site in the region that

well with the assemblage from Rangpur IIA and IIB

could be regarded as an important nodal site or sites

and Rojdi B. The Sorath Harappan pottery found at

wherein most of the cultural and economic activities

these sites mainly includes convex and straight sided

were centred around (see Table 1.1 and 1.2). This is

bowls, dish-on-stands and pots and jars. Apart from

true for both the Chalcolithic and the Historic period

these sites there are six more Chalcolithic sites where

sites in the region. Yet, as mentioned earlier, there

the Sorath Harappan pottery is not associated with

appears to have some amount of gradation among

the Prabhas pottery. Material relics from these sites

the sites at least in their size and structural remains

incorporate slightly later period artefacts, especially

(Table 2). For instance, the site at Prabhas Patan

those belonging to Rojdi- C and Rangpur- IIC, in

on the bank of Hiren river is the largest among the

addition to Rangpur- IIB. These sites also are small in

Chalcolithic and in the Historic period sites. The

size, in fact, even smaller than the sites having Prabhas

site in the Historic period is 9.00 hectares. Its overall

pottery. The sites do not actually suggest the presence

height is about 8 m from the ground level. It is not

of longstanding permanent structure in the form of

sure the Chalcolithic habitation at the site was as

remains of rubble stone or brick walls. They probably

widely spread as the Historic period was. It probably

had flimsy structures made of wattle and daub. The

may not be. Yet, the Chalcolithic occupation is

over all deposit at these sites is very thin and they

reasonably large and incorporates rubble stone

generally have few artefacts apart from pottery and

structures suggesting the importance of the site. The

skeletal remains of exploited fauna. These probably

site at Shapur probably was equally large during this

would be short-term pastoral campsites suggesting

period. The other site that shows similarity in size

that the sites were occupied periodically even during

and structural features is Lohij, further north-west

the Post-Urban phase.

of Shapur. The sites showing exclusively Late Sorath

- 66 -

A close look at the overall features of the sites of

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

Harappan features are much smaller in size ranging

(Majmudar 1960). That the Asokan rock inscription,

just 0.19 to 0.70 hectare.

followed by the inscriptions of the Kshatrapa and

The pattern is continued even in the Historic

the Gupta dynasties at Girnar hill, mention about

periods. The Nagro Timbo at Prabhas Patan once

the construction and periodic upkeep of a large

again is the largest site in the coast belonging to

reservoir by damming across the Suvarnarekha river

the Early Historic period marked by the prominent

to augment water supply, bespeak the importance

presence of the Red Polished Ware (RPW). The

that were attached to the settlements in this region.

uncanny resemblance of the RPW, both in technical

It is apparent that the coastal region of Saurashtra

perfection and vessel features, with Roman pottery,

must have played an important role in the Greco-

particularly the Samian pottery is unmistakable

Roman sea trade resulting into the establishment of

(Subbarao 1953). Dated from the first century ACE

economically sound settlements along the coast.

onwards, it shows wide distribution all along the

western coast in Gujarat. A recent survey lists 268 sites

the current survey belonged to the Early Historic

in Gujarat alone that has yielded the RPW (Patel and

period. Many of them are found to be very rich in

Rajesh 2007). In the current exploratory survey we

RPW and have substantial deposit. Three sites found

located/revisited seventeen Early Historic period sites

on either banks of the Saraswati river merit special

apart from Prabhas Patan. Six of these are found to

mention here.

Most of the Historic period sites discovered in

be reoccupied even during the Late Medieval period.

Many of the Early Historic sites are larger in size than

Cultural Chronology
and Summary

the Chalcolithic sites and have solid rubble stone and

mud-brick structures. The remains of some sort of
a perimeter wall is suspected at Humblio timbo and
Thumlio ni timbo in Umbrio village. There are four

The beginning of Chalcolithic food producing way

sites having an overall size falling between 3 and 4

of life in Saurashtra has been dated to the middle of

hectare belonging to this period. They are very rich

fourth millennium BCE from Padri, about 100 km

in artefacts, especially the RPW. Besides, the site at

east of Prabhas Patan, in Bhavnagar district. This most

Thumlio ni timbo has a few sherds of the so called

probably was an early regional Chalcolithic tradition

Roman amphora that have recently been identified as

whose remains were reported from a few more sites

the torpedo jars of Mesopotamia (Tomber 2007).

in the region suggesting its geographical influence

In contrast, the number of Chalcolithic period sites

(Paul et al. 1997). However, the early Chalcolithic

falling in this size category is just one. This probably

Padri remains are not reported from Prabhas Patan

is a reflection of the increased intensity in the

in spite of its geographical proximity with Padri.

economic activities during the Early Historic period.

The ceramic assemblage and the lithic artefacts

The Saurashtra coast, especially along Junagadh and

reported from the earliest levels at Padri are distinctly

Bhavnagar districts, was a major centre for both

different from their counterpart at Prabhas Patan.

political and social activities right from the time of

Besides, the beginning of Pre-Prabhas, the earliest

Asoka Maurya in the third century BCE onwards.

Chalcolithic remains at Prabhas Patan, has been dated

The region also was a major Buddhist centre during

to 3000/2900 BCE; about half a millennium later

this period as was evident from a number of both

than the Padri. The Prabhas assemblage that follows

structural and rock-cut architectural remains of

Pre-Prabhas at Prabhas Patan has been dated by C14

the Early Historic period found along the coast

estimation to 2200 BCE (Calibrated, Possehl 1993).

- 67 -

P. Ajithprasad et al.

The Pre-Prabhas assemblage therefore can be dated

of Saurashtra. The sites are essentially found within

between 3000 and 2600 BCE.

the narrow strip of fertile littoral corridor running

Very little is known about the geographical

between the seashore and the rocky ridges of the

spread of the Pre-Prabhas assemblage in Saurashtra.

inland Saurashtra. The fertile soil in this narrow

However, a site showing the Pre-Prabhas pottery

corridor supported development of village farming

has been reported from Datrana in north Gujarat,

communities very early in the Chalcolithic period.

about 400 km north of Prabhas Patan (Ajithprasad

The economic development during this period appears

2004). The Pre-Prabhas pottery at the site is found

to be limited primarily due to the geographical

associated with the Early Harappan Sindh related

constraints of the region. The Early Historic period

ceramics of north Gujarat. This is significant, as it not

let loose a set of economic forces based on new

only establishes the interaction between Saurashtra

economic production incorporating enough inputs

and north Gujarat but also help dating the Early

from trade. The Saurashtra coast during this period

Harappan Sindh related assemblages reported from a

was one of the major centres of political development

number of sites in north Gujarat in the same bracket

in Gujarat. No wonder the sites of this period are

as that of Pre-Prabhas.

larger and richer than the preceding Chalcolithic

period. To some extent this was to continue even in

The Early Harappan Sindh related pottery in the

current exploration has been reported from three sites

the Medieval period.

Warodra-I, Lohij-III and Shapur in the Saurashtra

coast of Junagadh district. Among these three sites,

Mesolithic sites was found in the region that we

the Early Harappan pottery is found independently

have explored so far. Could we have missed them

not in association with other regional cultural

while looking for Chalcolithic and Historic period

remains at Warodra-I and Lohij-III. It is found

sites? There is always an outside chance of such an

along with Prabhas pottery in the third site at Shapur.

eventuality. However, the rigorous and systematic

However, the site at Shapur is badly disturbed by the

nature of our survey strategy makes that possibility

recent construction of a boundary wall cutting across

quite remote. If at all there were a few sites they

the site through the middle. The construction works

are either completely buried under later habitation

may have brought up some of the Early Harappan

or obliterated by later habitation activities. This

remains from the lower levels to the surface and

is very intriguing as we do come across Mesolithic

thereby mixing it with the later Prabhas assemblage.

sites in several parts of Saurashtra in var ying

These discoveries firmly establish the presence of

ecological contexts. It is possible that the Mesolithic

Early Harappan sites along the Saurashtra coast prior

communities preferred forested rocky terrain of the

to the beginning of Prabhas tradition. The Prabhas

interior Saurashtra that better suited to their hunting

therefore appears to be a cultural tradition developed

and gathering strategies than the coastal plains.

parallel to the Sorath Harappan along the Saurashtra

That does not rule out different levels of interaction

coast drawing inspiration from the Early Harappan

between the Chalcolithic and Mesolithic sites in the

and the Pre-Prabhas. The Sorath Harappan on the

region. It has been shown to a reasonable degree of

other hand is rooted more in the interior Saurashtra.

confidence that such interactions were extremely

To summarize, the current survey has brought

important in the development of cultural patterns,

to light the importance of systematic and focussed

both in the Mesolithic and Harappan Chalcolithic

survey in locating both the Harappan Chalcolithic

settlements in Gujarat (Possehl 1980, Ajithprasad

and the Historic period sites in the Junagadh coast

2004). The presence of copper and a few crude

- 68 -

It is surprising that no microlith yielding

Archaeological explorations in the Saurashtra coast of Junagadh District, Gujarat

potsherds in the Mesolithic context at Langhnaj in

of the Indus Civilization. Books and Books, New Delhi.

north Gujarat (Sankalia 1965) and the presence of

pp.243-251.
Dhavalikar, M.K. and G.L. Possehl (1992) The Pre-Harappan

domesticated cattle and sheep/goat skeletal remains

Period at Prabhas Patan and the Pre-Harappan Phase in

from a few other Mesolithic sites (Bhan 1994) are

Gujarat. Man and Environment 17(1): 71-78.

a sure indication of the above interaction. Apart

Dhavalikar, M.K., M.R. Raval, and Y.M. Chitalwala (1996)

from these, interaction between the Chalcolithic

Kuntasi: A Harappan Emporium on West Coast. Deccan

settlements of the coastal region and the Sorath

College Post Graduate and Research Institute, Pune.

Harappan and Harappan settlements of interior

Hegde, K.T.M. and V.H. Sonawane (1986) Landscape and

Saurashtra were certainly responsible for the over all

settlement Pattern of the Harappan Culture Villages in

development of Chalcolithic way of life in Saurashtra.

the Rupen Estuary. Man and Environment 10: 23-31.

Two important results of the current survey are

IA R 1955 to 1977: Indian Archaeolog y - A Review.

the identification of six new sites showing the Prabhas

Archaeological Survey of India, New Delhi. IAR 1955-56:

pottery and three sites having the Early Harappan

7-8; IAR 1956-57: 16-17; IAR 1957-58: 18-23; IAR 1958-

Sindh related pottery along the coast. The former

59:19-21; IAR 1971-72: 12-13; IAR 1975-76: 13; IAR 197677: 17-18.

discovery provides much needed data for tracing the

Majmudar, M.R. (1960) Historical and Cultural Chronology

geographical spread/extension of Prabhas sites in the

of Gujarat. The M.S. University Oriental Institute

Saurashtra coast. This would help define the regional

Publication, Baroda.

character of this Chalcolithic tradition and its role

Majumdar, A. (2001) Emergence of Early Harappans in North

in the development of Chalcolithic settlements in

Gujarat. Man and Environment 26(1): 23-38.

Saurashtra. The discovery of Early Harappan sites

Majumdar, A and V.H. Sonawane (1996-97) Pre-Harappan

in the Saurashtra coast is even more significant as it

Burial Pottery from Moti-Pipli: A New Dimension in

shows the Early Harappan influence from the Indus

the Cultural Assemblage of North Gujarat. Prgdhar 7

valley extending up to the Saurashtra coast much

(1996-97): 11-17.
Merh, S.S. (1995) Geology of Gujarat. Geological Society of

earlier than the beginning of the Mature Harappan.

India, Bangalore. p.4.

Nanavati, J.M., R.N. Mehta and S.N. Chaudhary (1971)
Somnath - 1956. Department of Archaeology, Gujarat

References

State and M.S. University of Baroda, M.S. University

Ajithprasad. P. (2002) The Pre-Harappan Cultures of

Press, Baroda. pp.18-41.

Gujarat, in S. Settar and Ravi Korisettar (eds.) Indian

Pandya, P.P. (1958) In Indian Archaeology 1957-58 A Review

Archaeology in Retrospect vol.II Protohistory: Archaeology

(IAR 1957-58) Department of Archaeology, Government

of the Harappan Civilization. ICHR and Manohar, New

of India, New Delhi: 18-23.

Patel, A. and S.V. Rajesh (2007) Red Polished Ware (RPW) in

Delhi. pp. 129-158

Gujarat, Western India An Archaeological Perspective.

Ajithprasad. P. (2004) Holocene Adaptation of the

Prgdhar 17 (2006-07): 89-111.

Mesolithic and Chalcolithic Settlements in North

Gujarat, in Y. Yasuda and Vasant Shinde (Eds.) Monsoon

Paul, A .B., B. Tripathy and V. Shinde (1997) New

and Civilizations. Roli Books Pvt. Ltd., New Delhi.

Archaeological Discoveries in the Lower Shetrunji River

pp.115-132.

Basin, Bhavnagar District, Gujarat, in B.M. Khanduri

Bhan, K.K. (1994) Cultural Development of the Prehistoric

and Vinod Nauriyal (eds.) Him-Kanti: Archaeology, Art

Period in North Gujarat with Reference to Western

and History. Book India Publishing co., Delhi. pp.53-63.

Possehl, G.L. (1980) Indus Civilization in Saurashtra. B.R.

India. South Asian Studies 10: 71-90.

Publishing Corporation, New Delhi.

Dhavalikar, M.K. (1984) Sub-Indus Cultures of Central and

Possehl, G.L. (1993) Radiometric Dates from South East Asia

Western India, in B.B. Lal and S.P. Gupta (eds.) Frontiers

- 69 -

P. Ajithprasad et al.

(Unpublished list, Compiled by G. L.Possehl).

Shinde, V. and S.B. Kar (1992) Padri Ware: a New Painted

Possehl, G.L. (1997) The Transformation of the Indus

Ceramic Found in the Harappan Levels at Padri in

Civilization. Journal of World Prehistory 11(4): 425-71.

Gujarat. Man and Environment 17(2): 105-110.

Rajesh, S.V. and A. Patel (2007) A Gazetteer of Pre- and

Sonawane, V.H. (2000) Early Farming Communities of

Protohistoric Sites in Gujarat. Man and Environment

Gujarat, India, in P. Bellwood, D. Bowdery, D. Bulbeck,

32(2):61-136.

D. Bear, V. Shinde, R. Shutter, G. Summerhayes (eds.)

Rao, S.R. (1958) In Indian Archaeology 1957-58 A Review

Bulletin of the Indo-Pacific Prehistoric Association 19: IndoPacific Prehistory: Melaka Papers 3: 137-146.

(IAR 1957-58) Department of Archaeology, Government

of India, New Delhi: 13-17

Sonawane, V. H. and P. Ajithprasad (1994) Harappa Culture

and Gujarat. Man and Environment 20(1-2): 37-49.

Rao, S.R . (1963) Excavations at Rang pur and Other

Explorations in Gujarat. Ancient India 18-19: 5-207.

Subbarao, B. (1953) Baroda Through the Ages. The M.S.

Rao, S.R. (1985) Lothal: A Port Town (1955 - 62), vol.II.

University Archaeolog y Series 1. Maharaja Sayajirao

Memoirs of the Archaeological Survey of India no. 78.

University of Baroda, Baroda. pp.58-64.

Subbarao, B. (1958) The Personality of India. The M. S.

Archaeological Survey of India, New Delhi.

Sankalia, H.D. (1965) Excavations at Langhnaj: 1944-63. Part

University Archaeology Series 3. The Maharaja Sayajirao

I: Archaeology. Poona. (Building Centenary and Silver

University of Baroda, Baroda. p.132.

Jubilee Series 51). pp.41-64.

Tomber, R. (2007) Rome and Mesopotamia Importers into

India in the First Millennium AD. Antiquity 81: 972-988.

Shinde, V. (1992) Excavations at Padri - 1990-91: A

Preliminary Report. Man and Environment 17(1): 79-86.

- 70 -

Annual report of excavation at Kanmer 2007-08 and 2008-09

Annual report of excavation at Kanmer

2007-08 and 2008-09
J.S. Kharakwal
Department of Archaeology, Rajasthan Vidyapeeth, Udaipur

Y.S. Rawat
State Department of Archaeology, Gujarat, Gandhinagar

Toshiki Osada
Research Institute for Humanity and Nature, Kyoto

Abstract
Kanmer, a multicultural site, is located in Kachchh district of Gujarat state of India. Our controlled excavations have yielded four
fold cultural sequence at the site. During the Bronze Age it was a small Harappan settlement, which was protected by a very strong
fort wall. Among the interesting discoveries are Harappan objects like a few seal impressions, drill bits and unoxidised faience
beads, and historic materials such as Roman amphorae and Torpedo ware type pottery. Kanmer played a very important role in
trade or commercial activities right from the Harappan period.

INTRODUCTION

site is composed of Jurassic piedmont having current

bedded ferruginous sandstone and variety of shale. An

The Harappan site of Kanmer (2323N; 7052E) is

iselberg next to the site which is identical to an upside

situated close to the Little Rann of Kachchh in Rapar

down V-shaped hillock marks the site from a distance.

taluka in the Kachchh district of Gujarat (Figure 1).

Locally known as Bakarkot the site is situated on the

OBJECTIVES

right bank of a seasonal rivulet Aludawaro Vokro

which passes to the north of the present village of
Kanmer. The ancient mound was discovered by R.S.

The ancient mound at Kanmer is squarish measuring

Bisht and his colleagues of Archaeological Survey of

115 m from east to west by 105 m from north to south

India way back in the eighties of the last century (IAR

on plan (Figure 2) with a height of about 8 m from

1985-86: 15-19).

the ground level (i.e. 20 m above the mean sea level).

The peninsula of Kachchh is largely composed

It was chosen for excavation to understand the role

of marine and estuarine sediments and has a variety

of small fortified settlements, distinct features of

of rocks ranging from middle Jurassic to recent

architecture, regional variation in ceramic assemblage,

times. The area is marked by low lying bare rocky

de-urbanization process and to comprehend the

hills, sandy plains and vast tracts of naked tidal mud

Kachchh Harappan, besides to understand the role

flats, in the north-western and eastern parts, these

of contemporary rural cultures in the urbanization

mud flats are known as Great Rann and Little Rann

process. Till recently the Harappan archaeology has

respectively. The entire region is drained by seasonal

been dominated by the excavation of large sites, and

rivulets. Of course it is one of the poorest areas as far

as a result the role of smaller settlements like Kanmer

as rain fall is concerned. The landscape around the

has not been fully understood.

- 71 -

J.S. Kharakwal et al.

Figure 1 Location map of Kanmer

With the above cited objectives, controlled small

Period III: KMR III (Late Harappan)

scale excavations at Kanmer were conducted for two

Period IV: KMR IV (Early Historic)

field seasons 2005-06 and 2006-07, which allowed us

Period V: KMR V (Medieval)

In our earlier publications we have termed the

to identify the following fivefold cultural sequence

(Kharakwal et al. 2005, 2007, 2008, 2009):

cultural sequence as Pd I Early Harappan, Pd II

Period I:

KMR I (Early Harappan)

Mature Harappan and Pd III as Late Harappan. These

Period II:

KMR II (Mature Harappan)

cultural periods correspond to KMR I, KMR II and

- 72 -

Annual report of excavation at Kanmer 2007-08 and 2008-09

Figure 2 Contour plan of Kanmer

KMR III respectively. KMR stands for Kanmer.

4) To train young students of archaeology in field

This new terminology is preferred because the total

techniques.

Harappan deposit represents different stages of

culture. Of course we have local material including

bichrome potter y in KMR I, which gradually

India, Japan and other countries participated in the

disappear in the upper levels. KMR IV represents the

excavation.

The following scholars and students from

Early Historic deposit whereas KMR V is a medieval

one.

Prof. S.N. Rajaguru, Drs. P.P. Joglekar, Shusma Deo,

Besides, the remains of a massive fort, a few

Shahida Ansari, Mr. Pankaj Goyal, Miss C.V.Sharda,

Harappan and Historic structures, medieval silos

Miss Yun Chong Suk, Mr. Tilok Thakuria, Mr.

were also partially exposed. However to understand

Santanu Vaidya, Mr. Shingo Kameyama (all from

the results of these two initial seasons the excavation

Deccan College, Pune), Dr. Anil Pokharia, Birbal

work continued in the third and fourth seasons with

Sahni Palaeobotany Institute, Lucknow,

the following objectives:

Profs. Takao Uno, Dr. Hirofumi Teramura, Dr.

Kondo, Prof. Hideo Sakai, Prof. Toru Kishida, Miss

1) To understand the nature and history of the

Asuka Kanto and Dr. H. Chiba, Dr. Endo Hitoshi,

fortification wall.

Mr. Takashi Ito, Yukiko Takeuchi, Dr. Ainori Useugi

2) To understand the possible function of the site.

(all from RIHN Japan), Miss Carla Lancelotti,

3) To enquire if there is a lower town or any settlement

Cambridge, Miss Sarah Luddy, London.

out side the main mound

- 73 -

J.S. Kharakwal et al.

Mr. Lal Chand Patel, Mr. Hansmukh Seth, Mr.

understand nature of structural complexes that were

Rajesh Meena, Mr. Suresh Meena, Mr. Asif Hussain,

exposed partially during last season belonging to

Mr. Krishna Pal Singh, Mr. Bhim Raj Varhat,

periods KMR II and KMR III. In the north eastern

Mr.Sachin Dikshit, Mr. Digvijay Chauhan, Miss Rita

area trenches Z15, Z16, Z17 and AA16, AA17 and

Chauhan, Miss Rashmi Salvi, Mr. Rohit Menaria, Mr.

BB16, BB17 were selected for excavation mainly to

Hitesh Bunkar, Kalu and Ganesh (all from Institute

under stand the details of the inner face and corner of

of Rajasthan Studies, JRN Rajasthan Vidyapeeth,

the fortification.

Udaipur).
Mr. Na r a y a n Pa l i w a l a n d S a n j e e v K a t a r i a ,

STRATIGRAPHY

Banaskantha, Gujarat, (Draughtsmen)

Teachers and students from HNB Garhwal University,
Uttarakhand, M.S. University, Vadodara, Institute of

In the south central part layers one and two with

Heritage and Management, Delhi and Lok Vigyan

a combined thickness of about 70 cm were found

Kendra, Almora.

devoid of any cultural material, except for a few

rolled pot sherds. These layers were light greyish in
appearance and were found separated from each other

EXCAVATION

by a thin streak consisting of a few rolled pot sherds.

Layer three loose in nature was marked by a thin

In continuation of the previous seasons work it was

layer of ashy deposit ranging from 5 cm to 30 cm

decided to lay down more trenches in the south

in thickness. This deposit yielded medieval pottery

central, south eastern and north eastern parts of the

associated with a few flimsy structures which were

mound. In addition to these areas, to understand

partially exposed in the south eastern and southern

the constructional phases of the fort wall and the

area. One of partially exposed such structures oriented

deposit accumulated along its exterior a small area

roughly north-south (trenches AA34 and AA35), was

of the eastern face of the fort wall was exposed. To

found right raised on the southern arm of Harappan

have a better understanding of stratigraphy and

fortification. The western and eastern walls of this

layout of the settlement trenches X28, X29, X30,

house were measured 3.50 m each, indicating squarish

Y28, Y29, Z28, Z31, AA28, AA29, AA30, AA31, AA32

on plan. In the northern room of this house was found

and AA33 were laid in the south central area of the

a rotary quern. A circular stone structure (diameter

mound. All these trenches were in fact extension of

2.50 m) was found superimposed on this medieval

last seasons excavation in adjoining trenches Z29

structure. Besides, in the central area (in trenches

and Z30 and Y30. The horizontal excavation in these

Z29 and Z30) two large circular black burnt features

trenches allowed us to identify a pre- fortification

(diameter about 1 m) were noticed, which may have

deposit followed by construction of a fort and several

been used as fire places. In the south eastern area over

successive structural phases of the Harappan culture,

a dozen silos belonging to this phase were found,

besides a few potters kilns (belonging to the Early

which were dug into the Historic and the Harappan

Historic phase). In the south eastern area trenches

deposit.

such as FF29, FF30, EE29, EE30 and GG29 were laid

down to the west of earlier exposed area (represented

thick (average) deposit of dark greyish fine silt sticky

by trenches HH29, HH30 and II29, II30 partially

in appearance. It was identified as layer 4, which did

exposed). These trenches were excavated mainly to

not yield any cultural material. It was found deposited

- 74 -

These medieval remains were raised on a 50 cm

Annual report of excavation at Kanmer 2007-08 and 2008-09

in the central depression overlying the Historic

storage cannot be ruled out. These pits dug into the

remains. Therefore it was identified as a cultural break

Harappan deposit and structures underneath. It seems

between the Medieval and Historic periods.

that these pits were of no use during the later phase

The Early Historic remains were found in layer

of Historical settlements at the site as people started

5, which is about 1.50 m thick in the central part

throwing garbage in them. The regular dumping of

and gradually thinned down towards the peripheral

such garbage has been identified in the form of lenses

regions where it measured hardly a few centimetres in

in one such pit exposed in trench Z30 in the third

thickness. The Historical remains were identified on

season. Thus the total deposit of layer 5 was recorded

the basis of presence of terracotta seal impression with

about 1.55 m.

late Brahmi letters, Red Polished Ware, Red Ware

(Rang Mahal type) and a few potsherds of Roman and

many fragments of Roman amphorae and West

West Asian origin. This deposit was further divided

Asian Tarpedo Jar type pottery (Tomber 2007). The

into two sub layers levelled as 5a and 5b in the central

presence of these pottery types is good indicator of

part. Layer 5a was measured 55 cm in thickness and

long distance trade that was being operational from

yielded residential structures, which were partially

Kanmer. What commodities were traded from here is

exposed. It was composed of dark greyish soil generally

yet to be understood.

compact in nature. In northern part of trench Z29 and

AA29 a large number of yellow patches were noticed

central part of the fortified area, when the Historic

in layer 5a. According to the villagers, the yellowish

people arrived at the site. This depression may have

clay may have been brought from a distant source

been formed due to continuous structural or other

for pottery making. Interestingly, a huge dumping

activity of the Harappans in the peripheral area

of partially broken bowls with featureless rim and

towards the end of their occupation. This depression

tapering body was found in the north western corner

was covered by a very thick layer of dark grey soil

of trench Z29 right on a floor. It is not clear if it was

which was coarse in nature. The thickness of this

a potters house or storage of such pots. Layer 5b

dark grey layer gradually decreases as we approach

(thickness about 1 m) was largely ashy and appeared to

from central to the peripheral region. Considering its

be a formed due to regular dumping activity.

contour, it appears that it may have been actually more

In the central area of this ashy deposit more than

than 1.50 m thick in the central part whereas it was

half a dozen roughly circular burnt features (diameter

found only a few centimetre thick in the peripheral

about 3 m) were exposed. Excavation of two such

zones. Thus these historic remains were underlined by

features revealed that they were actually conical

a uniform deposit of dark grey soil, which was black

shaped pits with highly burnt base or beds. In these

in appearance and coarse in nature. Levelled as layer

pits was found large quantity of historical pottery.

(6) this deposit was found overlying almost all the

The purpose of such pits is not clear to us, though

Harappan structures. It was devoid of any direct or

they were identical to the modern potter's kiln. In

indirect human activity therefore has been identified

absence of vetrified, ill baked potsherds or potsherds

as desertion between the Early Historic and Harappan

with highly burnt surface or with black patches, we

deposits. The composition of this deposit indicates

are not sure about their identification as kilns. Of

that most of its contents were result of re-deposition

cource one can not miss a large quantity of ash spread

of the eroded material. From the south central part

in and around these pits. Possibilities of some of these

layer 6 was largely thrown away by the Historical

pits were used for grain storage or any other kind of

people to make their kilns (?). Therefore it survived

- 75 -

Besides historical pottery, layer 5b also yielded

There was a very large depression in the south

J.S. Kharakwal et al.

only in patches measuring in thickness about 35 cm.

thickness of this ashy deposit gradually increases

Some places such as in trenches AA34, EE29 and

towards the peripheral regions from central area. It

FF29 it was found about 80 cm thick.

indicates that perhaps the peripheral regions along the

fort wall remained occupied for longer period than

Layer 7 varies from light greyish to ashy in

appearance. Its thickness varies from 10 to 40 cm

the central part.

and was traceable only in trenches X28, X29 and

AA28, AA29 and AA30. It has been noticed that this

Harappan material identical to urban phase along

deposit also got disturbed due to large pits dug by

with Sorath Harappan and Anart elements. It yielded

the people of historic period for carrying out their

a large variety of Harappan objects and pottery,

various activities. As the result at places the historic

structures and floor levels including seals and seal

period material was found resting directly upon

impressions. In the lower part of this layer (in trenches

the Harappan deposit. However the undisturbed

AA28 and Z28) was found a lense of white clay (about

deposit of layer 7 yielded Harappan pottery, often

10 cm thick) which looked like redeposited washed off

ochreous in appearance, along with black and red

white plaster. While in the exposed section (section

(Ahar type?) and very Coarse (looks Gritty) Red

facing south) of trench Y29 a series of lenses of fine

Ware (Misra et al. 1997). The ceramic assemblage

clay, course sand and floors were noticed. Besides

is dominated by high neck jar and the Coarse Red

these, several lenses of burnt soil were also noticed.

Ware pots which is often brittle. The fine variety of

This deposit has been identified as KMR IIB.

Red Ware was of course treated with red slip which

has either peeled off or washed away and the pottery

material, though a few pieces of bichrome and other

gives ochreous colour while handling it. This kind of

local types appeared in the layers 10 and 11. The

pottery has concentration in south eastern part of the

combined thickness of these three layers was measured

mound. This pottery assemblage has been identified

about 2.50 m. These three layers have been identified

as belonging to KMR III. Neither goblet nor script

as part of cultural phase KMR II A.

material was discovered from this level.

Layer 8 is also ashy and loose in composition. Its

in nature. This deposit is more than 1.0 m in thick

thickness varies from 0.50 m to 2.60 m in the central

in the central part and appears to decrease gradually

part whereas in the peripheral area, particularly

towards western part. It is largely composed of

near the inner face of western fort wall, it measured

yellowish fine sandy clay, to which the local term

more than 3.0 m. This layer is largely formed due to

used by the villagers is pila lilva. It appears that the

dumping of large quantity of ash in the northern

Harappans used this clay as binding and plastering

and western area. Successive episodes of dumping

material in their stone houses. This layer has yielded,

have been identified in this deposit. In these dumps

in addition to the cultural material identical to the

evidences of local/ in situ burning have also been

urban Harappan phase, course pottery types similar to

noticed. On the basis of the cultural material

Anarta assemblage.

recovered from the central, NE, SE and western parts

as well as the nature of deposit, it seems that layer 8

loose and less sandy in comparison to layer 9. It has

of the central part corresponds to layers 5, 6, 7 and

small lenses of pinkish and whitish clay. These lenses

8 of western part (trench Q28), layers 4, 5 and 6 (so

have caused possibly due to mixing of powdered shale

far excavated in Z 16, 17, AA 17) of the northern part

with clay for some purpose (floor making ?). Besides

and layer 4 of south eastern area of the mound. The

urban phase cultural material, this layer also yielded

- 76 -

Layer 8 also yielded various residential structures,

Layers 9, 10 and 11 yielded identical cultural

Layer 9 is yellowish in appearance and compact

Layer 10 is also yellowish in appearance but it is

Annual report of excavation at Kanmer 2007-08 and 2008-09

a few pot sherds of bichrome and unslipped red ware

site like Kanmer is one of the most impressive and

with a band of corrugated lines on the shoulder and

significant discoveries at the site. The outer face of

body part of jar/ pots and bowls with perforated

the fortification was partially visible on the northern

knob.

as well as on the southern slopes of the mound. This

Layer 11 is largely composed of coarse sand,

fortification stratigraphically belongs to KMR IIA.

which appears to have been formed due to some water

Interestingly, the fortification of Kanmer was not

action. In the eastern and southern sections of AA28 it

oriented along the cardinal directions as has been

had greenish appearance perhaps indicating stagnancy

noticed at most of the Harappan sites. It is rather

of water. It is hardly 25 cm thick. The cultural material

oriented in NNW-SSE direction. Hence there is a

of this layer was also identical to layer 10, of course the

deviation of about 26 (off the north-south line).

quantity and bichrome, Cream slipped, Coarse Red

Therefore to understand its precise plan and growth

ware, unslipped red ware increased.

a few trenches were laid down at the south eastern,

Layer 12 is about 35 cm thick (average) and is

north eastern and north western corners of the

composed of light greyish soil. Its general appearance

mound. Excavation of the wall face, in both of these

is whitish. It has several lenses of fine clayey material

areas, revealed that the main wall was provided with

with yellowish, whitish and pinkish in appearance.

a revetting wall all along its outer face. Two corners,

These lenses indicate that perhaps the inhabitants

the north-eastern and the south eastern, were exposed

mixed locally available powdered shale with the

internally whereas the north eastern and north

different coloured soil for various day to day purposes.

western corner was exposed externally. In a cutting

This layer is resting right up on the ferruginous bed

(trenches Y15, Y16 and Y17) near the north-eastern

rock. It has yielded ceramic assemblage which appears

corner it was realized that the core of the wall was

to be dominated by localise pottery types such as

actually made of mud bricks (mostly brown sandy

bichrome, Cream slipped, Red and chocolate slipped,

clay) and it was veneered on either side by large semi-

un-slipped types and Coarse Red ware. Harappan

dressed blocks of lime stone and sand stone procured

material in small quantity is also available. This kind

from the nearby hillocks. Some of the blocks placed

of material was also recovered from the deposit

in the lower part of the wall measured about 2 m in

comprising layers 14 and 15 of trench Q28 in a trench

length and 50 cm in breadth and 40 cm in thickness

laid along the inner face of the western fort wall and

(Figure 3). The outer faces of the northern and eastern

belong to the pre- fortification phase at the site. On

side along with the later additions were exposed up

the basis of its stratigraphical position and the type of

to a length of about 35 m and 36 m respectively. As

material it yielded this deposit has also been identified

many as 10 courses of eastern wall and 6 courses of

as cultural period KMR I.

the addition were exposed in order to understand if

Layer 10, 11 and 12 were exposed in a small area in

the construction pattern matches with others, besides

trenches AA28 and Z28 in the central part and Q28

deposit outside. The eastern wall was exposed up to

in the western part there fore the material recovered

a depth of 3.20 m and its construction pattern was

from these layers is also small in quantity.

found identical to rest of the walls.

A few small cuttings put on the northern and

southern slopes proved that the addition to the wall

FORT WALL

was actually a supportive wall (revetting) provided

to the main fortification. This support was found

Presence of a strong fortification around a small

running all along the exterior of the fortification. The

- 77 -

J.S. Kharakwal et al.

Figure 3 Stones places in the northwestern corner of the fort wall

Figure 4 Inner face of the fort wall, western arm

- 78 -

Annual report of excavation at Kanmer 2007-08 and 2008-09

construction was such that no visible gap between the

to the exterior of wall on all three sides appear to be a

revetting and the wall on the eastern and southern side

protective measure taken up after the calamity.

was noticed but in the northern side a narrow gap was

noticed which widened gradually from east to west.

just below about 2.20 m from the highest point has

At places (near the north western corner) the gap

been measured about 18 m. As the wall had an upward

was found more than 1m wide which was found filled

tapering, it is likely that the width at the base would

with earth and rubble. Interestingly the main wall and

measure more than 20 m. Compared to its northern

the revetting both have been found standing on the

and eastern counterparts the western wall is thinner as

same bed rock. This kind of situation leads to think a

its width measured only 15 m at a depth of 3.70 m from

slight change in orientation or planning at a later date

the highest point. On plan, two arms of the fort at

too! The exterior of the western wall does not reveal

the south eastern corner form an obtuse angle (about

any revetting like the other three walls. The height of

100) while another two arms at north eastern corner

western wall is not preserved much. Comparatively

form an acute angle. Thus these corner angles suggest

its extant height is about 3 m lower than its eastern

that on plan the fortification could have formed a

counterpart. It seems it has been robbed by the local

parallelogram. During the precise documentation

villagers for building graves as a grave yard is located

of the fortification the external length of eastern,

on the western and south- western slope of the

western, northern and southern arm was measured 116

mound.

m, 109.50 m, 107 m and 113.50 m respectively. Hence,

To understand the history of the defense wall, the

the average east west length was about 110 m whereas

inner face of western wall was exposed (trenches P27,

north south was about 115 m. Similarly, the inner

P28, Q27, Q28) near a modern shrine of Hanuman

length of these four walls was also found different

which is located on the extant top of the wall. In Q28

from each other. The inner length of northern,

the inner face of the fort wall was found overlying the

southern, eastern and western arm was measured

earliest cultural deposit, which has been identified

72.50 m, 79.50 m, 78.50 m and 74.50 m respectively.

as KMR I or pre-fortification deposit. In this area

Therefore the average east-west inner length measured

the inner face of the wall has survived up to a height

76.50 m and north-south inner length measured 76 m.

of 5.70 m with 29 stone courses. In the construction

of the stone veneer the Harappans had used the

dimensions and the total area covered by the

yellowish clayey sand (locally known as pila lilva) or

habitation including the fort wall was about 12,430

the powdered white shale (locally known as dhodi

sq m. Out of this total area about 6,616 sq m (more

mati) as binding material. The veneer on the either

than 50%) area was covered by the fortification itself.

side shows a battering as a result of which the breadth

Thus only about 5,814 sq m space was available inside

of the wall gradually decreases with increase of the

the fortification for residential or other activities. It is,

height. This inner face of the western wall shows

therefore, obviously a difficult question to address why

a considerable tilt towards west in its upper levels.

such a miniscule settlement was protected so well by

In fact, barring the 8 courses at the base, the entire

a strong fortification. Further excavation may provide

wall seems to be repaired in this part (trench Q28;

some clue to it.

The average width of northern and eastern walls

Thus, all the four arms were found with different

see Figure 3). It appears that some serious damage

TEST TRENCH N3

was caused to the fort wall perhaps soon after its

construction. Was it an earthquake (?), which brought

A few trenches were laid on the northern and

about this damage ? Perhaps the aforesaid additions

- 79 -

J.S. Kharakwal et al.

Figure 5 Outer face of the eastern fortification wall

TRENCH Q28

southern slopes to know the precise orientation of the

main wall and the extension of revetting (addition)
provided to it. These cuttings helped us to understand

A laser survey carried out by us indicated a gap in the

the nature of main wall and the revetment. In trench

western fort wall near the modern shrine of Hanuman

N3, located on the north western slope right on the

which is situated on the top of the wall. To examine

outer face of the fortification, the extant height of

the possibilities of an entrance at this point and also

the north side wall was measured 4m with as many

to reach the natural soil in this area a few trenches

as 15 semi-dressed stone courses which looks very

were laid here. These trenches numbered as P27, P28,

impressive. In this cutting revetment was found

Q27 and Q28 were laid across the wall, the extant top

sealed by layer 5.

of which was visible right on the surface.

Layer 6 represented the filling of about 1 m gap

In these trenches layer 1 compact in nature was

between the outer arm and the addition. This gap was

found composed of dark greyish soil. It was about 10

measured about 1.10 m at the base whereas about 90

cm thick and yielded mixed pottery comprising of

cm at the top surviving course of addition, indicating

both Historic and Medieval vintage. Layer 2, light

that it was gradually decreasing towards top. It was

greyish to pale yellowish in colour, was slightly loose

found resting on a 30 cm thick soil, dark greyish in

compared to layer 1. Its thickness was measured about

colour. It is locally known as kaanp. It was levelled

35 cm. Two residential structures were exposed in this

as layer 7 resting on the bed rock (locally known as

layer one in Q 27 and another in Q 28. Of these the

bhokar.

one discovered in Q 27 was found just below 15 cm of

the surface (at a depth of 3.60 m from the reference
- 80 -

Annual report of excavation at Kanmer 2007-08 and 2008-09

Figure 6 Seal with geometric design

point). It is a rectangular room measuring 3.70 m by

3.20 m and built with rubble stone walls of which two

western arm near the temple as was indicated by

courses are extant. A considerably large patch of ash

the laser survey. Of course, its upper courses were

(1.30 m from east to west) and 70 cm (from north to

perhaps removed by the Historical people to prepare

south) was found close to the northern wall of this

a staircase. Therefore it appears that this point could

house, besides a small heap of broken pot sherds of

have been used as an entrance during the historical

red ware of Historical period. The purpose of such

period.

heap could not be ascertained. Another wall, oriented

east-west, was found joining this structure at its north

pottery belonging to KMR III. A stone structure was

western end. It is available up to a length of 3.20 m

partially exposed in the eastern section of this layer. It

only and could have joined the inner face of the fort

was found resting on layer 5. Layer 4 in the southern

in trench P27. The second room was discovered about

half of trench P 27 yielded a rectangular steatite seal

2.50 m to the south of the first one in trench Q28.

with perforated knob on the reverse. It has been found

The eastern and western walls of this room are of

engraved with a cross like sign. This seal appears to be

inferior quality whereas the northern one was built

quite interesting as it has coating of faience (Figure 6).

by erecting massive stone blocks perhaps robbed from

the fort wall. The southern wall was found 40 cm

deposited in various successive episodes. As many as

in width. This Historical deposit was underlined by

three major episodes represented by layers 5a, 5b and

a 25 cm thick dark greyish layer representing break

5c were identified on the basis of nature of dumping,

between the Historic and Harappan deposit. It was

its colour and signs of in situ burning. All these

numbered as layer 3. The Historic people prepared a

layers deposited against the inner face of the fort

chute by cutting down layer 3 between the aforesaid

were forming a steep slope. This direction of slope

rooms roughly in east-west orientation. The edges of

suggested that this ashy material was being thrown

this chute were lined with flat stone slabs. Perhaps it

from the top of the fort wall. Episode 5a was more

was a covered drain as some cap stones were found in

whitish compared to 5b, in which thin lenses of burnt

the exposed eastern end. The western end of the drain

earth and bits of charcoal were found. The thickness

passed through the then extant top of the western arm

of this episode gradually decreases as one goes away

of the fort. This channel perhaps served the purpose

from the face of the wall, besides the surface tends to

of an out let.

become gentler. At the base of this layer was found

- 81 -

Excavation did not revealed any gap in the

Layer 4 was loose ashy deposit which yielded

Layer 5 was represented by a huge ashy dump

J.S. Kharakwal et al.

Shifting of northern
inner arm

a thin lense of burnt earth, which looks almost

horizontal but actually has a gentler slope close to
the face of the wall. Episode 5c deposited against 15th
and 16th course of the wall is more or less horizontal

Towards the end of the KMR II B the width of

and is composed of thin lenses of burnt earth and ash.

northern wall was narrowed down by shifting its inner

The combined thickness of these three episodes was

face about 10 m further north. It is identified as the

measured about 3m. Layer 6 was light grey and pale

third major structural phase of the fort wall (Figure

yellowish soil mixed with ash, the surface of which

7). It appears that the outer face was maintained as

was burnt black. It was loose in composition much

per original either right on the revetting wall or the

harder than any episode of layer 5. Layer 7 composed

original wall face. This new inner face was exposed

of compact white clayey material was only about 10

in trenches Y15 and Z15 up to a length of 8.50 m. Its

cm thick. Layer 8 loose in composition consists of

extant height measured 1.15 m with 5 stone courses.

dark greyish soil mixed with brown sandy clay. Layers

It was made up of large stone blocks some of which

9, 10, 11 and 12 were primarily composed of brown fine

measured about 80 cm in length and 35 cm in height.

sandy clay. These were separated mainly on the basis

In trench Z15 it was found joined by a north-south

of their compactness and colour variation. Layers

oriented retaining wall (5.40 m long ) at the right

12 and 13 were found deposited against the earliest

angle. Only seven courses of retaining wall are extant

courses of the wall. These were composed of floor

with a height of 1m. At its southern end it takes

levels and patches of powdered shale. The cultural

a turns towards east and joins another wall (2.50

material, particularly pottery, of layers 5, 6, 7 and 8

m long ) roughly oriented in NNW-SSE. Initially

is comparable to layer 8 of central area or KMR II B

it was suspected that it may be corner of an inner

whereas layers 9, 10, 11, 12 and 13 yielded material that

bastion. Therefore, in the third and four field seasons

is grouped under KMR II A.

more area was exposed to understand the structure

Underlying the fort wall 14, 15 and 16 belong to

fully. During the excavation the above mentioned

the pre-fortification phase at the site. This deposit

another wall oriented roughly in NNE-SSW was

measuring about 70 cm in thickness yielded fragments

exposed up to a length of 6.80 m. It was found built

of bichrome ware and coarse red ware pottery, besides

of huge stone blocks. Its northern half shows an

a few sherds of unslipped red ware. This deposit is

eastward curve while the southern half is straight and

comparable to KMR I material of central part. Layer

oriented roughly north-south and appears to joining

16 was a thin layer of soil resting on top of ferruginous

the eastern inner arm of the fort. As many as seven

sand stone bed rock (locally known as bhokar). A few

courses of the face were exposed, but we could not

flakes of agate were discovered from this layer.

reach the base as it was right in the core of the fort

wall of earlier phase.

It is interesting to note that successive floor levels

were noticed along the inner face of the fort wall

In the northeastern corner were found several

suggesting that residential structures were built very

parallel revetments, oriented in NNE-SSW. Two

close to the wall and as such there seems no scope for

of them were found curved towards west on their

any street along the same.

south western end. One of them located close to the

inner corner of the fort was exposed up to a depth of
2.40 m (14 courses), but we could not reach its base,
because it was erected right in the core of the fort
wall. These may represent local repairing of fort wall.
- 82 -

Annual report of excavation at Kanmer 2007-08 and 2008-09

Figure 7 Third structural phase of the fort wall, northern arm

Somewhat similar parallel stone alignments were also

which is still unknown to us. Presence of the ash

noticed in the north western corner of the fort wall.

dumps on either side of the fortification seems to be

Perhaps these indicate sub-structural phases of the

result of such burning activities.

fort construction and may belong to KMR IIB.

After exposing these parallel walls, it seems that

width (on top) of the northern arm may have been

there was no inner bastion at the north eastern corner,

reduced to about 6 m. It is yet to be ascertained

at least after the end of urban phase KMR IIB. Could

whether the width of other three arms of the fort were

there be a watch tower kind of structure, squarish

reduced like the northern wall in order to create more

on plan, the corners of which were rounded? Either

space inside the fortification or for other unknown

these parallel walls were result of successive veneer

reasons. It can be confirmed only after further

added to the northern arm of fort at its eastern end

excavations, though the construction pattern, size of

or were simple constructions is yet to be ascertained.

stones, binding material found in the northern and

Some such indications were also exposed in the north

southern exposed corners of eastern inner are quite

western corner too.

different in appearance.

It appears that the 10 m wide space created

Therefore, towards end of the urban phase the

Thus the shifting of northern inner face further

towards the end of the urban phase KMR IIB, i.e.

north was identified as the third major construction

after shifting the inner face of northern wall further

phase of the fortification. By going through the

north may have been used for some activity. It has

construction method of Harappans it is likely that

been observed that large scale burning activities were

the fort wall at Kanmer may not have been 20 m

carried out regularly in large pits for some purpose

wide or 9 m high right from the beginning rather its

- 83 -

J.S. Kharakwal et al.

width increase gradually with addition provided from

result it appears in two sets. It is not clear if it was

external as well as internal side. Similarly its height

caused due to an earthquake? However, this event

may have been raised with each successive addition.

could not be corroborated by any other such evidence

These types of gradual additions have been reported

in the habitation deposit.

from Dholavira also. Of course, to understand the

precise history of growth of the fort wall, one has

shale was gradually washed off and in due course

to cut through the wall to identify its construction

of time largely got deposited on the tapering outer

phases.

surface of the fort wall. While exposing the outer

The outer face of the southern arm, barring the

face of the fort wall such material sticking on the wall

lower courses, was largely missing. However, the core

betrays coating. It may have eventually strengthened

of the wall was exposed at several locations. It was

the veneer as it was too difficult to expose such area of

noticed that the lower part of its core was built mainly

the wall.

of mud bricks. The bricks were prepared of brown

sandy clay (locally known as lilva). In the upper part

appears to have survived with maximum height of

of the construction of core, however, bricks made of

about 9.70 m with 30 courses. It was sealed with white

yellowish sandy clay (pila lilva) were also added to

plaster prepared from white shale.

It seems that in monsoon seasons the powdered

The eastern end of the northern arm of the fort

the brown sandy ones. The same clays were also found
used as cement in the stone veneer. Besides white,

POSSIBLE LOCATION OF THE GATE

pinkish powdered shale was also used as binding

material towards the end in the fort wall.
The evidence found near the south eastern corner

Excavation of the eastern half of the northern arm

indicates that the top of the fort wall was sealed

and northern half of eastern arm of the fort wall

perhaps with a thick layer of clay. Here, on the top of

ruled out the possibility of entrance to the settlement

the wall a 65 cm thick deposit consisting of successive

from these sides. Similarly, close examinations of

layers of a variety of clays (variety of sandy clays such

surface features of the mound indicate that there

as brown, greyish, yellowish and white and pinkish

may not be any entrance on the eastern and western

powdered shale) was exposed (in trenches II29, HH29

arm either. However, the contour map of the site

and HH30). The average thickness of each layer

shows that there is a general slope towards south.

was found about 4 cm. It is interesting to note that

Horizontal excavation in the south central part also

repetition of such layers more than once and sealed

reveals a gentle slope towards same direction. Thus it

every time by white or pink powdered shale suggests

seems that the entrance to the Harappan settlement

phases of regular repairing or maintenance of the fort

was provided from south through the southern arm.

wall. Of course, they may have to mine large quantity

Unfortunately, outer arm at the possible location

of shale available locally. This feature is evidenced

of entrance is largely missing or destroyed by the

only at the top of south eastern and north eastern

villagers. Perhaps further excavation across the wall at

corners. Thus the 9 m high strong fort with white

this point of arm may provide some evidence of the

plastered top must be a very impressive construction

entrance.

of its builders. It is a bit difficult to asses how much

time, labour and energy went into its erection. It is
interesting to note here that the horizontal layers of
thick clay deposit show a slight vertical sinking as a
- 84 -

Annual report of excavation at Kanmer 2007-08 and 2008-09

Figure 8 Residential structures in the central part

RESIDENTIAL STRUCTURES
IN THE CENTRAL AREA

The earliest residential structure at Kanmer was

partially exposed in the north eastern half of trench

Z30 at depth of 3.80 m below the surface. It was

Excavation at the site revealed five major phases of

exposed up to a length of 2.20 m. Oriented in NNW-

Harappan structural activities. Of them the first

SSE its width was measured around 45 cm. Beneath

three have been identified in the central area (Figure

this structure were of course found four floor levels

8) while the two later ones, belonging to the Late

(at a depth of 4.50 m from surface). Interestingly, the

Harappans were exposed in the south eastern area

successive floors show a gradual shift towards north to

of the site. Each of these major structural phases has

a street or open space.

sub phases suggesting constant addition or alteration

by various generations. The few partially exposed

the same trench (Z30) overlying the first one. It was

structures of these phases revealed that most of them

represented by a partially exposed room (4.80 m

were largely disturbed or destroyed by the activities of

by 3.0 m) with a platform (1.53 m by 1.23 m) on the

the subsequent occupants of the site.

street or open space to its north. The walls of this

In the first cultural period, i.e. KMR I no

structure join each other at right angle and are made

structure could be discovered so far as the excavation

of stone. Yellowish sand was used as binding material.

was restricted to very small area. However, as many as

The extant height of its northern wall has measured

four successive floor levels were identified in trenches

88 cm (6 courses) whereas the height of western

AA28, Z28, and Z30 in the central part. The earliest

wall measured 45 cm with 4 courses. A platform

among them was prepared right on the bed rock.

seems to be provided at the entrance of the house. A

- 85 -

The second structural phase was identified in

J.S. Kharakwal et al.

medium sized jar found placed near its outer north-

measured 1.20 m. This wall measured 3.85 m in length.

eastern corner appears to be a sullage jar. About 7m

It seems all these structures belong to sub phases of

north of the platform was found a room of another

second structural phase.

house (northern house) measuring about 2.25 m by

1.50 m in dimension. Exposed partially it was also

stones. The locally available yellowish fine sandy clay

found roughly oriented in NNW-SSE direction thus

(pila lilva) was used as binding material. The breadth

conforming to the general orientation of the site.

of walls was found ranging from 45 cm to 50 cm. The

A stone alignment (3.56 m) was discovered in the

lay out plans of these structures indicate that they

western half of Z28, which seems to be contemporary

restrictedly followed the orientation of the main fort,

to the aforesaid house. Thus the supposed street or

thus indicating a planned settlement inside.

open space was exposed up to 5m from east to west in

length and up to 7m from north to south in width. It

It seems that by now the open space (exposed in Z29)

was found continuing towards further west.

had encroached upon by structures. Several structures

A sub-phase of this complex was identified

of this phase were exposed in a large area (trenches

with rectangular rooms situated to the south of

X28, X29, X30, Y28, Y29 and AA29) located on

open space or street. As many as three rooms were

the previous open space or street. In this phase the

partially exposed in trenches AA30, Z31 and Y30. The

average breadth of walls was found about 50 cm.

aforesaid open space appears to be a by-lane towards

Use of yellowish sandy clay was stopped foe binding

south, perhaps connected to the main entrance of

material in constructions. In X30 a button shape

the settlement as the gentle slope in the stratigraphy

steatite seal with perforated knob was discovered in

would indicate. This structural complex was possibly

layer 8. More than half a dozen rectangular rooms of

located to the east of this lane.

this phase were identified in the central part. In one

of the rooms a gap found in the western wall indicates

About 1 m to the east of the northern house

All these structures were made of semi-dressed

The third structural phase belongs to KMR IIB.

was found an alignment (3.20 m) of stones, oriented

that entrance was provided from the west side.

NNW-SSE, the southern end of which was missing.

To the west of its existing southern end a small

were finely plastered possibly with the powdered shale

squarish platform was found whereas to the east of

mixed with cow dung. Even today the local villagers

this structure was found a floor and a large circular

prepare their house floors in such way. Therefore the

patch of ash (1.50 m by 1.25 m). Further to the east

local folks were easily able to identify the ancient

of this structure was found another stone alignment

floors during excavation.

The floors prepared by levelling and filling of soil

at about 3 m distance. What is interesting to note

here that these structures appear to have been made

SOUTHEASTERN AREA

by encroachment on the open space or the lane.

Therefore, these partially exposed structures show
some kind of change in planning. A terracotta sealing

In continuation of the second seasons excavation a

was found from the SW corner of Z28 on the floor

few more trenches such as GG29, GG30, FF29, FF30,

levels of layer 9.

EE29 and EE30 were laid in south eastern area of the

A few more walls, belong ing to different

mound order to understand the partially exposed

rectangular rooms, were discovered to the east and

large structural complex near the south eastern corner

northeast of the house exposed with a platform. The

of the fort. During the excavation a large number (a

extant height of one of the walls of these rooms has

dozen) of circular pits were exposed just below the

- 86 -

Annual report of excavation at Kanmer 2007-08 and 2008-09

surface. All these pits were made of fine clay, perhaps

brought from near by lake or pond and were coated

were three successive structural phases of stone walls.

with white fibrous material. Many of them yielded

It has been noticed that each time the complex may

layers of sand and ash and medieval pottery and

have 4 or 5 rooms. In case of the latest complex

charred grains. Therefore these pits were identified

numbered as LH 1, the width of wall increased up to

as silos and were assigned to cultural period V of

60 cm whereas in older structures it was hardly 50 cm.

Based on the stratigraphy, it appears that there

Kanmer. Underlying these pits were discovered a

few minor structures with one or two courses. These

NORTHEASTERN AREA

structures perhaps belonged to Historical phase as

some of them were cut by the pits. These structures
were found in FF30, EE30 and EE29 whereas the pits

Barring a few lower courses, the outer face of north

were somewhat evenly distributed throughout the

eastern corner of the fort was missing. As the extant

exposed area.

top of the revetment was visible close to the corner

Like other parts of the mound the dark grayish

we decided to expose outer arm of the fort as well as

layer, represented by layer 2 in this area, was found

the revetting wall to understand the constructional

overlying the Harappan remains. In the northern half

phases (Kharakwal et al. 2008). In continuation of

of GG29 was exposed a circular platform made of

the last seasons work the outer addition or revetting

stones in layer 3, the diameter of which was found 1.50

was exposed up to a length of 28 m. The entire face

m. On the southern side of this circle were discovered

was found with a considerable southward tilting.

two medium sized pots of Coarse Red or Gritty Red

The addition or revetting was largely made of semi

Ware. It appears that perhaps this could be a floor

dressed stones of the average size of about 30 x 25 x

level inside a house. However the exact use of the

8 cm. However, some of the stone blocks were quite

circular structure is not known. In the north western

large as their length measured 60 cm and thickness 10

corner of the same trench was discovered a highly

cm. Such large stones were mostly used in the lower

burnt red feature in a small area (60 cm by 40 cm) on

part of the wall. It has been observed that the original

the floor of a house. The eastern and southern walls

exterior of the fort was made of well dressed large

of this house were also partially exposed in the same

stone blocks than to the latter addition or revetting.

trench.

Therefore the original wall looks very impressive. Near

In the southern half of FF29 half a dozen stone

the north eastern corner the fort height seems to have

structures, possibly belonging to equal number of

survived intact as the plaster of white shale, perhaps

houses in layer 3 have been recorded. Some of them

applied on its top is still visible. The total height of

have been found destroyed by later pits. Among these

the outer face was found 9.50 m.

at least four walls appear to be superimposed on the

older ones suggesting stratigraphically a succession

BB16 and BB17, located in the north eastern corner

of structures in south eastern area. These walls have

of the mound, were basically selected in order to

survived with courses ranging from 3 to 5 with length

expose the inner corner of the fort to understand its

from 1.50 m to 3 m. In the south western corner of

inner growth. Soon after the removal of surface soil in

FF30 also stone walls of three different phases were

trenches Z16, AA16 and BB16 it was realised that this

noticed. In the northern half of FF29, close to the

whole area was perhaps used as dumping space by the

northern section facing south, also two structures

Historic people as it yielded large amorphous shaped

were discovered jetting out of the section.

burnt patches. In fact it seems that the Historic

- 87 -

Trenches Z16, Z17, Z18, AA16, AA17, AA18 and

J.S. Kharakwal et al.

people dug the core of the Harappan fort wall made

like powdered shale mixed with yellowish sandy clay.

few pits also. Therefore the historical material was

It was found against the eighth course of the fort wall.

found resting directly on the Harappan fort wall.

Perhaps it was the binding material of the wall which

However, layer 1 in trenches Z17 and Z18, AA17 and

appears to have washed off and deposited against the

BB17 yielded mixed pottery belonging to Historic

wall. Layer 7a was formed of brownish sandy material

and medieval period. Layer 2 was represented by

and compact in nature whereas 7b was loose. As

thick deposit of dark greyish soil, which was coarse

many as 13 courses of the inner face of the fort wall

in nature. The nature of this layer was very well

were exposed in this trench with a total height of

comparable to layer 6 of central part, layer 3 of

about 2 m. In this area it seems that the inner face was

western area and layer 2 of south eastern area. It has

repaired as the size of semi dressed stones used was

been identified as a break between the Early Historic

much smaller (20 x 10 x 8 cm average) compared to

and the Harappan deposit. In layer 3 a retaining wall,

other exposed areas. No binding material was found

oriented roughly east-west, was partially exposed,

between the stones. Excavation could not continue

against which was found a floor level. Its extant height

further down in this area due to discovery of a furnace

measured about 1m. To the east of this floor level

with a central column. All these layers were found

was found a residential structure (LH 3), oriented

sloping towards south. Perhaps this entire deposit

roughly north-south, which was partially exposed (up

seems to be thrown down from the top of the fort

to a length of 3.50 m) in trenches AA16 and AA17. It

wall. Layer 7b in which the furnace was discovered

has survived up to a height of about 1 m with seven

close to the face of the wall was the last activity carried

courses. The southern wall of this house oriented

out inside.

roughly east-west has survived up to a length of 3.40

m. The average width of these walls measured about

trench AA17. The eastern inner arm was exposed up

60 cm. The eastern and northern walls of the room

to a length of 5.50 m, whereas the northern inner arm

are missing. From the floor levels fragments of dish

was exposed up to a length of 8 m. The northern inner

on stands, jar and Gritty Red Ware, besides featureless

arm appears to have been repaired close to the corner

bowls were recovered. Most of this material was

as it is made of much smaller stones compared to the

ochreous. Thus on the basis of stratigraphy and

eastern one and neither has the binding material like

ceramic assemblage the aforesaid structures and floor

the surviving eastern one. The eastern wall was located

levels and pottery was assigned to cultural period

about 1.80 m below the surface. It was made of large

KMR III.

semi dressed stone, some of which were measured

In order to expose the inner face of the fort wall

about 60 to 80 cm in length, 25 to 45 cm in breadth

excavation continued in AA16 also. In this trench

and 25 cm in thickness. The Harappans often used

layers 4, 5, 6 and 7 were found deposited against the

white and pinkish powdered shale and yellow sandy

inner face of the fort wall. The total height of this

clay locally known as pila lilva as binding material.

The inner corner of the fort was exposed in

exposed section was measured about 3 m. Among

these excavated layers, layer 4 was composed of ash

FURNACE

whereas layer 5 was light greyish in appearance and

was composed of brownish sand mixed with sand and
also bits of charcoal. Layer 6 was composed of various

A bulb-shaped plan of a furnace with a central

lenses of burnt earth, charcoal etc. One of the lenses

cylindrical hollow column (diameter 31 cm, depth

of layer 6 was composed of white clay which looked

35 cm) (Figure 9) was exposed in trench Z17 about

- 88 -

Annual report of excavation at Kanmer 2007-08 and 2008-09

Figure 9 Furnace with a central column

10 m west of north eastern corner of the fort. It

was found inside the column. Several tubular faience

was built close to the inner face of the wall and was

beads were recovered from the furnace area and near

oriented in NNW-SSE direction. It measured 1.40

the square platform. Therefore, it is likely that this

m from north to south and 96 cm from east to west.

was a faience bead making furnace. This furnace was

The clay walls of the furnace were barely 4 cm thick

found sealed by layer numbered 4. This layer consists

and the area between the column and outer clay wall

of several episodes of ash which was possibly thrown

was found completely filled with ash. The column

from the top of the fortification wall by the people

with an opening only from top was found filled with

also belonging to KMR IIB phase. The furnace seems

whitish ashy material. The burnt red colour of the

to have been built during the initial stage of this

cylindrical column, the outer clay wall and the earth

phase.

around the furnace indicate that the temperature

raised in the furnace may have been more than 700

MINOR OBJECTS

C. Indications are that the heat generated in the

furnace also damaged the face of the fort wall. A small
square (90 x 90 cm) platform made of flat stones just

A large variety of minor objects such as beads of

to the east of the furnace also seems to be associated

terracotta, paste, semiprecious stones, gold, shell,

with it. This platform was prepared on a floor level.

seals, seal impressions, terracotta cakes, dices,

A thick cubical sandstone block was found lying on

gamesman, amulets, bead polishers, drill bits, rough-

the eastern margin of the floor to which was found

outs and weights have been discovered. Terracotta

sticking some whitish substance identical to what

cakes were found distributed throughout the deposit

- 89 -

J.S. Kharakwal et al.

5cm

Figure 10 Terracotta objects with a seal impression from Kanmer

in the urban phase (KMR II). They are smaller in

impressions, found in a room in layer 9, is circular in

size in the beginning and gradually were made bigger

shape, flat on the obverse and convex on the reverse.

during the later part of the urban phase (KMR IIB).

The impression was made by a square seal, which

A few of them were also discovered in the upper

may have been about 26.7 x 25 mm. The diameter of

levels with KMR III deposit. Beads outnumber all

this seal impression was measured 34.52 mm. It has

other finds as they have been discovered over 20,000

a perforation (diameter 4.15 mm) which is slightly

in number. The most striking discovery of the site

off the center. On the obverse is depicted a unicorn,

appears to be three identical terracotta objects with

besides a few Harappan letters. The reverse shows

a seal impression, which appear to be sort of identity

a signature. The reverse signature in each case is

cards or passport, perhaps used by Harappans during

different than the others.

their long distance trade (Figure 10). One of the seal

Among others objects a button shaped steatite

seal depicting a donkey (?) appears to be quite

Table 1 Beads discovered at Kanmer

significant. The discovery of over 158 drill bits of

Material

Number

Ernestite (Figure 11) and about half a dozen of

1. Steatite

17580

Rohri chert is perhaps the other significant finding

2. Faience

328

considering the size of site and duration of work

3. Paste

14

carried out at the site. The Rohri material appeared

4. Agate

31

at the site from KMR II. The find of large quantity of

5. Carnelian

59

raw material, particularly agate, variety of drill bits,

6. Lapis

11

7. Amazonite

rough outs of carnelian, debitage, chips, unfinished

8. Serpentine

9. Sandstone

10. Limestone ?

11. Terracotta

30

12. Shell

141

13. Bone

others as more than 18,000 have been discovered so

14. Gold

far. They have been recorded as micro-beads, mini

products, beads polishers all strongly indicate that

perhaps bead making was one of the most popular or
prized occupation of Harappans at Kanmer.

The majority of Harappan beads type are disc,

bicone and tubular. Beads of steatite outnumber all

- 90 -

Annual report of excavation at Kanmer 2007-08 and 2008-09

2cm

Figure 11 Drill bits from Kanmer

2cm

Figure 12 Carnelian beads from Kanmer

- 91 -

J.S. Kharakwal et al.

disc type, disc type, cylindrical or tubular, segmented,

inferior. A few examples of shell, bone and metal ones

wafer type and triangular ones. Except for one hoard

have been found from the Mature and Late Harappan

from a house of KMR III level most of them have

levels. The following Table 1 shows quantity of beads

been found from streets. Some of the tubular beads

of different material at Kanmer discovered in the

have been decorated with circular marks. In KMR

Harappan deposit.

III, the quantity of tubular beads decreases and

sometimes they have been found coated with green

LITHICS

or blue colour, perhaps to look like faience. Faience

beads are disc type, short tubular or tubular and
blackish, greyish and greenish in appearance. The

Hundreds of lithic objects such as flakes, cores,

blackish or greyish examples appear to be the result of

lunates, notches, long parallel sided blades were

ill or over firing. Most of the tubular ones are 2 cm to

discovered from Kanmer. They are being studied by

3 cm in length and have been discovered from north

H. Endo of Research Institute for Humanity and

eastern area (trenches Z16, Z17, Z18, and AA17). It is

Nature, Kyoto. The raw material of lunates, flakes

likely that they were manufactured in the northern

and cores was mostly agate and local chert, perhaps

area of the mound, which strongly appears to be used

from Mardhak Bet. However notches, Ribbon blades

for craft activities. The paste beads, though very few in

(Figure 13) were made of Rohri chert. Interestingly no

quantity when compared to steatite, were found quite

raw material or any core of Rohri chert was found at

attractive. Most of them are either black or red or

the site. Perhaps the Harappans of Kanmer received

white and globular, disk type or cylindrical in shape.

finished blades of Rohri. What is interesting is that

The beads of semiprecious stone have been

we have been able to locate as many as 9 drill bits of

identified as carnelian (Figure 12), agate, lapis lazuli,

Rohri from the site. It seems that besides Ernestite

chalcedony, serpentine and bloodstone. The site

only Rohri chert was used for preparing drill bits at

yielded raw material of agate besides, chipped, rough

Kanmer.

outs, grinded, unpolished bead blanks. A large variety

of shapes such as cylindrical, disc type, biconical,

POTTERY

truncated bicone, circular, barrel, globular, circular,

flat bicone, diamond were found. All these findings
indicate that a variety of beads were manufactured

The ceramic assemblage was mainly represented

at the site. Perhaps the raw material was brought

by Red Ware right from the beginning, which was

from nearby Mardhak Bet, about 20 km northeast

treated with a variety of slip e.g., Red, chocolate,

of the site in the Little Rann. Except for lapis lazuli/

buff, cream, black and reserve. Except for black

sodalite, source of raw material for all these beads

slip and finer variety of reserve slip (grey variety),

types could have been the Little Rann and its adjacent

all other types were introduced at the site right

areas. Lapis may have been brought either from Ajmer

from the beginning. Both fine and coarse varieties

in Rajasthan or from Afghanistan. The beads of

remained in use throughout the habitation, of course

carnelian are disc type, globular, bicone and tubular

the coarse variety accounts for a small percentage

and etched ones. These beads were drilled after

of the assemblage in the early levels and it gradually

polishing as the find of many broken but polished

increased in the subsequent phases with a mat surface.

examples would indicate. In the Late Harappan phase

Besides Red Ware, Reserve Slipped with grey core,

the quality of surface treatment of such beads was

Black-and-Red, Buff ware were some of the other

- 92 -

Annual report of excavation at Kanmer 2007-08 and 2008-09

Figure 13 Lithics from Kanmer

10cm

Figure 14 Painted pottery from KMR I

- 93 -

J.S. Kharakwal et al.

types used. The fine variety of Red Ware (in case of

types gradually increased in the later levels. A variety

bowls, dishes and medium sized jar/pots) was very

of bowls with flared body and inverted rim or with

often treated either with red slip or with chocolate

slightly pronounced body and featureless rim or with

slip or sometimes with both. In the lower levels

carinated body and everted rim or with stand were

(KMR I and KMR IIA) the slip was uniformly

part of this assemblage. Two different types of stands

applied on external surface whereas later on very often

such as cylindrical shape and with flared body were

the outer surface appears to have been moistened

discovered. Perhaps the former one was part of dishes.

with slip. Sometimes slip was not applied to the entire

Both types were often decorated with bichorme

outer surface rather wide panels were prepared all

paintings. Bowls with carinated body and featureless

round the body leaving major area unslipped. Though

rim types continued in the subsequent cultural

pottery was fired at high temperature, it seems that

periods too but with little change both in material

the uniform firing was not maintained in all the types.

and shape. Basins with carinated body appear to have

Towards the end most of the pottery types appear

been popular in the lower levels. Often the fine variety

ochreous. It is not ascertained if this kind of character

was wheel thrown, though slow wheel and handmade

is result of poor firing technique.

types pre-dominate the assemblage. The bichorme,

KMR I was represented by both slipped and

cream slipped and coarse type pottery is also common

unslipped red ware, bichorme and polychrome, cream

in Anarta complex of North Gujarat (Ajithprasad

slipped, chocolate and buff slipped pottery (Figure

2002).

14). The jar/pots were dominated by globular body

shapes often with concave neck and everted rim.

triangles, chess board patterns , vertical registers of

The bichrome types were treated with black or dark

wavy lines or squares, diamond shape designs, fish

greyish or light brown or red slip. Very often a white

scales discovered in bichrome type jar pots in from

panel was prepared on the neck or shoulder portion

Kanmer (KMR I) appear to be identical to those

on which variety of geometric patterns were executed

ones found from Amri Period I by Casal (1964).

with black, brown or red colour. In case of some

For example a panel of triangles (Period IA Fig. 44

polychrome decorations sometimes orange colour

type 49, Period I D Fig 66 type 232), chess board

was also used either to fill gaps or to make boarder.

patterns (Fig 44 type 52, Period IC Fig 56 type 159),

It looks as if it was used to highlight certain design.

vertical register of way lines or squares from Period

Compared to bichrome the polychrome sherds are

IB, diamond shape designs (Period IC Fig 57 type

only a few.

162), fish scale (Period I D Fig. 64 type 214) of Amri

have exact parallels at Kanmer in the first or earliest

Sometimes unslipped Red Ware or buff slipped

A variety of design patterns such as panel of

medium sized jar/pots were decorated with a band

cultural phase (KMR I).

of thin grooved lines all around the shoulder perhaps

by a comb like instrument. These sherds remind the

carinated body, with flared sides and featureless rim,

Kot Diji type pottery. Jar/pots with coarse fabric

bowls with stand, with vertical sides and featureless

were generally left unslipped and the lower part of

rim, jar/pots with concave neck and everted rim,

their body was rough. From this assemblage, except

basins and dish on stand (?) with nail headed rim also

for bichrome and polychrome types, jar/ pots with

have parallels at Amri in Period I and II (see e.g., Fig.

grooved decoration tend to disappear in KMR II as

51 type 112; Fig 64 type 216, Fig 65 type 228; IC Fig.

the discovery of only a few stray fragments indicates.

54 type139; Fig. 62 types 200, 202, 203; Fig 65 type

On the other hand the chocolate and red slipped

226; Fig. 70 type 282 Fig 69 type 273; Fig 74 tye 317;
- 94 -

Besides, a few bowls with everted rim and

Annual report of excavation at Kanmer 2007-08 and 2008-09

Fig 79 type 347).

motifs become visible only after making them wet.

In KMR IIA most of the shapes of KMR I

Did the Harappan use such pots for storing water?

continue with little change, but now the assemblage

It seems that the painting may have been visible only

is dominated by fossil types identical to urban phase

when there was water in them.

e.g., S shaped jars, perforated jars, dish on stands,

large jars with clubbed rims, terracotta cakes, besides

of jar/ pots, bowls were treated with red or orange

a few fragments of goblets with flattened base.

slip and their core was whitish. Their appearance

In case of bowls a few fragments of handles with

is different than buff pottery. It seems that the

perforation were found, besides Black-and-Red Ware

Harappans mixed powder of white or pink shale with

with featureless rims and with pronounced body. In

clay to make such pottery.

case of medium and large jar/pots generally one or

sometimes two parallel lines were depicted running

and jar with elongated neck appear to be common.

all around the shoulder or body part. Of course some

The entire assemblage appears ochreous. The gritty

of the medium sized jar/pots were decorated with

sometimes appears brittle.

In this assemblage, a few rim and body sherds

In KMR III most of the KMR IIB types continue

natural motifs such as papal leaf or some other leafs or

branch like motif.

FAUNAL REMAINS

In KMR IIB the pottery of earlier phase continue,

but some new types were introduced at the site

such white painted Black and Red Ware and Sorath

The faunal analysis is being carried out by P.P. Joglekar

Harappan material. Bowls with featureless rim and

and Pankaj Goyal from Deccan College, Pune.

vertical sides, sometimes with stud handles and star

Though faunal remains were recovered from almost

motif were common. Besides, a very coarse Red Ware,

all the trenches, they were not evenly distributed. In

which may also be called as gritty Red Ware appears

most of the trenches bones were encrusted with soil,

in this phase. It has often been treated with thick red

but they were able to identify cut marks on bones

slip on the shoulder and rim part whereas the lower

and charred bones in some cases. They have identified

part of the body is very rough or seldom appears

several animal taxa, which include mammals, birds,

rusticated. A variety of deep incised geometric design

fish, reptiles and molluscan species. Among the

patterns have been engraved on the shoulder portion

domestic animals, cattle, buffalo, sheep, goat and

of such pots with some sharp object. In fact together

pig were identified. More than a dozen wild animals

with white painted Black and Red ware this pottery

were identified in the collection, including the nilgai,

type reminds the Ahar pottery of Southern Rajasthan

antelopes, deer, carnivores and rodents.

(Misra et al. 1997). It is possible that the idea of such

pottery types travelled to this region from Ahar

predominate like in the other cultural phases at the

complex. Large and medium sized jars of this gritty

site. Among the domestic animals the cattle and

type are often found burnt black on the inner side,

buffalo constitute a majority and many of them bear

discovered particularly from south eastern part of the

cut marks or are charred suggesting that they were

mound. It is possible that they were used for heating

consumed. The wild animals represented in the Late

bead blanks or raw material at the site. We carried out

Harappan phase include a large bovid, wild pig ,

some such experiment at the site which yielded similar

antelopes, deer, carnivores and small mammals like

result. Most of the pottery types become finer during

hare and rodents (9.60%). In the Mature Harappan

this period. In some medium sized pots painted

phase 96% faunal remains belong to cattle and buffalo

- 95 -

In the Late Harappan phase domestic animals

J.S. Kharakwal et al.

PALAEOETHNOBOTANICAL
INVESTIGATIONS

(81.09%), followed by sheep and goat (13.93%).

Besides these, wild animals like nilgai (blue bull),
four-horned antelope, a suidae species and the
porcupine were identified.

In order to understand subsistence of the Harappans,

Thus as many as twenty-four species were

Anil K . Pokharia (Birba l Sa hni Institute of

identified, out of which the domestic mammals

Palaeobotany, Lucknow, India) continued the study of

were represented by seven species (cattle, buffalo,

palaeobotanical remains, particularly of charred grains

sheep, goat, pig, dog and cat) wild mammals were

in the third and fourth seasons. He has been able

represented by 14 species (Nilgai, wild pig, antelopes

to conduct preliminary examination of some of the

(Blackbuck, Chinkara and Four-horned antelope),

samples collected by him from the site. He primarily

deer (Sambar, Chital, and Mouse deer), a felid species,

employed water flotation technique to retrieve the

porcupine, hare and rodents (house rat and desert

charred grains at the site. He was assisted by several

rat). Besides these, a few birds, reptiles, fish and shells

research students of Rajasthan Vidyapeeth and water

were also identified.

was brought from a distance by bullock cart to wash

Evidence of charring, butchering and cut marks

the samples. Initially we attempted quantitative

has been found on a large number of bones which

control on our sampling, but gradually we realized we

perhaps indicate that these animals may have been

require some strategy for varied context. For example

part of their diet. The relative proportion of charred

from our floors or general habitation levels ten

bones was larger in the Mature Harappan phase

ghamelas soil was sampled for floatation. Ghamela is

compared to the Late Harappan and Historic. Some

a deep basin kind of metal utensil. From hearth/ fire

of the bones of cattle/buffalo and sheep/goat were fire

places generally two ghamelas and from burnt patches

hardened and it is likely that they may have been used

five ghamelas of soil sample was employed for washing

as tools.

to retrieve the grains. One ghamela contains about 5

Goyal and Joglekar (in press) conclude that

kg of soil. Soil samples for wet flotation were selected

domestic animals were the most important resource

from north eastern, central, south eastern and western

to the animal-based subsistence throughout the

areas. Each sample was poured into the body of water

occupational history. The higher percentage of

in a tub that was agitated so that light material is

domestic animals like cattle, buffalo, sheep and goat

buoyed to the surface and skimmed off through a 25

may suggest that stock raising and pastoralism was

mesh sieve. Floatation allowed recovery of all size

an important component in the subsistence activity

and classes of botanical material preserved in the

of the community in all the periods. They further

sediment, making qualitative and quantitative analysis

point out that the proportion of wild mammals used

possible. A large amount of carbonized material was

in the Late Harappan phase was more than any other

gleaned from well stratified trenches.

cultural phase, clearly indicating an increase on the

dependence on the wild mammals. Of course one

found in utterly fragile, highly burnt and mutilated

has to keep in mind that in all our trenches we have

state of preservation. The inferences of the study,

exposed Late Harappan deposit more than any other

therefore, are based on a small fraction of the

cultural phase. Eventually the faunal assemblage of

material. The remains were sorted out under the

the Late Harappan outnumbered all other phases.

stereo-binocular microscope and thereafter cleaned in

Therefore, this conclusion, as the authors have

acid-alcohol (glacial acetic acid 10% + ethyl alcohol

pointed out, is tentative.

50% in equal volume). The identification, Pokharia,

- 96 -

A sizeable amount of botanical material was

Annual report of excavation at Kanmer 2007-08 and 2008-09

Table 2

Botanical remains recovered from Harappan Kanmer identified by Anil Pokharia of Birbal Sahni Institute of

Palaeobotany, Lucknow, Uttar Pradesh, India

S. no Area of mound Trench and Qdt

Cult. Pd

Layer

Depth in meter Botanical Remains

Central

Z30 SW

KMR I

12

10.02

Vigna, Macrotyloma

Central

Z30 SW

KMR I

12

9.62

Central

Z30 SW

KMR I

12

9.42-9.67

Ho rd e u m , Vi g n a , M a c r o t y l o m a ,
Trianthema
Hordeum vulgare, Triticum aestivum,
Trianthema triquetra, Polygonum sp.,
Ziziphus nummularia, Vicia sp.

Central

Z30 SW

KMR I

12

9.33

Central

Z30 SW

KMR II A 11

9.23

Central

Z30 SW

KMR II A 11

9.06

Central

Z30 SW

KMR II A 11

9.00-9.06

H o r d e u m v u l g a r e , Po l y g o n u m s p . ,
Tr i a n t h e m a t r i q u e t r a , Z i z i p h u s
nummularia
Z iziphus nummul aria, Trianthe ma
triquetra
Z iziphus nummul aria, Trianthe ma
triquetra
Vigna sp., Pisum sp., Trianthema sp.,

Central

Z30 SE

KMR II A 10

9.9

Hordeum, Vigna sp., Trianthema, Ziziphus

Central

Z30 SE

KMR II A 10

9.4

Hordeum, Setaria, Trianthema, Ziziphus

10

Central

Z30 NE

KMR II A 10

9.2

Hordeum, Vigna sp., Trianthema

11

Central

Z30 NE

KMR II A 10

9.09

Hordeum, Oryza, Trianthema

12

Central

Z30 SE

KMR II A 10

9.09

Trianthema

13

Central

Z30 NE

KMR II A 10

9.04

Hordeum, Setaria sp.

14

Central

Z30 NW

KMR II A 10

9.04

Hordeum, Triticum, Ziziphus, Trianthema

15

Central

Z30 NE

KMR II A 10

8.9

Setaria sp.

16

Central

Z30 NW

KMR II A 9

9.93-9.98

Ziziphus sp., Setaria sp.

17

Central

Z28 SE

KMR II A 9

7.83

Ziziphus

18

Central

Z28 SE

KMR II A 9

7.53

Ziziphus

19

Central

Z28 NW

KMR II A 9

7.51

Hordeum, Trianthema, Polygonum

20

Central

Z28 SW

KMR II A 9

7.4

Hordeum, Ziziphus, Trianthema

21

Central

Z28 NE

KMR II A 9

7.33

Hordeum, Ziziphus, Trianthema

22

Central

AA28 NW

KMR II A 9

7.33

Hordeum, Trianthema

23

Central

AA28 SW

KMR II A 9

7.59

Vigna radiata, Trianthema

24

Central

AA28 SW

KMR II A 9

7.38

Vigna radiata

25

Central

AA28 NW

KMR II B 8

7.11

Linum

26

Central

AA28 SW

KMR II B 8

7.11

Gossypium

27

Central

Y28 SE

KMR III

5.73

Hordeum, Gossypium, Trianthema

28

Northeastern

Z17 SW

KMR II B 6

5.39

Hordeum vulgare, Triticum aestivum,

Chenopodium sp., Setaria sp., Trianthema
triquetra, Ziziphus nummularia

29

Northeastern

AA17 NW

KMR II B 6

5.4

Sorghum bicolor, Sesamum indicum

30

North eastern Z18 NW

KMR II B 4

3.95

31

North eastern Z18 NW

KMR II B 4

3.9

32

North eastern Z18 NW

KMR II B 4

3.82

Vigna radiata, Ziziphus nummularia,

Polygonum sp., Trianthema triquetra,
Setaria sp., Scleria sp., Acacia sp.,
Triticum aestivum, Oryza sativa, Vigna
radiata, Linum usitatissimum, Sesamum
indicum, Setaria sp.,
Hordeum vulgare, Triticum aestivum,
Oryza sativa, Pennisetum typhoides, Vigna
radiata, Lathyrus sativus, Macrotyloma
uniflorum, Sesamum indicum, Gossypium
arboreum/herbaceum

- 97 -

J.S. Kharakwal et al.

S. no Area of mound Trench and Qdt

Cult. Pd

Layer

Depth in meter Botanical Remains

Linum usitatissimum, Ziziphus
nummularia, Setaria sp., Polygonum
sp., Solanum sp., Trianthema triquetra,
Asphodelus sp., Vicia sp.
Triticum aestivum, Oryza sativa, Vigna
radiata, Pisum arvense, Macrotyloma
uniflorum, Sesamum indicum, Linum
usitatissimum, Gossypium arboreum/
herbaceum, Eleusine ?, Cucumis sp., Setaria
sp., Asphodelus sp., Solanum sp., Polygonum
sp., Trianthe ma triquetra, Z ziphus
nummularia, Vicia sativa
Vigna sp., Ziziphus nummularia, Setaria
sp., Trianthema triquetra
Hordeum vulgare, Setaria sp., Cyperus sp.,
Trianthema triquetra

33

North eastern Z18 NW

KMR III

3.84

34

North eastern Z18 NW

KMR III

3.72

35

North eastern Z18 NW

3.6

36

North eastern Z17 SW-NE

cultural
break

37

North eastern Z17 NE

cultural
break

38

North eastern AA11 SW

KMR II B 4

5.22

39

South eastern FF29 NW

KMR III

Hordeum vulgare, Triticum aestivum, Vigna

radiata, Lathyrus sativus, Pisum arvense,
Linum usitatissimum, Sesamum indicum
Hordeum vulgare, Triticum sp., Polygonum
sp., Ziziphus nummularia, Trianthema
triquetra
Pennisetum typhoides, Sorghum bicolor

40

South eastern FF29 SW

KMR III

3.79

Vigna radiata

41

South eastern FF29 SE

KMR III

3.77

42

South eastern FF30 NE

KMR III

3.12

43

South eastern FF30 NW

KMR III

3.23

Pennisetum typhoides, Ziziphus sp., Setaria

sp.
Sorghum bicolor, Pennisetum typhoides,
Ziziphus sp.
Vigna radiata, Ziziphus nummularia

44

South eastern GG29 NW

KMR III

3.16

Hordeum vulgare, Oryza sativa, Ziziphus

nummularia, Trianthem triquetra

45

South eastern GG29 NW

KMR III

3.14

46

South eastern GG29 NE

KMR III

2.9

47

South eastern EE29 NE

KMR III

4.1

Vigna radiata, Setaria sp., Trianthema

triquetra
Vigna radiata, Chenopodium sp., Indigofera
sp., Setaria sp.
Setaria, Polygonum

48

South eastern EE 29 NW

KMR IV

3.25

49

North eastern AA 17 SW

KMR IV

1 (dump) 4.08

50

North eastern AA16 NW

KMR IV

1 (dump) 2.86

Vicia sp., Andropogon sp., Trianthema

triquetra, Solanum sp., Indigofera sp.,

51

Central

cultural
break

6.47

Coix sp.

52

North eastern Z 16 SE

KMR IV

2.99

Hordeum vulgare, Pennisetum typhoides,

Macrotyloma uniflorum, Linum
usitatissimum, Scleria sp.

Z 29 SE

- 98 -

Pennisetum typhoides, Sorghum bicolor,

Macrotyloma uniflorum, Vigna radiata,
Linum usitatissimum, Ziziphus sp.,
Ipomoea sp., Polygonum sp., Dactyloctenium
aegyptium, Chenopodium sp., Acacia sp.,
Barley, Chenopodium sp., Trianthema
triquetra

Annual report of excavation at Kanmer 2007-08 and 2008-09

is based on the morphological details preserved in

the dominant dry thorny vegetation.

the carbonized grains and seeds by comparing them

with the corresponding parts of extant plants of the

marina, Mysistica malabarica and Wrightia tinctoria,

same species. These remains represent the accidentally

which may have been used for some activities other

scorched portion of a larger amount.

than fuelling their hearths. For example, Lancelotti

In the first field season several cereals and

and Madella suggest that the presence of Wrightia at

pulses such as barley (Hordeum vulgare), bread-

the site can be associated with dyeing activities as its

wheat (Triticum aestivum), dwarf-wheat (Triticum

leaves produce a blue dye similar to Indigo. With their

sphaerococcum), rice (Oryza sativa), field-pea (Pisum

preliminary observations they have postulated that

arvense), grass-pea (Lathyrus sativus), and green-

intensive exploitation of dry scrubland and riverine

gram (Vigna radiata) were identified, besides cotton

species, combined with increasing aridity may have

(Gossypium arboreum/herbaceum), some weeds and

led to the environmental deterioration.

wild taxa (see Kharakwal et al. 2007). In the second

field season, largely focused to understand the history

Kanmer given in Tables 3 and 4.

of fortification wall, we could retrieve number of

charred grains of jowar-millet (Sorghum bicolor),

different cultural period and chronology at Kanmer.

Pearl millet (Pennisetum typhoides), sesame (Sesamum

To our surprise there are three dates at our disposal

indicum), and horse-gram (Macrotyloma uniflorum).

from the upper levels of Kanmer, which fall around

Besides, remains of weeds and other wild taxa were

4th millennium BCE. However, these early dates

also encountered reflecting the ecological conditions

may be errors of some kind or due to contamination

and ground vegetation. Table 2 presents summary

as there is no indicator in the archaeological remains

results of samples so far analysed by Pokharia,

which points to such early dates.

They have also identified species such as Avicennia

We have so far received 27 radiocarbon dates for

These dates appear to support our identification of

belonging to third and fourth field season.

Besides charred grains, Carla Lancelotti studied

CONCLUSION

a few samples of charcoal from Kanmer in order to

understand the fuel system of the Harppans with an
idea if the shortage in fuel accelerated the decline of

In fact out of 555 known Harappan and Chalcolithic

the mighty Harappans (Lancelotti and Madella in

settlements in Gujarat as many as 63 are located in

this volume). They have identified a large number

Kachchh (Possehl 1999; Seth et al. 2007; Rajesh and

of woody and grass plant remains belonging to dry

Patel 2007). Among them, only a few sites, such

scrubland and riverine species. They have pointed out

as Dholavira (IAR 1967-68: 14-16; Bisht 1989a, b,

that a variety of wood species flourished at Kanmer

1991, 1997), Surkotada ( Joshi 1990) and Juni Kuran

during KMR II or urban phase. The most abundant

(Pramanik 2003-2004), have been subjected to

species identified by them are Ziziphus, Salvadora

horizontal excavation whereas Desalpar, Pabhumath,

and Capparis, of course belonging to dry scrub

Sikarpur were subjected to small scale excavation.

land. On the basis of their charcoal evaluation and

Of course Shikarpur is now being re-excavated by

phytoliths studies, they suggest that there may have

M.S. University of Baroda. Sikarpur is also a small

been watercourse near the site which supported good

settlement like Kanmer and has a similar fortified

vegetation cover and coast line was much closer to the

enclosure, of course made of mud bricks. Besides,

site. According to them, the Harappans were able to

Khirsara, located in western Kachchh is under

exploit riverine specie and mangrove forest, besides

excavation by the Archaeological Survey of India.

- 99 -

J.S. Kharakwal et al.

Table 3

14

C dates of Kanmer by the Birbal Sahni Institute of Palaeobotany

no Trench

Depth
(cm)

Layer Lab Ref. No.

BSIP

14

C date
(yrs BP)

Calibrated date Calibrated date Sample from cultural

(BP)
(BC/AD)
deposit

1
2
3
4
5

930
466
280
224
350-380

19
7
3
2
6

419080
3870100
361090
81090
125080

4838-4573
4344-4262
4080-3735
726-722
1277-1062

Y-30
R-21
HH-30
GG-31
S-21

BS-2619
BS-2627
BS-2628
BS-2692
BS-2618

2888-2623 BC
2470-2149 BC
2130-1785 BC
1224-1239 AD
673-888 AD

KMR I
KMR II B
KMR II A
KMR V
KMR IV

Table 4 AMS radiocarbon dates of samples from Kanmer by Paleo-Labo, Japan

no

Layer/
Depth
Trench (cm)/
Cultural
period

PLD No.

13C

central
trench/-

3-5/
-/
KMR V

PLD-13115

-25.240.18

170816

central
trench/-

3-5/
-/
KMR V

PLD-13116

-22.430.22

cenral
trench/-

7/
-/
PLD-13117
KMR III

central
trench/-

8/
-/
PLD-13119
KMR IIB

Conventional Conventional
radiocarbon age radiocarbon age
(not rounded)
(rounded)

Caribrated age
1 range

2 range

171015

263AD(14.4%)277AD
330AD(53.8%)383AD

257AD(28.7%)301AD
317AD(66.7%)395AD

170817

171015

263AD(14.6%)278AD
330AD(53.6%)383AD

257AD(29.0%)301AD
317AD(66.4%)395AD

-23.240.18

173917

174015

255AD(10.0%)265AD
273AD(58.2%)335AD

242AD(93.0%)351AD
369AD( 2.4%)379AD

-24.100.12

393118

393020

2474BC(33.0%)2454BC
2419BC(13.2%)2406BC
2377BC(22.0%)2351BC

2479BC(95.4%)2345BC

central
trench/-

8/
-/
PLD-13120 -24.820.14
KMR IIB

370118

370020

2135BC(14.9%)2119BC
2096BC(53.3%)2040BC

2188BC( 1.0%)2184BC
2142BC(94.4%)2031BC

central
trench/-

8/
-/
PLD-13121
KMR IIB

-24.310.13

366119

366020

2121BC(26.0%)2095BC
2041BC(27.1%)2016BC
1996BC(15.1%)1980BC

2133BC(35.2%)2083BC
2057BC(60.2%)1960BC

central
trench/-

11/
-/
PLD-13122
KMR IIA

-24.150.12

389518

389520

2459BC(42.3%)2401BC
2382BC(25.9%)2348BC

2466BC(88.3%)2335BC
2324BC( 7.1%)2301BC

NE
6/
trench/
363/
PLD-14748 -25.080.25
AA17 KMR III

378025

2278BC(22.7%)2251BC
2229BC( 6.0%)2221BC
2211BC(13.6%)2195BC
2176BC(25.9%)2145BC

2292BC(95.4%)2137BC

SE
3/
trench/
385/
PLD-14749 -25.500.28
FF29 KMR III

375025

2204BC(62.6%)2134BC
2077BC( 5.6%)2064BC

2279BC( 7.2%)2250BC
2230BC( 1.5%)2220BC
2211BC(68.2%)2121BC
2095BC(18.5%)2041BC

378226

375127

- 100 -

Annual report of excavation at Kanmer 2007-08 and 2008-09

no

Layer/
Depth
Trench (cm)/
Cultural
period

10

SE
3/
trench/
385/
PLD-14750 -25.540.21
FF29 KMR III

11

central
trench/
Z30

PLD No.

13C

Conventional Conventional
radiocarbon age radiocarbon age
(not rounded)
(rounded)

Caribrated age
1 range

2 range

376924

377025

2271BC( 8.0%)2259BC
2206BC(60.2%)2141BC

2287BC(92.7%)2133BC
2079BC( 2.7%)2061BC

-22.760.13

398424

398525

2564BC(37.9%)2534BC
2494BC(30.3%)2472BC

2571BC(56.2%)2513BC
2504BC(39.2%)2466BC

12

central
10/
trench/ 904/
PLD-14752 -24.260.13
Z30 KMR IIA

393524

393525

2476BC(35.6%)2437BC
2421BC(12.4%)2404BC
2379BC(20.2%)2349BC

2560BC( 3.2%)2536BC
2491BC(92.2%)2341BC

13

SE
4/
trench/
458/
PLD-14753 -24.530.24
EE30 KMR IIB

368224

368025

2132BC(44.8%)2084BC
2058BC(23.4%)2030BC

2141BC(89.2%)2011BC
2001BC( 6.2%)1977BC

west
11/
14 trench/ 800/
PLD-14754 -25.920.29
Q28 KMR IIA

389625

389525

2462BC(42.4%)2398BC
2384BC(25.8%)2346BC

2467BC(95.4%)2299BC

15

west
12/
trench/
820/
PLD-14755 -26.070.14
Q28 KMR IIA

386625

386525

2455BC(16.2%)2420BC
2406BC(15.5%)2377BC
2350BC(36.5%)2291BC

2464BC(90.4%)2281BC
2250BC( 3.9%)2231BC
2219BC( 1.1%)2212BC

16

central
trench/
Z28

PLD-14756 -25.450.20

387324

387525

2455BC(19.0%)2419BC
2406BC(17.3%)2377BC
2350BC(31.9%)2296BC

2465BC(94.9%)2286BC
2247BC( 0.5%)2243BC

17

cenral
9/
trench/
788/
PLD-14757 -11.410.22
AA28 KMR IIA

380923

381025

2286BC(68.2%)2205BC

2337BC( 2.2%)2323BC
2308BC(86.7%)2196BC
2172BC( 6.5%)2146BC

18

west
13/
trench/
845/
PLD-14758
Q28 KMR IIA

-21.250.18

389224

389025

2459BC(42.1%)2391BC
2385BC(26.1%)2346BC

2466BC(95.4%)2298BC

19

west
11/
trench/
809/
PLD-14759 -25.290.14
Q28 KMR IIA

389822

390020

2462BC(42.8%)2402BC
2382BC(25.4%)2348BC

2468BC(87.9%)2335BC
2324BC( 7.5%)2301BC

central
20 trench/
AA28

12/
972/
KMR I

12/
891/
KMR I

PLD-14751

8/
711/
IIB

PLD-14760 -22.330.15

387523

387525

2455BC(20.0%)2419BC
2406BC(17.7%)2377BC
2351BC(30.5%)2298BC

2464BC(95.4%)2288BC

west
trench/
Q28

10/803

PLD-14761 -25.680.17

388627

388525

2457BC(68.2%)2342BC

2466BC(95.4%)2292BC

central
22 trench/
AA28

12/924

PLD-14762 -24.220.20

381424

381525

2288BC(18.5%)2267BC
2260BC(49.7%)2206BC

2344BC(90.6%)2196BC
2172BC( 4.8%)2146BC

21

- 101 -

J.S. Kharakwal et al.

Most of these settlements belong to urban and post

during our archaeological diggings and the results

urban phase of the Harappan Civilization. The

of digital recording indicates that the shape of the

concentration of sites appears to be in the eastern part

fortified settlement was like a parallelogram. The total

of the peninsula, which is believed to be an arm of sea

thickness of the wall was measured about 18m in the

in the past (Gupta and Pandya 1980). It is held that

middle level of its height.

the accessibility of marine resources, semiprecious

stones and arable land were major factors for the

KMR III were found overlying the remains of KMR

expansion of enterprising Harappans in Gujarat.

II. The pottery assemblage of KMR III generally

Though researches on the Harappans in Gujarat have

appears ochreous. It is likely that the Late Harappans

unfolded various new facets, e.g., elaborate water

also used the existing rampart initially but did not

structures, dock yard (!), diverse craft industries,

maintained it as it was not required then and finally

regional variation in architecture, ceramic and so

they raised their houses right on top of the rampart.

on, there are still various challenging issues yet to be

Thus there was no cultural discontinuity throughout

addressed, such as function of strongly fortified small

the deposit of Harappan vintage.

sites, decline/ de-urbanisation of the Harappans, role

of non Harappans cultures in civilisation process.

remained unoccupied for about 1600 years. This long

The earliest settlement (KMR I) was raised on a

period of time was represented by a thick layer of dark

thin soil cover overlying the bed rock, which may also

grey sandy soil, often loose in composition and devoid

be stated as pre-fortification deposit. It was identified

of any structural remains. It was identified as layer

on the basis of presence of Anarta types coarse Red

6 in the central part whereas layer 2 in the northern

Ware, bichrome and polychrome types, cream slip

area and layer 3 in the western part. The Historic

pottery, Red ware (sometimes decorated with a panel

structures, belonging to KMR IV, were raised on

of grooved decoration on the shoulder or body part)

this break. The discovery of large number of sherds

together with some Harappan type fossils. Except for

of West Asian Tarpedo Jar and Roman Amphorae

Coarse Red, all other types gradually disappear in the

indicate that Kanmer, directly or indirectly, was also

subsequent cultural deposit. The urban phase deposit

included in the Indo-Roman trade network. The site

has been levelled as Period KMR II, which is further

seems to be again deserted for a few hundred years

sub-divided in two parts Period KMR IIA and

after the Historic period. During this long spell of

Period KMR IIB on the basis of appearance of new

abandonment the site again got deposition of fine

ceramic types, nature of deposit, change in planning

silt mixed with clay indicating that Aeolian soil was

of settlement at the site. It seems during this period

getting deposited in the large depression surrounded

a planned settlement was built at Kanmer, which

by the ruins of the fort and subsequently getting

was secured with a massive and strong fortification.

reworked by the rain..

An addition was made all along the outer arm of

the fort wall during period KMR I perhaps to check

like a lower town could be located. Of course there

some serious damage to the wall. It was identified as

is a very thin deposit of Late Harappan period to

the second major structural phase of the fort wall.

the east of the main mound and the laser survey

Towards the end of period KMR II the height of

has indicated an outer defence wall like feature to

fort was further raised but its width was reduced.

the east of the main mound. From the deposit of

It was identified as the third or last constructional

the Mature and Late Harappan phase considerable

phase of the fort wall. The lay out plan documented

quantity of nodules of agate have been recovered.

- 102 -

The remains of post urban phase identified as

After the decline of the Harappans the site

Outside the rampart area no other settlement

Annual report of excavation at Kanmer 2007-08 and 2008-09

These were perhaps brought from Mardhak Bet, an

Project sincerely thanks Archaeological Survey of

island, located in the Little Rann about 20 km to the

India for granting permission to continue further

northeast of Kanmer. It is likely that this raw material

studies at the site. We thank various authorities

was used for making variety of beads, blades and

of JRN R ajasthan V idyapeeth, Udaipur, and

weights. The faunal and floral remains clearly indicate

Government of Gujarat for encouraging us and

that agriculture and cattle breeding were among the

providing administrative support for field research in

important occupation at the site right from the third

Kachchh. The authors would like to express sincere

millennium.

gratitude to Profs. D.P. Agrawal, R.S. Bisht, R.S.

From the early levels at Kanmer, i.e. pre-

Fonia, Ajithprasad and Randle Law for enlightening

fortification deposit nodules of agate, flakes, blades,

on various issues related to Kanmer. We would also

and rough-outs were discovered. In the second

like to record special thanks to Prof. B.S. Garg, Prof.

cultural period, i.e. KMR II (contemporary to urban

Divya Prabha Nagar of Rajasthan Vidyapeeth for

phase) a large number of drill bits, rough outs, variety

constantly supporting for analysis of the excavated

of beads of semi-precious stone were discovered. One

material in the institute at Udaipur. The authors also

may argue that the local rural non Harappan folks or

thank Mr. Hansmukh Seth, Rajesh Meena, Suresh

the early Harappans knew the source of raw material.

Meena, Asif Hussain, K.P. Singh, Narayan Paliwal,

The aforesaid nature of discovery indicates that they

Rohit Menaria, Soyeb Qureshi and Miss Deepshikha

might be aware of the technology of bead making

for lending us helping hand in the institute while

too. The Harappans, being enterprising people, may

working on the excavated material.

have realized the importance of source of raw material

and bead making process. It is therefore likely that
they may have begun the marketing or raw material
as well as finished products. Besides this, a large

References

number of faience beads have been discovered from

Agrawal, D.P., Kharakwal, J.S., Y.S. Rawat, T. Osada, and

the site and many among them are un-oxidized. It is

P. Goyal (2010) Redefining the Harappan Hinterland.

likely that they were also supplying faience otherwise

Antiquity vol. 84 (323) on line publication.

they would not buy un-oxidized faience. Perhaps to

Agrawal, D.P. and J.S. Kharakwal (2003). Bronze and Iron

protect the resource or raw material and the craftsmen

Ages in South Asia. Aryan Books International, Delhi.

the Harappans required a strong protection. This

Ajithprasad, P. (2002) The Pre-Harappan Cultures

may be a possible answer to understand why they

of Gujarat, in S. Settar and Ravi Korisettar (eds.)

spent so much of energ y to build massive fort at

Protohistory: Archaeolog y of the Indus Civilization.

Manohar Publication, Delhi. pp.129-158.

Kanmer for their miniscule settlement. The discovery

Bhan, K.K. (1994) Cultural development of the prehistoric

of seals, seal impressions , sealing suggest that they

period in North Gujarat with reference to Western India.

were involved in trade. There are many other fortified

South Asian Studies 10: 71-90.

sites in Kachchh, which may have played similar role.

Bisht, R.S. (1991) Dholavira: new horizons of the Indus

Thus it may be stated that the Harappans of Kanmer

civilization. Puratattva 20: 71-82.

were engaged in some kind of resource management

Bisht, R.S. (1989a) A new Model of the Harappan town

(Agrawal et al. 2010).

planning as revealed at Dholavira in Kutch: a surface

study of its plan and architecture, in Bhaskar Chatterjee

Acknowledgements

(ed.) History and Archaeolog y (Prof. H.D. Sankalia

The entire team of Kanmer Archaeological Research

felicitation volume). Ramanand Vidya Bhawan, Delhi.

- 103 -

J.S. Kharakwal et al.

pp.397-408.

and Nature, Kyoto. pp.21-46.

Bisht, R.S. (1989b) The Harappan colonisation of Kutch:

Kharakwal, J.S., Y.S. Rawat and T. Osada (2008) Preliminary

an ergonomic study with reference to Dholavira and

observations on the excavation at Kanmer, Kachchh,

Surkotda, in Krishna Deva and Lallanji Gopal (eds.)

India 2006-07, in T. Osada and A. Uesugi eds.)

History and Art. Ramanand Vidya Bhawan, Delhi. pp.

Occasional Paper 5. Research Institute for Humanity and

265-72.

Nature, Kyoto. pp.5-24.

Bisht, R.S. (1997) Dholavira excavations: 1990-1994, in

Kharakwal, J.S., Y.S. Rawat and T. Osada (2009) Excavation

J.P. Joshi (ed.) Facets of Indian Civilization: Recent

at Kanmer: A Harappan site in Kachchh, Gujarat.

Perspectives. Aryan Books International, New Delhi. pp.

Puratattva 39: 147-164.

107-120.

Lancelotti, C and M. Madella (2010) Fuelling Harappan

Casal, J.M. (1964) Fouilles DA mri, vol. 1. Librairie C.

Hearths: Fuel exploitation and use during the Mature

Klincksieck, Paris.

Harappan phase at Kanmer (Kachchh, Gujarat, India),

Dikshit, M.G. (1950) Excavation at Rangpur 1947. Bulletin of

Deccan College Research Institute 11(1): 3-55.

in this volume. pp.129-142.

Misra, V.N., V.S. Shinde, R.K. Mohanty, Lalit Pandey and

Ghurye, G.S. (1939) Two sites in Kathiawar. Journal of the

Jeewan S. Kharakwal. (1997) Excavations at Balathal,

University of Bombay 8 (1): 3-12.

District Udaipur, Rajasthan (1995-97), with special

reference to Chalcolithic Architecture. Man and

Gupta, S.K. and S. Pandya (1980) Kutch Harappans-their

communication routes- a discussion. Vidya 23 (1): 39-46.
IAR: Indian Archaeology - A Review. Annual Publication of the

Environment 22(2): 35-59.

Possehl, G.L. (1992) The Harappan Civilization in Gujarat:

Archaeological Survey of India, New Delhi.

the Sorath and the Sindhi Harappans. The Eastern

Anthropologist 45 (1-2): 117-154.

Joglekar, P.P. (2007) Report of the faunal remains recovered

from Kanmer, Kachchh, Gujarat, during the first

Possehl, G.L. (1999) The Indus Age. Oxford, Delhi.

season (2005-06), in T. Osada (ed.) Occasional Paper

Pramanik, Shubhra (2004) Excavation at Juni Kuran: 20032004 a preliminary report. Puratattva 34: 45-67.

2. Research Institute for Humanity and Nature, Kyoto.

pp.47-76.

Rajesh, S.V. and Ambika Patel (2007) Gazetteer of Pre and

Protohistoric sites in Gujarat. Man and Environment 32

Joshi, J.P. (1972) Exploration in Kutch and excavation at

Surkotada and new light on Harappan migration. Journal
of the Oriental Institute 22 (1-2): 137-143.

(2): 61-136.
Rao, S.R . (1962-63) Excavations at Rangpur and other

Joshi, J.P. (1990) Excavation at Surkotada 1971-72 and

Exploration in Kutch. Archaeological Survey of India,

explorations in Gujarat. Ancient India 18-19: 5-207.

Seth, H., L.C. Patel and B. Varhat (2007) Harappan sites in
Gujarat, in T. Osada (ed.) Occasional Paper 2. Research

New Delhi.
Kharakwal, J. (2005) The Indus Civilization: an overview,
in T. Osada (ed.) Occasional Paper. Research Institute for

Institute for Humanity and Nature, Kyoto. pp.111-126.

Sonawane, V.H. and P. Ajitprasad (1994) Harappan Culture
and Gujarat. Man and Environment 19 (1-2): 129-39.

Humanity and Nature, Kyoto. pp.41-85.

Kharakwal, J.S., Y.S. Rawat and T. Osada (2005) Harappan

Tomber, R. (2007) Rome and Mesopotamia importers into

sites in Kachchh and new opportunities of tourism, in

India in the first millennium AD. Antiquity 81 (2007):

Heritage Tourism. Government of Gujarat. pp.35-43

972988

Kharakwal, J.S., Y.S. Rawat and T. Osada (2007) Kanmer: A

Vats, M.S. (1935) Trial excavation at Rangpur, Limdi State,

Harappan sites in Kachchh, Gujarat, India, in T. Osada

Kathiawar. Annual Reports of Archaeological Survey of

(ed.) Occasional Paper 2. Research Institute for Humanity

India 1934-35: 348.

- 104 -

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

Report on the faunal remains recovered from

Kanmer, Gujarat, during the third field season (2007-08)
Pankaj Goyal and P.P. Joglekar
Department of Archaeology
Deccan College Post Graduate and Research Institute, Pune, India

ABSTRACT
The three seasons of excavation at Kanmer revealed a large number of animal skeletal remains. This paper presents the results of the
analysis carried on the excavated faunal material during the third season (2007-08) from Kanmer. More than 25,000 bone fragments
were collected during this season at Kanmer; however, this report is based on a part of this collection mainly the material coming
from two trenches Z29 and Z30 located in the central part of the mound. A total of 2984 fragments were analyzed from these two
trenches belonging to different cultural periods. The bone identification revealed the presence of both domestic and wild mammals.
The domestic species include cattle, sheep, goat, pig, horse (found only in the Medieval Phase-KMRV), ass and cat. The wild mammals include wild buffalo, nilgai, deer (Sambar, spotted deer, barking deer and hog deer), antelopes (blackbuck, gazelle and fourhorned antelope), pig, porcupine, hare, elephant and carnivores (wolf and jungle cat) and house rat. The non-mammalian species
identified are domestic fowl, peafowl, pond heron, monitor lizard, Chitra turtle and Indian sawback turtle. Any signature of bone
modifications observed is also recorded. So far, information regarding the change in the animal based subsistence pattern from the
Mature Harappan to the Late Harappan phase is virtually unknown. It is on this background the faunal remains unearthed from
Kanmer are important.

Introduction

whereas the northeastern and southwestern corners

form acute angles. The site is very close to the well-

Kanmer, locally known as Bakar Kot is an impor-

known Harappan sites like Surkotada and Dholavira

tant addition to the list of Mature Harappan sites in

and might have played an important role as well. Ex-

Gujarat (Figures 1 and 2). The ancient site was first

cavations have been going on at Kanmer for last three

discovered by R.S. Bisht, Y.S. Rawat and other Ar-

years under the supervision of J.S. Kharakwal, Y.S.

chaeological Survey of India (ASI) team members in

Rawat and Toshiki Oshada. (Kharakwal et al. 2005;

1985. The site has an impressive mound that measures

2007; 2008a and b; 2009)

about 110 m E-W and 120 m N-S. The total thickness

of the deposit is 11 m. The stratigraphy of the periph-

vated from the site of Kanmer allowed the excavators

eral area of the site is different from that of the central

to identify the following fivefold cultural sequence at

part of the mound. The site revealed a very large forti-

the site:

Study and analysis of the cultural material exca-

fication wall. The excavation and GPS and GIS survey

conducted at the site showed that this fortification

Period I:

KMR I

is a parallelogram. The northwestern and southeast-

Period II:

KMR II

ern corners of this fortification make obtuse angles

Period III: KMR III

- 105 -

P. Goyal and P.P. Joglekar

Figure 1 Map showing Kanmer with other Harappan sites in Gujarat

Figure 2 A general view of the mound at Kanmer

- 106 -

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

Period IV:

KMR IV

different phases of Harappan occupation at the site.

Period V:

KMR V

This report deals with the faunal remains un-

earthed at Kanmer in the third season of excavation

KMR stands for Kanmer and KMR I, II and III

(2007-08). The previous two reports dealing with the

represents the different phases of Harappan occupa-

faunal material recovered at Kanmer in 2005-06 and

tion at the site. This terminology is preferred by the

2006-07 have already been published ( Joglekar 2007;

excavators as the total Harappan deposit represents

Goyal and Joglekar 2008). The authors have also pre-

different stages of culture. KMR IV represents the

sented research papers connected with the results of

Early Historic deposit, whereas KMR V the Medieval

the faunal analysis done so far in different conferences

one. In addition, two cultural breaks have also been

( Joglekar 2006; Joglekar and Goyal 2008; Goyal and

noticed. The first break was noticed between KMR

Joglekar 2008; 2010a; 2010b).

III and IV represented by layer 6 of trenches Z29 and

Z30. This layer, dark grayish in colour, revealed no

Methodology

structural activity and yielded mixed types of pottery

(Historic and Late Harappan). Therefore, it has been
designated as a break between the Historic and the

The process of retrieval of the faunal material at the

Late Harappan phase (Kharakwal et al. 2008). The

site was supervised by the first author. It is widely rec-

other break was found between KMR IV and KMR

ognized that the bones of small animals are frequently

V represented by layer 4 of the same trenches. How-

missed during excavation (Payne 1975). Therefore,

ever, no such break has been recognized between the

utmost care was taken to recover all the micro-faunal

Figure 3 Cleaning of the faunal material

- 107 -

P. Goyal and P.P. Joglekar

Figure 4 A top view of trenches Z29 and Z30

Figure 5 Faunal material in situ

- 108 -

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

remains. The recovered faunal material was washed

den Driesch (1976).

and dried in the shade at the site itself (Figure 3).

After cleaning and drying, the bones were packed

were identified to the exact anatomical area and an

first in the paper bags and then in cloth bags with

attempt has been made to properly record all types

proper context written on each bag. The micro-faunal

of cut marks. Identifiable specimens were compared

remains have been packed separately in small plastic

with the modern specimens of known species housed

boxes with proper labelling to avoid their breakage

at the Deccan College Archaeozoology Laboratory at

during the transportation.

Pune and Maharaja Sayajirao University of Baroda. In

Trenches Z29 and 30 were taken up for this study

addition, help from the published literature was taken

since these trenches represent all the cultural phases in

like Joglekar et al. (1994), Miguad (1992), Gupta et. al.

a well-stratified context (Figures 4 and 5). A prelimi-

(1987, 1990), Higham (1975) and Schmidt (1972), etc.

nary study of the material was conducted at the site

Epiphyseal fusion and tooth wear stages have been

itself and only the remaining skeletal fragments have

recorded to find out kill off pattern in the case of do-

been taken to the Deccan College Archaeozoology

mestic animals such as cattle, sheep, goat and pig.

Laboratory for further processing. At first, an initial

sorting was done and fragments were categorized into

and very difficult to differentiate if suitable morpho-

identifiable and non-identifiable parts. The identifi-

logical markers are not present. Therefore, fragments

able parts were then identified up to a specific skeletal

of ribs, vertebrae and other very fragmented bones

element and whenever possible to the species level.

not containing any morphological markers were

The bone fragments that could not be identified were

recorded in the category of Bos/Bubalus. The same is

termed as the unidentified material (UF) and were

done for sheep and goat and a category called Capra

further categorized into three different types, i.e. large

hircus/Ovis aries was created. To attain a fine level

unidentified fragments (UL), medium unidentified

identification of sheep and goat published literature

fragments (UM) and small unidentified fragments

like Prummel and Frisch (1986) and Pawankar and

(US) based on the relative sizes of the skeletal frag-

Thomas (2001) was consulted. In addition, categories

ments. Broken skeletal fragments of a same element

like small mammal, medium mammal, general deer,

were refitted and considered as a single unit following

general reptile, small sized birds and medium sized

Clasons (1972) method of counting the specimens.

birds were also created.

The skeletal fragments containing cut marks

The bones of cattle and buffalo are very similar

No attempt of refitting was done if the fragments belonged to different stratigraphic units. Criteria given

The Faunal Material

by Sadek-Kooros (1975) to describe intentional fracturing were used. Whenever possible, measurements

The present report is based on the faunal material

of the recovered material were taken following Von

Table 1 Details of faunal material examined

Phase

Contexts

NISP

UF

TF

NISP %

KMR V-The Medieval Phase

Z29 (3), Z30 (3)

25

29

54

46.29

KMR IV-The Historical Phase

Z29 (5), Z30 (5)

315

206

521

60.46

KMR III-The Late Harappan Phase

Z29 (7), Z30 (7)

159

124

283

56.18

KMR II-The Mature Harappan Phase

Z29 (8,9,10, 11)

1317

809

2126

61.91

1816

1168

2984

60.85

Total

- 109 -

P. Goyal and P.P. Joglekar

recovered from two trenches Z29 and Z30 (Figure

(n=1168) (Figure 6). Bones belonging to both axial

4) located in the central part of the mound. A total

and appendicular skeleton were present in the faunal

of 3020 fragments were analyzed belonging to four

assemblage; however, in case of all the animals, the

cultural periods - Mature Harappan Phase (KMR

axial skeletal parts were scantly represented in all the

II), Late Harappan phase (KMR III), the Historic

cultural phases. Maximum representation was of ribs

Period (KMR IV) and the Medieval Period (KMR V)

and vertebrae. Phalanges, astragalus, calcanium, carpal

(Table 1). The material (n=36) belonging to the first

and tarsal bones were well-preserved and mostly com-

two layers of trenches Z29 and Z30 were not included

plete.

in the analysis as this layer represented the modern

soil cover. The layer 1 did not contain any faunal ma-

The Species

terial except a second molar of Bos indicus (K1319)

and few unidentified skeletal fragments. A handful
of fragments recovered from the break between the

The faunal material analysed this year revealed pres-

Medieval (KMR V) and the Historic Phase (KMR

ence of a large number of species (Tables 2 - 3, Figures

IV) were not analyzed. Similarly, the cultural break

7 - 10) that include mammals, birds and reptiles. The

recognized between the Historic (KMR IV) and the

material also revealed a large number of fish fragments

late phase of Harappan culture at the site (KMR III)

and a small number of molluscan species; however,

revealed a few fragments of bones and was not in-

these were not included in the present report. These

cluded in this analysis.

fragments have been kept separately and would be

The other layers revealed 2984 skeletal elements,

analysed independently. Nine species of domestic

out of which 1816 (60. 85%) could be identified (Table

mammals have been identified this year the cattle,

1). The bone fragments that could not be identi-

buffalo, sheep, goat, pig, horse, ass, dog and cat.

fied were termed as the unidentified material (UF)

The wild mammals include 16 species the wild

Figure 6 Chart showing faunal material from Kanmer

- 110 -

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

Figure 7 Bones from the Mature Harappan context: a. First phalanges of cattle, b. Buffalo bones,
c. Antilope cervicapra bones, d. First phalanges of Gazella bennetti

Figure 8 Bones from the Mature Harappan context:

a. A complete humerus of Axis axis, b. Sus scrofa bones, c. Bird bones, d. Reptile bones
- 111 -

P. Goyal and P.P. Joglekar

Figure 9 Bones from the Historic context: a. Bos indicus bones (from Medieval context), b. Second phalanges of buffalo,
c. A. cervicapra bones (first two); Gazella bennetti bones (last two), d. Humerus of dog

Figure 10 Bones from the Late Harappan context: a. Capra hircus bones (A-C); Scapula of Ovis aries (D), b. Astragalli of Axis axis (AB); Tibia of Antilope cervicapra (C) and Tibia of Gazella bennetti (D), c. Lepus nigricollis bones (A-D), d. Lepus nigricollis bones (A-C)
- 112 -

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

buffalo, nilgai, elephant, antelopes (blackbuck, gazelle

information about the way in which these fragments

and four horned antelope), deer (spotted deer, hog

were discarded/deposited. Most of the bone frag-

deer, sambar and barking deer), wild pig, wolf, jungle

ments were well-preserved; however, bones belonging

cat, porcupine, hare and house rat. There was also a

to the layers located deep in the trenches (layers 9-11)

species of equid that could not be identified with cer-

were impregnated and encrusted with minerals. The

tainty. Similarly, a species of deer could not have been

effects of this varied noticeably from fragment to frag-

identified up to species level. The faunal material also

ment, but almost all fragments were affected to some

revealed the presence of birds and reptiles. However,

degree. Layer 5 revealed a large number of skeletal

it was not possible to identify many such fragments.

fragments in which flaking of the outer crust could

The identified birds include domestic fowl, peacock

be seen. This can be attributed to their exposure to

and pond heron, while the reptiles include monitor

Sun for a long period of time. In case of a few bone

lizard and two species of turtle. These fragments have

fragments from the same layer, root impression was

been kept separately for further examination in the

noticed on them.

laboratory.

A number of bones revealed the marks of modern

breaking. For example, the distal end of first phalanx

of Axis axis recovered from the Mature Harappan

Bone Modifications

phase (layer 10) showed the evidence of modern

breaking. Similarly, a skeletal fragment belonging to

A careful recording of bone modifications (Table 4) is

the radius of Antilope cerviapra (layer 9) were broken

very important as these can provide us with some vital

into two fragments. Some of these fragments were

Figure 11 A. Bone modifications; B. Bone modifications; C. Bone tools; D. Bone object

- 113 -

P. Goyal and P.P. Joglekar

Figure 12 Chart showing bone modifications observed at Kanmer

broken during excavation, while some other broke

charring noticed. A number of such fragments were

during their transportation to the laboratory at

found from all the cultural phases revealed; however,

Deccan College. The bones like carpals, tarsals and

the Mature Harappan phase revealed the maximum

phalanges were better preserved, and very few cases

number of charred fragments. Some peculiar bone

of bone modifications were noticed on these bones,

fragments charred at the anterior end only were no-

which can be attributed to their non meat-bearing

ticed in the Mature Harappan phase last year (Goyal

character.

and Joglekar 2008); however, no such fragment was

No cases of embryonic bones getting associated

noticed this time. Butchering and cutting marks have

with the faunal assemblage have been noticed this

also been noticed from all the periods. Cutting and

year. Sixteen cases of porous bones were noticed

butchering marks are produced during various stages

this season, 2 belonging to the Medieval phase, 2 to

of carcass utilization and a proper recording of these

the Historical phase and 12 to the Mature Harap-

marks can throw light on various aspects of food

pan phase. Few cases of bone gnawed by rodents

preparation and food processing activities.

and carnivores were also noticed. A large number

of bones having charring marks were noticed from

served from Kanmer which seemed to be used as tools

all the cultural phases. These charred fragments have

or made for some other purpose. The Mature Harap-

been recorded as three different types, i.e. charred,

pan yielded a total of five such fragments. All these

completely charred and charred and turned white

were made on bones of cattle/buffalo and sheep/

(calcified) based upon the relative proportion of

goat, mainly the metapodia. For example, an object

- 114 -

A small number of modified bones have been ob-

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

Table 2 Animal species identified at Kanmer in the third seasons material

Domestic mammals

Wild mammals

Non-mammals

Bos indicus (Cattle)

Bubalus arnee (Wild buffalo)

Gallus domesticus (Domestic fowl)

Bubalus bubalis (Buffalo)

Boselaphus tragocamelus (Nilgai)

Ardeola grayii (Pond heron)

Capra hircus (Goat)

Tetracerus quadricornis (Chowsingha)

Pavo cristatus (Peafowl)

Ovis aries (Sheep)

Antilope cervicapra (Blackbuck)

Sus domesticus (Pig)

Gazella bennetti (Gazelle)

Canis familiaris (Dog)

Axis axis (Spotted deer)

Equus caballus (Horse)

Axis porcinus (Hog deer)

Equus asinus (Ass)

Cervus unicolor (Sambar)

Felis catus (Cat)

Muntiacus muntjak (Barking deer)

Varanus bengalensis (Monitor lizard)

Kachuga tecta (Indian sawback turtle)
Chitra indica (Chitra turtle)

Sus scrofa (Wild pig)

Canis lupus (Wolf )
Felis chaus ( Jungle cat)
Hystrix indica (Porcupine)
Elaphus maximus (Elephant)
Rattus rattus (House rat)
Lepus nigricollis (Hare)

made on the metapodium (K770), recovered from

buffalo bones in all the layers belonging to the Mature

layer 9 has a polish on its outer surface. Similarly, a

Harappan phase. Similarly, the bones of goats were

metacarpal (K773) of cattle revealed sharp cuts at the

more in number than sheep.

proximal end and was polished on both sides. One

such fragment was noticed in the Medieval phase. The

finding. This layer brought to light a number of iden-

exact function of these modified bones is yet to be

tified skeletal fragments belonging to very young cat-

determined.

tle (Table 5). There were many more fragments which

Layer 9 of trench Z30 revealed a very interesting

seem to be of the same cattle. As these fragments were

not having any identifiable morphological features,

KMR II:
The Mature Harappan Phase

these were grouped in the category of unidentified

fragments (UF). Measurements of only few of these
could be taken due to their badly preserved condition.

The bulk of the faunal material examined from two

These skeletal elements were recovered just outside a

trenches belonged to the Mature Harappan phase,

large Mature Harappan structure. It is also important

and the number of species identified is also large. A

to mention that the density of bones outside struc-

total of 1317 fragments were identified out of 2126

ture was much higher than inside the structure. It

fragments recovered from the Mature Harappan

does not seem that this animal was used for food as

phase (Table 1). The contribution of domestic animals

not a single bone show any mark of human activity

is very large in the total assemblage. The domestic ani-

(cut marks and charring marks). On the other hand,

mal contributed 87.15% of which a major part comes

it seems that there were two skeletons of very young

from the cattle and buffalo (62.65%), followed by goat

cattle, which were disturbed due to post-depositional

and sheep (24%). The cattle bones were more than the

factors.
- 115 -

P. Goyal and P.P. Joglekar

Table 3 Summary of identified species at Kanmer

Species

Medieval Phase

Domestic Mammals

NISP

Historic Phase

NISP

Late
Harappan Phase

NISP

Mature
Harappan Phase

NISP

Bos indicus

12

28

8.89

3.77

98

7.44

Bubalus bubalis

10

3.17

0.63

11

0.84

Bos/Bubalus

13

52

189

60

72

45.3

714

54.2

Capra hircus

0.32

3.14

24

1.82

Ovis aries

0.32

0.63

10

0.76

Capra hircus/Ovis aries

20

54

17.1

34

21.4

282

21.4

Equus caballus

12

Equus asinus

0.08

Sus domesticus

0.15

Canis familiaris

0.32

0.08

Felis catus

0.3

Wild Mammals

Bubalus arnee

0.32

Boselaphus tragocamelus

0.32

0.3

Cervus unicolor

Equus sp.

0.15

Elaphus maximus

0.15

Axis axis

0.32

0.63

10

0.76

Axis porcinus

0.08

Deer general

0.08

Sus scrofa

0.23

Muntiacus muntjak

0.15

Tatracerus quadricornis

0.3

Antilope cervicapra

12

3.81

2.52

14

1.06

Gazella bennetti

1.59

1.89

18

1.37

Canis lupus

0.63

Felis chaus

0.23

Hystrix indica

0.63

Lepus nigricollis

0.95

21

13.2

17

1.29

Rattus rattus

0.32

0.63

15

1.14

Medium Mammal

0.63

0.08

Small Mammal

1.27

2.52

39

2.96

Reptiles

Chitra indica

0.08

Kachuga tecta

0.08

Varanus bengalensis

0.63

General Reptile

0.08

Birds

Gallus domesticus

0.32

0.63

0.61

Pavo cristatus

0.23

Ardeola grayii

0.63

0.46

Medium sized birds

0.63

11

0.84

Small sized birds

0.23

25

100

315

100

159

100

1317

100

Total

- 116 -

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

Table 4 Bone modifications observed in different phases

Phase
Comments
Charred Bones
Completely Charred Bones
Charred and Turned White
Butchering Marks
Cut Marks
Gnawing Marks
Carnivore Marks
Rodent Marks
Porous Bones
Bone Tool
NISP

The Medieval Phase

Total
7
0
0
0
1
0
0
0
2
1
11

The Historic Phase

% of NISP
63.64
0.00
0.00
0.00
9.09
0.00
0.00
0.00
18.18
9.09
100

Total
37
50
21
5
13
0
3
8
2
0
139

% of NISP
26.62
35.97
15.11
3.60
9.35
0.00
2.16
5.76
1.44
0.00
100

The Late Harappan

Total

The Mature

Phase
Harappan Phase
% of NISP Total % of NISP
27
57.45
73
28.74
2
4.26
41
16.14
2
4.26
4
1.57
4
8.51
12
4.72
8
17.02
87
34.25
2
4.26
0
0.00
2
4.26
9
3.54
0
0.00
11
4.33
0
0.00
12
4.72
0
0.00
5
1.97
47
100
254
100

The contribution of domestic pig was almost neg-

people as indicated by the human activity noticed on

ligible as only two bone fragments (second phalanx

the skeletal fragments. A complete humerus of Axis

and ilium part of pelvis) of this animal were found

axis (K825) was recovered from the Mature Harap-

from layer 8 and 9, respectively. No butchering and

pan phase. The jungle cat was represented in the form

charring marks had been observed on the pig remains

of three fragments only and no human activity was

excavated from the Mature Harappan phase. The

noticed on these fragments. Elephant was represented

previous reports have also revealed the same picture

in the collection by the presence of two fragments of

( Joglekar 2007; Goyal and Joglekar 2008). However,

ivory. The Mature Harappan people also exploited

looking at previous such cases at other Harappan

the small game like Indian hare (1.29%).

sites, it is certain domestic pigs were used for dietary

purpose. The other domestic animals include ass, dog

bones, two representing Chitra indica and Kachuga

and cat. The occurrence of domestic ass from the Ma-

tecta, while one could not be identified up to genus

ture Harappan suggests its use as a beast of burden.

level and termed as general reptile. Although, a large

Domestic dog was identified in the form of only one

number of bird remains could not be identified up to

fragment of mandible unearthed in layer 9. The do-

species level, it seems that they played an important

mestic cat was represented in the form of four skeletal

part in the subsistence activity of the Mature Harap-

fragments, i.e. distal portion of a humerus (K790), a

pan people. Birds contributed 2.4% of the total as-

complete calcaneum (K791), distal portion of radius

semblage. The identified birds include domestic fowl

(K1030) and distal portion of the metapodium. There

(Gallus domesticus), peafowl (Pavo cristatus) and pond

is no evidence to show that ass, dog and cat were used

heron (Ardeola grayii). A large number of these bird

for food purpose as no human activity is noticed on

fragments were having charring marks giving a clear

these fragments.

indication of their use as an item of food. A clear

The wild animals represented in this phase in-

picture of the contribution of birds will become clear

clude nilgai, deer, antelopes, wild pig, elephant, jungle

after the analysis of other trenches. No evidence of

cat and other small mammals (10.25%). Of these the

crab has been noticed so far at Kanmer, though these

most numerous were fragments of gazelle (Gazella

have been reported from a number of other Harappan

bennetti), blackbuck (Antilope cervicapra) and spotted

sites in Gujarat.

deer (Axis axis) forming 3.20% of the NISP (Table

3). These animals were included in the diet of the
- 117 -

Reptiles were identified in the form of only three

P. Goyal and P.P. Joglekar

Table 5 A detailed description of the identified skeletal elements belonging to two young cattle
Reg. No.

Skeletal Element

Portion

Side

Fusion

K902

Astragalus

Complete

Left

K903

Astragalus

Complete

Left

K911

Ulna

Proximal

Right

Unfused

K914

Matacarpal

Distal

Indeterminate

Unfused

K915

Metatarsal

Proximal

Left

Unfused

K916

Mandible

Condyle portion

Indeterminate

K918

Calcanium

Complete

Left

Unfused

K919

Pelvis

Ilium

Indeterminate

Unfused

K920

Humerus

Proximal

Left

Unfused

K921

Metapodial

Distal

Indeterminate

Unfused

K922

Astragalus

Complete

Right

K923

Femur

Proximal

Indeterminate

Unfused

K924

Femur

Distal

Indeterminate

Unfused

K925

Tibia

Proximal

Indeterminate

Unfused

K926

Humerus

Distal

Indeterminate

Unfused

K927

Calcanium

Proximal

Indeterminate

Unfused

K928

Femur

Proximal

Indeterminate

Unfused

K929

Pelvis

Ilium

Indeterminate

Unfused

K930

First Phalanx

Complete

Right

Unfused

K931

Humerus

Distal

Indeterminate

Unfused

K938

Pelvis

Ischium

Indeterminate

Unfused

KMR III:
The Late Harappan Phase

The other domestic animals reported earlier ( Joglekar

2007; Goyal and Joglekar 2008) like cats and pigs
were not noticed in this seasons faunal sample.

Faunal material from this phase contained 283

The wild mammals formed the next major com-

skeletal fragments of which 159 (56.18%) could be

ponent in the subsistence activity of Late Harappan

identified (Table 1). The analysed material revealed the

people since these animals together formed 23.25% of

presence of both domestic and wild mammals, reptiles

the total bones, which is quite a significant propor-

and birds like the previous cultural phase (Table 3).

tion. The wild mammals include spotted deer, black-

The domestic mammals include cattle, buffalo, sheep

buck, gazelle, wolf, porcupine, hare and house rat.

and goat. As noticed in the other cultural phases,

Hare (13.2%) dominated the group of wild mammals

cattle/buffalo (49.7%) dominated the assemblage fol-

and was represented by both axial and appendicular

lowed by goat/sheep (25.2%). Cattle were more than

fragments. Most of these fragments were having char-

buffalo as the later one is represented by three bones

ring marks and few of them were with butchering

only. Likewise, goats were more than sheep. However,

marks indicating the use of this small game as food.

the lowest proportion of cattle/buffalo was noticed in

In addition, most of the skeletal elements found were

this phase in comparison to other cultural phases. All

fragments with high meat value. It seems that hare

of these animals were included in the diet of people.

was an important food item during the Late Harap- 118 -

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

Table 6 Measurements of scapulae from Kanmer

Reg. No.

Trench

Layer

Species

K669

Z30

70.51

48.95

57.22

Bos indicus

K1041

Z30

11

25.54

16.36

19.43

Ovis aries

K1177

Z30

32.33

18.37

23.81

Ovis aries

K990

Z30

11

20.75

22.56

17.89

Antilope cervicapra

K1297

Z30

10.56

8.88

6.52

Lepus nigricollis

K1313

Z30

10.87

8.12

8.42

Lepus nigricollis

3: Maximum length of the articular process; 4: Width of the articular process; 5: Length of the glenoid cavity
Table 7 Measurements of humerus bones from Kanmer
Reg. No.

Trench

Layer

GL

Bp

Bd

Td

GLC

Species

K825

Z30

180.00

35.79

28.00

24.12

29.35

Axis axis

K985

Z30

33.92

--

--

Axis axis

K1151

Z29

--

69.13

--

--

Bos indicus

K1310

Z30

--

26.44

24.18

--

Capra hircus

K1216

Z30

32.81

--

--

29.00

Gazella bennetti

Bp: Maximum proximal width; Bd: Maximum distal width; WTH: Width of the distal trochlea

Table 8 Measurements of radius bones from Kanmer

Reg. No.

Trench

Layer

Bp

Tp

Bd

Wd

Species

K753

Z29

25.86

15.86

--

--

Antilope cervicapra

K1152

Z29

78.00

38.11

--

--

Bos indicus

K772

Z30

69.26

38.85

--

--

Bos indicus

K759

Z29

--

--

63.30

48.64

Bos indicus

K807

Z30

26.29

13.79

--

--

Gazella bennetti

Bp: Maximum proximal width; Tp: Proximal width; Bd: Maximum distal width; Wd: Width of the distal articular surface

pan phase.

porcupine in the assemblage. As noticed in the case

Spotted deer (Axis axis) made its presence in

of wolf, this also seems to be an accidental inclusion

the assemblage in the form a complete astragalus

in the habitation deposit. Blackbuck (Antilope cer-

(K828) only recovered from layer 7 of trench Z30.

vicapra) was present in the collection in the form of

Similarly, wolf (Canis lupus) was present in the form

four bones. These include a fragment of tibia (K795),

of one bone fragment only, i.e. distal portion of radius

a fragment of femur and two astragali, one complete

(K1194). The presence of wolf seems to be a chance

(K829) and other broken at the distal end. Three frag-

inclusion in the habitation deposit as this fragment

ments of gazelle (Gazella bennetti) were identified,

does not bear any mark indicating that it was used for

i.e. two first phalanges (K796 and K1195) and distal

food. A very badly preserved and weathered molar of

portion of a tibia (K994). Both blackbuck and gazelle

porcupine (Hystrix indica) was recovered from layer

were used for food purpose as suggested by the marks

7 of trench Z30. This was the lone representation of

noticed on these fragments.

- 119 -

P. Goyal and P.P. Joglekar

Table 9 Measurements of femora from Kanmer

Reg. No.

Trench

Layer

Bp

Bd

Td

GLC

TC

Species

K766

Z30

10

--

24.20

41.98

--

--

A. cervicapra

K668

Z30

--

74.43

--

--

--

Bos indicus

K1299

Z30

19.14

--

--

7.80

7.25

Lepus nigricollis

Bp: Maximum proximal width; GLC: Length of the capitulum; TC: Width of the capitulum; Bd: Maximum distal width

Table 10 Measurements of tibia from Kanmer

Reg. No.

Trench

Layer

Bp

Tp

Bd

Dd

Species

K1073

Z30

--

--

25.48

20.93

Antilope cervicapra

K1075

Z30

--

--

25.26

19.60

Antilope cervicapra

K776

Z30

10

--

--

21.44

18.36

Antilope cervicapra

K795

Z29

--

--

27.02

22.10

Antilope cervicapra

K1081

Z30

--

--

55.07

--

Bos indicus

K1034

Z30

11

39.13

38.60

--

--

Capra hircus

K994

Z30

--

--

20.69

17.49

Gazella bennetti

K1250

Z30

--

--

11.53

6.37

Lepus nigricollis

Bp: Maximum proximal width; Tp: Proximal width; Bd: Maximum distal width; Dd: Maximum distal thickness

Table 11 Measurements of Bos indicus astragalus bones from Kanmer

Reg. No.

Trench

Layer

Gll

GLm

Bd

Dl

Dm

Species

K901

Z30

66.71

60.83

39.20

36.03

35.30

Bos indicus

K902

Z30

51.7

47.85

30.05

26.16

26.32

Bos indicus

K903

Z30

51.35

47.58

29.44

26.87

--

Bos indicus

K1097

Z30

62.64

57.42

41.76

33.58

32.51

Bos indicus

K1121

Z30

69.90

63.63

42.47

33.60

35.70

Bos indicus

K1159

Z30

70.51

62.50

43.48

35.63

36.70

Bubalus bubalis

K868

Z30

28.04

26.90

19.28

15.62

16.98

Axis axis

K829

Z30

27.24

26.23

17.15

15.04

--

Antilope cervicapra

K1134

Z30

24.67

23.59

15.08

13.40

13.14

Tatracerus quadricornis

K803

Z30

24.60

23.80

14.82

13.43

16.00

Gazella bennetti

K804

Z30

25.80

25.18

15.80

12.27

--

Gazella bennetti

GLl: Lateral Length; GLm:Medial length; Bd: Maximum distal width; Dl: Maximum lateral thickness; Dm: Maximum medial
thickness

Reptiles and birds do not seem to have made a

the Late Harappan phase in the form of cervical verte-

significant contribution to the food economy of the

bra belonging to monitor lizard (Varanus bengalensis).

Late Harappan people as evidenced by the recovered

The birds made their presence felt in the assemblage

fragments. Only one reptile could be identified from

in the form of one bone fragment each of domestic

- 120 -

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

Table 12 Measurements of Bos indicus calcaneum bones from Kanmer

Reg. No. Trench Layer

Maximum Length

Maximum Width

Maximum Height

Species

K800

Z30

--

14.61

19.52

Capra hircus

K801

Z30

57.37

16.36

19.52

Capra hircus

K982

Z30

66.07

19.45

25.43

Axis axis

K1143

Z29

51.18

15.04

21.94

Gazella bennetti

K786

Z30

--

14.28

21.56

Antilope cervicapra

K780

Z30

10

54.42

15.80

21.77

Antilope cervicapra

K802

Z30

51.58

15.67

22.58

Antilope cervicapra

fowl (Gallus domesticus) and pond heron (Ardeola

nilgai, spotted deer and house rat were represented

grayii). Though the skeletal fragments belonging to

in the form of one fragment each. Wild buffalo was

birds do not bear any marks, perhaps the identified

present in the form of a complete second phalanx

birds were used as food.

(K1146) recovered from layer 5 of trench Z29, while

Axis axis was in the form of distal portion of a right
sided first phalanx (K1076). Blackbuck dominated

KMR IV:
The Historic Phase

the wild mammals and was represented in the form of

12 skeletal fragments. It contributed 3.81% to the total
faunal assemblage recovered from this phase. After

A total of 521 fragments belonging to this phase were

blackbuck, gazelle seems to be an important animal

analysed of which 315 (60.46%) could be identified

and 5 bone fragments of blackbuck were identified.

(Table 1). Five domestic animals have been identified

The nature of skeletal fragments belonging to black-

from the Historic phase - cattle, buffalo, sheep, goat

buck and gazelle suggest that these were hunted for

and dog. Cattle/buffalo (72.06%) dominated the as-

dietary purpose. Hare (n=3) also seems to have been a

semblage followed by sheep/goat (17.75%). Human

food item in the food economy of the Historic phase

activities on the skeletal fragments of these domestic

people of Kanmer. All the house rat skeletal elements

mammals suggest that all these (cattle, sheep and goat)

identified from this phase were later intrusions.

were used for food purpose. Dog was represented in

the form of distal portion of humerus unearthed from

phase people at Kanmer include only birds. The

layer 5 of trench Z30 only. This fragment was devoid

reptiles were completely absent from the assemblage.

of any marks and therefore, it can be said that this

Only three fragments of birds have been identified of

fragment did not belong to the food refuse at Kan-

which only one could be identified up to species level,

mer. Pig, cat and horse were identified from this phase

i.e. Gallus domesticus, while the other two fragments

earlier ( Joglekar 2007; Goyal and Joglekar 2008),

were identified as of medium-sized birds.

The non-mammals exploited by the Historic

however, these were not represented in the collection

this year.

KMR V:
The Medieval Phase

Wild mammals identified from this phase include

Bubalus arnee (wild buffalo), Boselaphus tragocamelus

(nilgai), Axis axis (spotted deer), Antilope carvicapra
(blackbuck), Gazella benetti (gazelle), Lepus nigricol-

A very small number (n=54) was recovered from this

lis (hare) and Rattus rattus (house rat). Wild buffalo,

phase, out of which only 25 could be identified (Table

- 121 -

P. Goyal and P.P. Joglekar

Table 13 Measurements of first phalanges from Kanmer

Reg. No.

Trench

Layer

GL

Bp

Tp

Bd

Species

K671

Z30

--

26.20

32.90

--

Bos indicus

K752

Z29

57.31

26.23

26.55

25.16

Bos indicus

K762

Z30

10

54.03

26.53

33.06

24.92

Bos indicus

K778

Z30

10

--

--

--

18.61

Bos indicus

K905

Z30

56.02

26.2

31.96

25.85

Bos indicus

K906

Z30

56.69

25.9

32.16

27.28

Bos indicus

K907

Z30

56.28

28.29

30.55

27.32

Bos indicus

K908

Z30

55.52

26.65

31.46

26.46

Bos indicus

K930

Z30

--

--

--

14.80

Bos indicus

K977

Z30

56.63

25.54

31.74

25.90

Bos indicus

K1048

Z30

--

36.26

42.14

--

Bos indicus

K904

Z30

61.58

32.15

35.18

32.29

Bubalus bubalis

K978

Z30

55.36

27.47

29.55

25.15

Bubalus bubalis

K1049

Z30

--

36.22

42.50

--

Bubalus bubalis

K1090

Z30

64.30

35.66

35.35

33.56

Bubalus bubalis

K1110

Z30

57.70

36.17

34.54

34.67

Bubalus bubalis

K782

Z30

36.52

11.85

12.86

11.72

Capra hircus

K823

Z30

39.86

14.46

15.28

13.04

Capra hircus

K952

Z30

37.85

13.36

14.91

12.5

Capra hircus

K979

Z30

40.17

12.62

13.96

11.75

Capra hircus

K1238

Z30

40.10

14.23

16.06

12.62

Capra hircus

K1259

Z30

--

--

--

11.02

Capra hircus

K760

Z29

--

12.90

15.11

--

Capra/Ovis

K1076

Z30

--

--

--

14.06

Axis axis

K779

Z30

10

31.9

9.25

13.02

7.49

Gazella bennetti

K781

Z30

33.5

10.35

14.08

7.93

Gazella bennetti

K796

Z29

36.52

10.89

14.06

--

Gazella bennetti

K917

Z30

33.66

9.85

14.13

8.14

Gazella bennetti

K980

Z30

35.48

9.71

13.37

8.46

Gazella bennetti

K988

Z30

11

32.93

10.03

12.29

9.46

Gazella bennetti

K1187

Z30

33.43

10.10

14.27

8.52

Gazella bennetti

K1195

Z30

28.39

9.37

11.41

10.00

Gazella bennetti

K1331

Z30

10

--

--

--

8.15

Gazella bennetti

K666

Z30

35.8

10.28

14.79

8.84

A. cervicapra

K667

Z30

35.98

10.38

14.53

8.96

A. cervicapra

GL: Maximum length; Bp: Maximum proximal width; Tp: Proximal thickness; Bd: Maximum distal width

- 122 -

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

Table 14 Measurements of second phalanges from Kanmer

Reg. No.

Trench

Layer

GL

Bp

Tp

Bd

Species

K1080

Z29

39.78

27.11

27.87

23.94

Bos indicus

K1309

Z30

--

23.47

24.86

--

Bos indicus

K1386

Z30

11

--

--

--

12.68

Bos indicus

K1407

Z30

28.99

23.18

27.41

20.81

Bubalus bubalis

K981

Z30

33.56

26.78

27.22

23.66

Bubalus bubalis

K1144

Z30

39.78

37.87

37.56

35.84

Bubalus bubalis

K1146

Z30

48.19

35.53

40.16

30.37

Bubalus bubalis

K1138

Z30

30.12

20.77

22.01

17.02

B. tragocamelus

K799

Z30

10

36.07

22.4

28.89

25.11

B. tragocamelus

K1181

Z30

20.07

9.41

10.52

7.55

Axis porcinus

K763

Z30

10

18.69

8.94

10.53

7.27

Gazella bennetti

K1406

Z30

19.71

8.72

11.31

7.67

Gazella bennetti

K764

Z30

10

18.95

7.44

9.89

6.64

Gazella bennetti

K1182

Z30

16.99

8.99

9.57

6.95

Muntiacus muntjak

K783

Z30

17.94

11.63

11.65

8.50

Sus domesticus

K754

Z29

22.45

8.80

11.78

7.58

A. cervicapra

GL: Maximum length; Bp: Maximum proximal width; Tp: Proximal thickness; Bd: Maximum distal width
Table 15 Measurements of incisors from Kanmer
Reg. No.

Trench

Layer

Length

Width

Species

K1337

Z30

11

12.59

6.74

Bos indicus

K1388

Z30

11

8.93

4.20

Bos indicus

K1402

Z30

11

13.24

7.62

Bos indicus

K1403

Z30

11

12.98

6.45

Bos indicus

K1404

Z30

11

13.27

6.90

Bos indicus

K890

Z30

10

7.23

2.43

Capra/Ovis

1). The identified material belonged to the domestic

and two metapodia. The nature of the material shows

mammals only. The cattle dominated the whole as-

that cattle, sheep and goat were used for food, whereas

semblage and contributed 68% of the total assemblage

the horse was not consumed.

recovered from this phase (Table 3). The rest of the

assemblage was represented by goat/sheep (20%)
and horse (12%). In the absence of any distinguishing

Bone Measurements

marker, it was difficult to separate faunal fragments

(Tables 6-18)

belonging to goat from those belonging to sheep.

Therefore, all of these have been kept together in the

Recording bone measurements is an essential part of

category of Capra/Ovis. Horse was represented by

standard archaeozoological methodology. The bone

three skeletal fragments, i.e. proximal end of a femur

measurements provide a tool to look at the sizes of

- 123 -

P. Goyal and P.P. Joglekar

past animals, particularly the domestic ones. With the

ferent animal groups (Table 20) revealed that the

estimates of the size and height of the animals at the

type of animals in different cultural periods were

withers, it is possible to compare the domestic stocks

different. It is clear from Table 20 that wild animals

at various archaeological sites and in different cultural

played a significant role in the animal based subsis-

periods ( Joglekar 2000-2001). All the measurements

tence economy in the Late Harappan phase. The wild

were recorded in mm using a digital caliper with a

animals form a significant proportion (23.25%) in the

least count of 0.01 mm. It has been possible to esti-

whole assemblage belonging to this phase. The previ-

mate the height of the cattle at the withers using the

ous report (Goyal and Joglekar 2008) also indicated

medial length of the astragalus (Zalkin 1970). A few

a slight increase of wild animals in this phase. A look

point estimates of the height of cattle found at Kan-

at the composition of wild animals in this particular

mer are given in Table 19. Earlier Joglekar (2007) has

phase revealed that Indian hare (Lepus nigricollis)

reported the height of cattle at withers found from

contributed appreciably forming a share of 13.2% in

the Late Harappan phase at Kanmer to vary from110

the whole assemblage (Table 3). The other trenches

to 117 cm.

currently under investigation are showing a similar

The distal width of the trochlea of the humerus

trend and revealing good number of skeletal frag-

can be converted to estimate the fat-free carcass

ments of Indian hare from the Late Harappan phase.

weight (kg) using the factor given by Noddle (1973).

This animal is widely distributed in India and usually

A single estimate obtained from cattle humerus

lives in the neighbourhood of villages and areas under

(K1151) showed that cattle at Kanmer from the cul-

cultivation. It is also known for entering the agricul-

tural break between the Medieval and the Historic

tural fields and inflicts severe damage to crop yields.

Phase had fat-free carcass weight equal to 146.43 kg.

Slightly more than 60% material presented in this

report comes from the Mature Harappan phase contexts. The analyzed material revealed the dominance

Discussion and Conclusion

of cattle in all the cultural phases at Kanmer. The skeletal elements of cattle were present in all the layers an-

Animal remains throughout the occupation at Kan-

alyzed. The maximum representation of this animal is

mer indicate that the food economy was mainly based

noticed in layer 9 (Mature Harappan Phase) of trench

on the consumption of domestic animals. In general,

Z30. The high proportion of cattle bones indicates

the economy was mainly dependent on domestic

the value of this animal not only for food purposes

animals, but it was observed that exploitation of wild

but also for other secondary products. The cow dung

fauna was at its peak in the Late Harappan phase. A

was extensively used for plastering the floors and walls

similar pattern of animal exploitation was also seen at

of the houses as revealed by the excavation in almost

other Harappan sites in Gujarat like Rangpur (Nath

all the cultural phases. The presence of bifid spines of

1962-63), Surkotada (Sharma 1990), Nageswar (Shah

the thoracic vertebra suggests the presence of humped

and Bhan 1992), Kuntasi (Thomas et al. 1996) and

variety of cattle at Kanmer.

Shikarpur (Thomas et al. 1995) etc., where the exploi-

tation of wild fauna was at its maximum towards the

part of the faunal material unearthed in the third sea-

close of the Harappan period. The contribution of

son (2007-08) at the site of Kanmer. Therefore, these

reptiles and birds was very less in comparison to the

results are preliminary in nature like the previous

mammals (Figure 13)

reports ( Joglekar 2007; Goyal and Joglekar 2008).

A comprehensive report at the end of excavations at

A closer look at the relative proportion of dif- 124 -

As mentioned above, this report is based on a

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

Table 16 Measurements of premolars and molars from Kanmer

Reg. No.

Trench

Layer

Tooth

Length

Width

Species

K892

Z30

10

Maxillary

First molar

22.29

17.22

Bos indicus

K1391

Z30

Maxillary

Second molar

18.34

15.08

Bos indicus

K1401

Z30

11

Maxillary

Third molar

15.05

9.23

Ovis aries

K1400

Z30

11

Maxillary

Second molar

18.31

10.11

Bos indicus

K1005

Z30

Mandibular

First molar

29.80

12.28

Bos indicus

K983

Z30

Mandibular

Third molar

37.23

12.61

Bos indicus

K984

Z30

Mandibular

Third molar

12.36

6.93

Bos indicus

K1091

Z30

Mandibular

Third molar

40.30

14.84

Bos indicus

K997

Z30

Mandibular

Second premolar 18.36

13.18

Bos/Bubalus

K1328

Z30

Mandibular

Second premolar 8.48

5.14

Bos indicus

K945

Z30

Mandibular

Fourth premolar 19.02

11.46

Bos indicus

K996

Z30

Mandibular

First molar

13.53

7.70

Capra/Ovis

K1353

Z30

10

Mandibular

First molar

14.97

8.52

Capra/Ovis

K1245

Z30

Mandibular

First molar

14.49

10.19

Capra hircus

K1314

Z30

Mandibular

First molar

13.28

8.04

Capra hircus

K1333

Z30

10

Mandibular

Second molar

14.18

6.95

Capra hircus

K1387

Z30

11

Mandibular

Second molar

15.76

7.32

Capra hircus

K1026

Z30

11

Mandibular

Second premolar 17.02

10.71

Capra hircus

K1397

Z30

10

Mandibular

Second molar

14.08

Axis axis

22.08

Table 17 Measurements of metapodia from Kanmer

Reg. No.

Trench

Layer

Bone

Bp

Tp

Bd

Td

Species

K1122

Z29

Metacarpal

--

--

55.90

28.88

Bos indicus

K1123

Z29

Metacarpal

--

--

61.45

32.56

Bos indicus

K1071

Z30

Metacarpal

--

--

53.93

30.49

Bubalus bubalis

K761

Z30

Metacarpal

20.50

9.87

Ovis aries

K989

Z30

11

Metacarpal

20.42

15.9

--

--

Ovis aries

K1284

Z30

Metacarpal

19.94

14.87

--

--

Gazella bennetti

K915

Z30

Metatarsal

32.15

28.97

--

--

Bos indicus

K1154

Z29

Metatarsal

44.87

39.09

--

--

Equus caballus

K1155

Z29

Metatarsal

--

--

45.30

36.54

Equus caballus

K1124

Z29

Metatarsal

23.79

21.38

--

--

A. cervicapra

K1355

Z30

10

Metatarsal

14.69

14.63

--

--

A. cervicapra

Kanmer would project a full picture of animal-based

kwal, Y. S. Rawat and Prof. Toshiki Osada for entrust-

subsistence economy at Kanmer.

ing the faunal material from Kanmer for analysis to

us. The authors also thank the other team members of

Acknowledgements

the Kanmer Archaeological Research Project (KARP)

The authors are extremely grateful to Drs. J. S. Khara-

for their support during this research, without which

- 125 -

P. Goyal and P.P. Joglekar

Table 18 Measurements of mandibles from Kanmer

Reg. No.
K916

Trench
Z30

Layer Condyle width

9
37.98

dP2
--

dP3
--

dP4
--

Species
Bos indicus

K1040

Z29

35.54

--

--

--

Bos indicus

K1398

Z30

11

--

length: 4.81

length: 8.29

length: 14.59

Capra hircus

K1405

Z30

20.37

width: 3.06
--

width: 5.21
--

width: 6.27
--

Ovis aries

Table 19 Cattle height at the withers estimates (cm)

Phase

Height

The Medieval Phase

105.08

The Historical Phase

116.44

The Mature Harappan Phase

87.07, 87.56, 111.32

Table 20 Relative proportion (%) of different animal groups

Groups

Medieval

Break*

Historic Phase

Break**

Late Harappan Mature Harappan

Cattle/Buffalo

68

100

72.06

71.07

49.7

62.65

Sheep/Goat

20

17.75

21.38

25.2

24

Others Domestic

12

0.34

0.5

mammals
Wild Mammals

8.9

6.3

23.25

10.25

Reptiles

0.6

0.2

Birds

0.95

1.25

1.25

2.4

* between the Medieval and the Historic Phase, ** between the Historic and the Late Harappan Phase

Figure 13 Relative proportion (%) of different animal groups

- 126 -

Report on the faunal remains recovered from Kanmer, Gujarat, during the third field season (2007-08)

this work was not possible. The authors also acknowl-

Paper presented during the 3rd International Congress

edge the help given by the Department of Archaeol-

of the Society of South Asian Archaeology (SOSAA)

ogy, Maharaja Sayajirao University, Baroda during

held at University of Kelaniya, Colombo, Sri Lanka,

20th-21st August 2010.

their stay there. The first author thanks Dr. Arati

Gupta, S.K., D.N. Sharma and R .L. Bhardwaj (1987)

Deshpande-Mukherjee for her help time to time. The

Comparative Anatomy of the Hard Palate of

authors also thank C.V. Sharda for her help in taking

Buffalo (Bubalus bubalis) and Ox (Bos indicus).

the photographs of the bones. The villagers of Kan-

Haryana Veterinarian 26: 22-26.

mer also lend a hand in different ways and the authors

Gupta, S.K., D.N. Sharma and R .L. Bhardwaj (1990)

are thankful to them.

Comparative Anatomy of the Bony Palate of Ox,

Buffalo and Yak. Indian Journal of Animal Science
60(4): 436-438.
Higham, C.F.W. (1975) The Faunal Remains from the 1996

References

and 1998 Excavations at Non Nok Tha, Northeastern

Bartram, L.E.J., and C.W. Marean (1999) Explaining the

Thialand. University of Otago, Otago.

Klasies Pattern: Kua Ethnoarchaeology, the Die

Kelders Midle Stone Age Archaeofauna, Long Bone

Joglekar, P.P. (2000-2001) Applications of Mathematics and

Fragmentation and Carnivore Ravaging. Journal of

Statistics in Archaeozoological Research at Deccan

Archaeological Science 26: 9-29.

College. Bulletin of the Deccan College Post-graduate

and Research Institute 60-61: 109-136.

Clason, A .T. (1972) Some Remarks on the Use and

Presentation of Archaeozoological Data. Helinium

Joglekar, P.P., P.K. Thomas, Y. Matsushima and Seema

12(2): 139-153.

J. Pawankar (1994) Osteological Differences

Driesch von den, A. (1976) A Guide to the Measurement of

between the Forelimb bones of Ox (Bos indicus),

Animal Bones from Archaeological Sites. Harvard

Buffalo (Bubalus bubalis) and Nilgai (Boselaphus

University Press, Cambridge.

tragocamelus). Journal of Bombay Veterinary College

Goyal, Pankaj and P.P. Joglekar (2008) Faunal Remains at

5 (1-2): 1720.

Kanmer: A Multi-Cultural site in Gujarat. Paper

Joglekar, P.P. (2006) Preliminary Report of the Faunal

presented during the 36th Conference of the Indian

Remains from Kanmer, Gujarat. Paper presented

Society for Prehistoric and Quaternary Studies

during the First Conference of the Society of South

(ISPQS) held at Bhubaneswar (29-31 December

Asian Archaeology held at Mumbai (16-18 December

2008).

2006).

Goyal, Pankaj and P.P. Joglekar (2008) Report of the Faunal

Joglekar, P.P. (2007) Report of the Faunal Remains recovered

Remains recovered from Kanmer, Gujarat, during

from Kanmer, Gujarat, during the First Field Season

the Second Field Season (2006-07), in T. Osada

(2005-06), in T. Osada (ed.) Lingusitic, Archaeology

and A. Uesugi (eds.) Lingustic, Archaeology and the

and the Human Past: Occasional Paper 2. Research

Human Past: Occasional Paper 5. Research Institute

Institute for Humanity and Nature, Kyoto. pp. 47-

for Humanity and Nature, Kyoto. pp. 25-43

76.

Goyal, Pankaj and P.P. Joglekar (2010a). Animal Utilization

Joglekar, P.P. and Pankaj Goyal (2008) Faunal Remains from

Patterns at Kanmer, Gujarat, Paper presented during

Historic Phase at Kanmer, a Multi-culture Site in

the Bhuj Roundtable 2010: International Conference

Kachchh Region of Gujarat, India. Paper presented

on Gujarat Harappans and Chalcolithic Cultures

during the Second Conference of the Society of South

held at Bhuj, 28-31 January 2010.

Asian Archaeology held at Islamic Azad University of

Kazeroun, Shiraz, Iran (25-29 May 2008).

Goyal, Pankaj and P.P. Joglekar (2010b) Reconstructing

Animal Based Subsistence of Harappan Culture in

Kharakwal, J.S., Y.S. Rawat and T. Osada (2005) Harappan

Gujarat: A Case Study from the Region of Kachchh.

Sites in Kachchh and New Opportunities of

- 127 -

P. Goyal and P.P. Joglekar

Tourism, in Heritage Tourism. Government of

of Sheep and Goat. Journal of Archaeological Science

Gujarat. pp. 35-43.

13: 567-577.

Kharakwal, J.S., Y.S. Rawat and T. Osada (2007) Kanmer:

Sadek-Kooros, H. (1975) Intentional Fracturing of Bone:

A Harappan site in Kachchh, Gujarat, India, in T.

Description of Criteria, in A.T. CLason (ed.)

Osada (ed.) Lingusitic, Archaeology and the Human

Archaeoz oolog ical Studies. Nor th-Hol land

Past: Occasional Paper 2. Research Institute for

Publication, Amsterdam. pp. 139-150.

Schmidt, E. (1972) Atlas of Animal Bones. Elsevier,

Humanity and Nature, Kyoto. pp. 21-46.

Kharakwal, J.S., Y.S. Rawat and T. Osada (2008a) Preliminary

Amsterdam.

Report of the Excavation at Kanmer, Kachchh,

Shah, D.R. and K.K. Bhan. (1992) Faunal Remains, in

Gujarat 2006-07. Shodh Patrika 59(1-2): 99-132.

K.T.M. Hegde, K.K. Bhan, V.H. Sonawane, K.

Kharakwal, J.S., Y.S. Rawat and T. Osada (2008b) Preliminary

Krishnan and D.R . Shah (eds.) Excavation at

Report of the Excavation at Kanmer, Kachchh,

Nageswar, Gujarat: a Harappan shell working site

Gujarat 2006-07. Shodh Patrika 59(1-2): 99-132.

on the Gulf of Kutch. M.S. University Archaeology

Kharakwal, J. S., Y. S. Rawat and T. Osada (2009) Excavations

Series No. 18, Baroda. pp. 145-47.

at Kanmer: A Harappan Site in Kachchh, Gujarat.

Sharma, A.K. (1990) Animal Bone Remains, in J.P.

Puratattva 39: 147-164.

Joshi (eds.) Excavations at Surkotada 1971-

Mig uad, P. (1989) Discriminating Characters on the

72 a n d E x pl orati on i n Ku tch. Mem o ir s of

Acropodium of Domestic and Wild Pig. Acta

the Archae olo g ica l Sur ve y of India no.87.

Veternaria (Brno) 42: 109-133.

Archaeological Survey of India, New Delhi. pp.

Nath, B. (1962-63) Animal Remains from Rangpur. Ancient

372-383.

India 18-19: 153-160.

Thomas, P. K., P.P. Joglekar, A. Deshpande-Mukherjee and

Pawankar, S.J. and P.K. Thomas (2001) Osteological

S.J. Pawankar (1995) Harappan Subsistence Patterns

Differences b et we en Blackbuck (Antilop e

with Special Reference to Shikarpur, a Harappan

cervicapra), Goat (Capra hircus) and Sheep (Ovis

Site in Gujarat. Man and Environment XX (2): 33-

aries). Man and Environment 26(1): 109-126.

41.

Payne, S. (1975) Partial Recovery and Sample Bias, in A.T.

Thomas, P.K., Y. Matsushima and A. Deshpande (1996)

Clason (ed.) Archaeozoological Studies. North-

Faunal Remains, in M.K. Dhavalikar, M.H. Raval

Holland Publishing Company, Ltd., Amsterdam.

and Y.M. Chitalwala (eds.) Kuntasi: A Harappan

pp. 7-17.

Emporium on West Coast. Deccan College, Pune.

Prummel, W. and H.J. Frisch (1986) A Guide for the

pp. 297-330.

Distinction of Species, Sex and Body Side in Bones

- 128 -

Preliminary anthracological analysis from Harappan Kanmer

Preliminary anthracological analysis from Harappan Kanmer:

Human-environment interactions as seen through fuel resources exploitation and use

Carla Lancelotti
Department of Archaeology and Anthropology Instituci Mil i Fontanals
Spanish National Research Council (CSIC), Barcelona Spain/
Department of Archaeology University of Cambridge UK

Marco Madella
ICREA & Department of Archaeology and Anthropology Instituci Mil i Fontanals
Spanish National Research Council (CSIC), Barcelona Spain

Abstract
This paper presents the preliminary results of charcoal analysis conducted on samples collected at the site of Kanmer during the
2008 fieldwork season. The main aim of the sampling was to analyse and understand the exploitation and use of fuel resources
during the Mature Harappan Phase of occupation at the site. Sediments form fireplaces, pits, floors and ash accumulations were
floated in order to retrieve charred wood remains. The analysis has revealed the use of a surprisingly high variety of arboreal species,
including some that are not currently found nearby the site. This new data allows the formulation of hypothesis on the presence of
different environmental conditions at Kanmer during the Mature Harappan Period.

Introduction

and the Rann of Kachchh, which is about 5 km away

to the south. The archaeological mound measures

During the months of January and February 2008,

105 m x 115 m and rises 8 m above the surrounding

archaeobotanical sampling was carried out at the

plain (Kharakwal et al. 2007, 2008). It is encircled

site of Kanmer (Kachchh, Gujarat) with the aim

by a massive stonewall and its top is shallow with

of understanding the exploitation and use of fuel

peripheral regions slightly higher than the central

resources during the Mature Harappan Period.

part.

Investigation at the site of Kanmer has been

conducted since 2005-2006 by a joint team of

Aludawaro Vokro that originates in an inselberg-

researchers from the Institute of Rajasthan Studies,

type hill 2 km west of the mound, drains the area

JNR Rajasthan Vidyapeeth (Udaipur), the Gujarat

around the site. Rajaguru and Shuhsama (2008)

State Department of Archaeology and the Research

suggested that the nullah was permanently active

Institute for Humanity and Nature (Kyoto, Japan).

during Harappan times. Presently, the only source

The modern village of Kanmer is located in Rapar

of water, except deep ground water, is a large natural

Taluka, Kachchh District. The archaeological mound,

tank located to the southeast of the hill. It is not as

known locally as Bakar kot, stands to the north of

yet clear whether this water basin was in use during

the village (2323N; 7052E). A small natural hill

the Harappan period although findings of Harappan

oriented northeast southwest, lays between the village

pottery on its margins may suggest this. However,

- 129 -

An ephemeral stream (nullah), known as

C. Lancelotti and M. Madella

the basin could have also been dug into Harappan

many other dry/saline areas of India and Pakistan.

deposits in a later period, which could also explain the

presence of Harappan pottery along the shore.

Materials and methods

The geology of the area is dominated by sandstone

intercalated with white shale. These sediments are

often traversed by faults and linear features running

A total of 37 contexts were floated and 41 soil samples

west-northwest and east-southeast (Rajaguru and

were collected during the 2007-2008 field season. Of

Shushama 2008). The vegetation around the site is

these only 10 flotation samples were deemed suitable

today scarce, composed predominantly by Prosopis

for analysis within the scope of this research. This

juliflora, Azadirachta indica, Salvadora persica and S.

was due to the fact that most of the other samples

oleoides, Capparis decidua and Calotropis sp. (personal

collected belonged to historical or medieval periods

observations). These are the dominant species of the

and were thus outside the time frame of this work.

dry-thorn scrubland formation that characterises

this region. Trees are scarce and Prosopis juliflora

were processed by bucket flotation with water

(Sw.) DC dominates the assemblage. This species was

delivered on site by bullock cart from the nearby

introduced from South America in the last decades

tank (Figure 1). A mesh of 0.5 mm was used in order

of the 19th century as a powerful tool to re-vegetate

to collect small seeds as well as charcoal. Whenever

the area and now has become so invasive that it is the

possible, a standard quantity of soil (20 litres) was

most common shrub/tree in the territory as well as in

floated so as to have control on the quantitative

As there was not running water on site, samples

Figure 1 Bucket flotation at Kanmer where water was collected from the nearby tank and carried to the site by bullock cart.
Table 1 Kanmer: Samples divided by the trenches where they were collected from.
Trench

Charcoal samples (n.)

Sediment samples (n.)

AA17

FF29

Z17

11

Z18

Z30

- 130 -

Preliminary anthracological analysis from Harappan Kanmer

Figure 2 Kanmer: Kiln structure dated to the Mature Harappan Phase

unearthed in trench Z17 at a depth of 5.39 meters below the datum point.
Table 2 Kanmer: Distribution of samples per context type.

Context type

No. of charcoal samples

Fireplace

Floors

Ash accumulations

Occupation levels

analysis. The flots were dried in the shade and then

Schweingruber 1990).

transported to the G. Pitt-Rivers Laboratory for

Bioarchaeology, University of Cambridge, UK. Here

recovered from five trenches that cover the three

they were sorted under a low power microscope to

excavated areas. Table 1 shows the distribution of

separate charcoal from eventual seeds. The charcoal

samples per trench. Samples were recovered from

was then analysed at high magnification with a Leica

fireplaces, floors, ash accumulations, and occupation

microscope and identified using reference material

levels (Table 2).

appositely created for this work as well as published

material (Asouti and Fuller 2007; Fahn et al. 1986;

Z17 (Figure 2) was sub-sampled: one bulk soil sample

Hather 2000; Inside Wood Database 2004-onwards;

from the level just above the kiln; three sediment

- 131 -

The samples analysed during this work were

Due to its nature, the big kiln found in Trench

C. Lancelotti and M. Madella

samples from the kiln central cylindrical structure;

of indeterminate and determinate fragments in a

one bulk sample from just outside the fuel chamber;

sample. It is a parameter to evaluate taphonomy in

one bulk sample from the inside of the fuel chamber.

the samples and its value stands in the comparison

The flot from the latter sample contained no charcoal.

between the samples rather than in its absolute

value for each individual sample (Asouti 2003). The
Simpsons Diversity Index measures the presence

Anthracological analysis

and distribution of a specific taxon in the sample.

This formula can be used to distinguish between

A total of 1222 fragments recovered from 10 contexts

generalised (such as pits or accumulation of charcoal

were examined from the site of Kanmer. The contexts

refuse) and specialised (such as fireplaces) assemblages

chosen were the only ones that dated to the Mature

(Asouti 2003). Density, expressed as the total weight

Harappan phase of occupation. Although the

of charcoal material per litre of floated sediment,

context number is lower than the one recommended

provide informative comparisons on different samples

for anthracological analysis, the total number of

and can be a tool for a first assessment of taphonomic

fragments is adequate for quantification and statistical

processes over the samples when combined with other

analyses (Chabal 1997; Chabal et al. 1999; Scheel-

indices (Asouti 2003).

Ybert 2002).

The quantity of charcoal retrieved from each

sample is low ; the Density Index is below the

Taphonomy and preservation

average in 78% of the samples (Figure 3). However,

The percentage of undetermined fragments, the

the percentage of undetermined fragments is also

Simpsons Diversity Index and the Density Index

generally low and it is below the average in 70% of the

were used as a measure of the preservation conditions

samples (Figure 4).

of the charcoal fragments. The Percentage of

Unidentified Fragments shows the proportions

carefully due to the very high standard deviation

Although these figures need to be considered

Table 3 Kanmer: Mean, Standard Deviation and Median values for the Density, percentage
of undetermined fragments and Diversity Indices.

Density Index

Undetermined %

Diversity Index

MEAN

22.63

0.023

0.81

STDEV

11.56

0.024

0.09

MEDIAN

19.27

0.017

0.82

Table 4 Kanmer: Spearman's coefficient of correlation for values of Density Index and percentage of
undetermined fragments. There is no significant correlation at the 0.05 interval level (2-tailed).

Spearman's rho

Density

Undetermined %

Correlation Coefficient

Density

Undetermined%

1.000

-0.350

Sig. (2-tailed)

0.356

Correlation Coefficient

-0.350

1.000

Sig. (2-tailed)

0.356

- 132 -

Preliminary anthracological analysis from Harappan Kanmer

Table 5 Kanmer: Spearman's correlation test between Simpsons Diversity Index and percentage of
undetermined fragments. There is no significant correlation between the values at the 0.05 interval level (2-tailed).

Spearman's rho

Diversity

Undetermined %

Diversity

Undetermined %

1.000

-0.458

Sig. (2-tailed)

0.215

Correlation Coefficient

-0.458

1.000

Sig. (2-tailed)

0.215

Correlation Coefficient

Figure 3

Kanmer: Results of the Density Index (average 22.63). The contexts have been grouped according to their type: A) ash

accumulation; FL) floor deposit; FP) fireplace; HL) general occupation level. KMR FL002 does not have a result for this Index, as
the charcoal retrieved from the single floor sample analysed was hand picked.

of the data, they show that post-depositional

can be the result of burning conditions, of recovery

taphonomic processes should not be a major concern

methods or of a scarce use of wood fuel a priori. The

in the preser vation of Kanmer charcoals. The

last possibility seemed to be confirmed by the strong

Spearmans Coefficient of correlation between the

positive correlation between Density and Simpsons

Density and percentage of undetermined fragments

Diversity Indices (Table 6). This means that when

indices confirms this result (Table 4). This is further

the density increases, the possibility of finding a

supported by the total absence of correlation between

higher number of species augments, which means that

the percentage of undetermined fragments and the

arboreal species were probably abundant around the

Simpsons Diversity Index (Table 5), which proves

sites. As the density is very low it can be postulated

that all samples, independently of their preservation

that wood was not exploited as the preferential source

state, display the same range of variety.

of fuel.

The preservation of charred material during and

after the burning activity is more problematic. The

small volumes of charcoal preserved in the sediments
- 133 -

C. Lancelotti and M. Madella

Figure 4

Kanmer: Percentage of undetermined fragments over the total number of charcoal fragments identified in each context

(average 0.023). The contexts have been grouped according to their type: A) ash accumulation; FL) floor deposit; FP) fireplace;
HL) general occupation level.
Table 6 Kanmer: Spearmans correlation between Density and Simpson's Diversity Indices.
There is a strong positive correlation at the 0.01 interval level (2-tailed).

Spearman's rho

Density

Diversity

Density
1.000
.
9
0.831
0.006
9

Correlation Coefficient
Sig. (2-tailed)
N
Correlation Coefficient
Sig. (2-tailed)
N

Diversity
0.831
0.006
9
1.000
.
9

Plants as fuel: descriptive and

relative abundance of species relating to riverine

quantitative results

environments and to coastal habitats (mangrove

A total of 23 taxa were identified in the samples

forest) is remarkable (Figure 6). As explained in the

analysed, plus two that could not be identified, as they

introduction, at present the only water bodies near

were not found in the reference collections that were

the site are an ephemeral nullah that receives water

used for this work. The assemblage is dominated, both

during the monsoon season but is otherwise dry,

in terms of fragments count and ubiquity occurrence,

and the tank to the southeast of the modern village.

by the arid dry thorn scrubland species Ziziphus sp.,

The percentage and ubiquity occurrence of Acacia

Azadirachta indica and Capparis decidua and by the

nilotica suggests that this taxon was not imported

riverine species Acacia nilotica (Table 7 and Figure

but must have been growing in the vicinity of the site

5). In addition, there are some species, such as the

and therefore there must have been a watercourse

mangrove coastal forest Avicennia marina, that even

nearby (see Table 7 and Figure 5). Furthermore the

if not abundant in the record are very important as

identification of Dalbergia sissoo, Shisham or Indian

palaeoenvironmental indicators.

rosewood, confirms the presence of a watercourse

near the site (Orwa et al. 2009).

If the dominance of species relating to arid

environments is not surprising, the presence and

- 134 -

Even more significant is the presence of a

Preliminary anthracological analysis from Harappan Kanmer

Table 7

Kanmer: Ubiquity scores (number of samples in which the species has been identified over the total number of samples

analysed) of the major taxa in the samples analysed and their ecological significance (where known). ADT= Arid Dry Thorn
scrubland; R= Riverine; M= Mangrove forest. In brackets the total number of samples analysed.
Site
(10)
8
8
6
5
5
5
5
5
5
4
4
4
3
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1

Species
Azadirachta indica A. Juss. (ADT)
Capparis decidua (Forssk.) Edgew. (ADT)
Acacia nilotica (L.) Willd ex Delile (R)
Ziziphus nummularia (Bum. f.) Wight & Aru. (ADT)
Salvadora persica Wall. (ADT)
Prosopis cineraria (L.) Druce (ADT)
Acacia senegal Willd. (ADT)
Calotropis procera (Aiton) W.T. Aiton (ADT)
Acacia sp. (ADT)
Ziziphus sp. (ADT)
Salvadora sp. (ADT)
Avicennia marina (Forssk.) Vierh. (M)
Type A
Senna siamea Lamk. (ADT)
Clerodendrum sp. (ADT)
Monocotyledon
Cordia sp. (ADT)
Ziziphus mauritiana Lam. (ADT)
Salvadora oleoides Decne. (ADT)
Tamarix aphylla (L.) Karst. (R)
Dalbergia sissoo Roxb. (R)
cf. Wrightia tinctoria (Roxb.) R. Br.
Ficus sp. (ADT)
Myristica sp.
cf. Phoenix sylvestris Roxb. (ADT)
Type C
Leaf fragments
Bark fragments

Ash
(2)
2
2
1
1
1
1
2
0
2
1
0
1
0
0
1
0
1
1
1
0
1
0
0
0
0
0
0
1

Floor
(1)
1
1
1
0
0
1
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0

Fireplace
(3)
2
1
1
2
2
1
1
1
1
0
2
1
1
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0

Occupation
(4)
3
4
3
2
2
2
2
3
2
3
2
2
1
1
1
0
0
0
0
1
0
0
0
1
1
0
1
0

mangrove species, Avicenna marina, grey or white

whereas fireplaces contain the most homogeneous

mangrove, which occurs in intertidal areas. Naidoo

deposits (Figure 8). This is in accordance with the

(1987) showed that the optimum growth rate is

theory that sees fireplaces as short-term deposits and,

achieved at a level of salinity of about 20%, which

therefore, they contain a less varied range of species

marks out this species as one of the most salt tolerant

(Chabal 1997).

among mangroves (Tengberg 2002). Its wood presents

a very distinct transverse section with vessels in rows

sp. from the nutmeg family, have been found in a

of 4 or more and scanty paratracheal parenchyma. The

single context. The preservation of the fragments

rays are composed of mixed cells and are 1 to 3 cells in

and the lack of reference material did not allow

width. Vessels present simple perforation plates and

for identification to species level. Four species of

alternate pits (Figure 7).

Myristica are native to western India, all from the

These data show that people in Kanmer used a

Western Ghats, concentrated more or less between

great variety of wood resources. In fact, all the contexts

Mumbai and Goa (Gamble 1902). Among these, only

analysed show a very high Diversity Index value (see

one has been historically used for timber: Myristica

average in Table 3). Among the context types, ash

malabarica (Bombay mace or false nutmeg ), and

accumulations show the highest Diversity Index,

none are reported as good fuel sources. The wood of

- 135 -

A few fragments of an exotic taxon, Myristica

C. Lancelotti and M. Madella

Figure 5 Kanmer: Pie chart showing the most abundant species in the samples analysed.
Only species with overall scores >10% in the ubiquity test were considered
and the species belonging to the same genus were grouped together.

Figure 6 Kanmer: Pie chart showing the distribution of vegetation types.

this genus is characterised by solitary vessels or vessels

context KMR FP005 Z17. The wood of this tree is

in short radial rows of two, axial parenchyma in

used today for turning, carving and making toys

marginal bands, vessels with scalariform perforation

and its leaves are used for dying (Gamble 1902; Patel

plates (Figure 9) and alternate to opposite pits, rays

1971). Microscopically it is characterised by vessels in

of 1-2 cells wide composed by a body of procumbent

short radial rows of 2-4, axial parenchyma diffuse in

cells with 1 or 2 marginal rows of square/upright cells.

aggregates, vessels with simple perforation plates and

Another species identified at Kanmer, which

alternate pits, rays 1-2 seriate composed of a body of

prime use was probably not for fuel is Wrightia

procumbent cells with 4 or more marginal rows of

tinctoria. This small tree, which does grows in

upright/square cells (Figure 10).

Gujarat, albeit rarely (Patel 1971), was identified from

- 136 -

Preliminary anthracological analysis from Harappan Kanmer

Figure 7 Kanmer: Avicennia marina, TS, and RLS of a fragment from context A007 AA17 (magnification 200X).

Figure 8 Kanmer: Simpson's Diversity Index scores for the different context types.

Environment and vegetation at Kanmer

of humidity, as this species thrives well in brackish

The anthracolog ical data from Kanmer show

waters. It is possible that at least the mangrove was

that a great variety of wood species was exploited

imported to the site. Indeed, Avicennia wood is

during the Harappan period and that environments

considered a good fuel that might have been desirable

different from the current ones were present. The

for specific purposes and the leaves are used as fodder

most significant point raised by the anthracological

for camels and cattle (Selvam 2007).

analysis at Kanmer is the presence of more water

than is currently the case. The relative abundance of

the vicinity of the site, as well as higher water levels

species related to riverine environments indicates that

in the Rann, finds further confirmation in the recent

a river course was probably flowing nearby the site

geomorphological survey conducted by Rajaguru and

(Figure 6 and Table 7). The presence of mangrove,

Shushama (2008) on the fluvial sediments around

although only in one context, suggests that the waters

the site. Although in a preliminary form, the authors

of the Rann might have been more permanent than

conclude that the water levels in the Rann were

today or that the sediments maintained a higher level

higher during the Mid-Holocene and that the nullah

- 137 -

The suggestion of a permanent watercourse in

C. Lancelotti and M. Madella

Figure 9

Kanmer: Microphotograph of one of the fragments identified as Myristica sp. that shows part of a scalariform

perforation plate (magnification 200X). The identification was achieved through comparison with published material (Inside
Wood Database, 2004-onwards).

Figure 10 Kanmer: Microphotographs of the TS of Wrightia tinctoria charcoal from context FP005 Z17 (magnification 200X).

bordering the site presented, during the Harappan

species for their fuel collection (Figure 11).

Phase low energy fluvial activity consistent with a

perennial course of water. The anthracological data

a single field season, and the number of samples is

shows that, during the Mature Harappan Phase,

rather limited, it has not been possible to trace fuel

people seem to have constantly relied on riverine

exploitation during the entire occupation of the site.

- 138 -

Unfortunately as the samples collected represent

Preliminary anthracological analysis from Harappan Kanmer

Figure 11 Kanmer: Presence and trends of ecological zones exploitation in the different context types.

Correspondence analysis of charcoal

Avicennia marina separating them on the second

assemblages

dimension by clustering with the fireplace/occupation

Correspondence analysis ran on the charcoal

level group.

assemblage shows that the samples separate into

three distinct clusters (Figure 12). Floor deposits

Discussion and conclusions

are separated from the other context type on the

first dimension, which is mainly characterised by
Calotropis procera, Prosopis cineraria, Senna siamea

Anthracolog ical analysis at Kanmer sug g ests

and Ficus sp. These four species have a contribution

that preservation of charred remains might be a

to the inertia greater than 0.1. Fireplace deposits and

problem. The extremely low concentrations of

occupation levels separate from ash accumulations

charcoal retrieved from excavation contrasts with

on the second dimension. Acacia senegal, Ziziphus

the variability of the species identified. The poor

nummularia, Cordia sp. and Salvadora sp. all

preservation of charcoal might be linked to the high

contribute to the inertia of the second dimension by

salinity of the soils, as salt crystallization increase

more than 0.1.

the fractures of charred plant remains and, when

Species indicators of specific ecologies tend to be

dissolved during flotation, disaggregate the charcoal

equally distributed in the three groups. For example,

pieces. Dry sieving might therefore be a more suitable

the two riverine forest indicators identified in the

method for recovering charred plant remains at this

samples (Acacia nilotica, Dalbergia sissoo) cluster one

site. The low density of samples at Kanmer does

with floor deposits and one with ash accumulations.

not, however, impair the validity of the analysis. In

However, among those 2 species, Acacia nilotica sums

fact, it has been shown that sample diversity is not

up to over 7% of the entire assemblage against the 0.4

correlated to density (Table 6) and for this reason the

Dalbergia sissoo. Therefore, it can be said that riverine

interpretation of the anthracological analysis has to be

species characterise mainly floor deposits. The other

considered reliable.

two groups are characterised by a mixture of dry thorn

scrubland species with the mangrove forest indicator

the analyses represents a richer variety of wood

- 139 -

The number of woody taxa identified during

C. Lancelotti and M. Madella

Figure 12 Kanmer: Correspondence analysis scatterplot for wood assemblages with samples grouped by context-type.

species than the one observed around the site today.

woodland habitats (although nilgai also live in open

The most abundant species recovered, Ziziphus,

grassland) and the four-horned antelope, in particular,

Salvadora and Capparis belong to the dry-thorn

do not live very far from water courses (Mallon

scrubland environment that characterise the area at

2008). Finally, also geological survey and analyses of

present. The anthracological evidence also suggests

the area support the hypothesis of a higher presence

that near the site once flowed a watercourse able to

of water around the sites. The work of Rajaguru and

sustain good vegetation cover and that the water

Sushama (2008) demonstrated that the nullah that

regime of the Rann might have been different from

flows nearby the sites was probably permanent at the

today, with higher water levels or longer periods

time of the Mature Harappan settlement.

of inundation. People exploited riverine species

and coastal resources in addition to the dry thorn

riverine species, combined with increasing aridity,

scrubland vegetation, which was dominant around

might have led to environmental depletion. The

the site. This reconstruction is also supported by

absence of Avicennia marina from the area today

the archaeozoological analyses that have identified

can be seen as a combination of environmental

Boselaphus tragocamelus Pallas (nilgai) and Tetracerus

and anthropic causes. Indeed, it has been shown

quadricornis Blainville (four-horned antelope)

in the United Arab Emirates that the combination

(Goyal and Joglekar 2008). These two species prefer

of environmental stresses, such as aridity and high

- 140 -

The intensive exploitation of dry scrubland and

Preliminary anthracological analysis from Harappan Kanmer

London.

salinity, with harvesting of plants for fuel and fodder

Chabal, L. (1997) Fortes et Socits en Languedoc (Nolithique

has a major impact over the reproduction of mangrove

Final, antiquit Tardive). LAnthracologie, Mthode

species (Howari et al. 2009).

et Paocologie (Documents dArcheologie Francaise

The presence at the site of species like Myristica

63). ditions de la Maison des Sciences de

malabarica, and Wrightia tinctoria, which were not

lHomme, Paris.

primarily exploited as fuel (although mangrove are

Chabal, L., L. Fabre, J.-F. Terral and I. Thry-Pariso (1999)

sometimes used as fuel as well), provides information

Lanthracologie, in C. Bourquin-Mignot, J.E.

on various activities. First, the Bombay Mace or false

Brochier, L. Chabal, S. Crozat, L. Fabre, F. Guibal,

nutmeg (Myristica malabarica) indicates contacts

P. Marinval, H. Richard, J.-F. Terral and I. Rhry

between this part of Gujarat and the Maharashtra

(eds.) La Botanique. Editions Errance, Paris.

coast south of Mumbai. As its name suggests, this

Fahn, A., E. Werker and P. Baas (1986) Wood anatomy and

species fruits are used as a surrogate for the more

identification of trees and shrubs from Israel and

expensive nutmeg and its presence at the site can be

adjacent regions. The Israel Academy of Sciences

probably related to the commerce of this spice. It is

and Humanities, Jerusalem.

improbable that this species was imported to be used

Gamble, J.S. (1902) A Manual of Indian Timbers: An Account

as fuel. More likely, the fragments detected in the

of the Growth, Distribution, and Uses of the Trees

sample came from a tool, a small object or an artefact

and Shrubs of India and Ceylon with Descriptions of

that was imported and subsequently burned. The

Their Wood-structure. S. Low, Marston & co. ltd.

Goyal, P. and P.P. Joglekar (2008) Report on the faunal

same can probably be said of Wrightia tinctoria whose

remains recovered from Kanmer, Gujarat, during

wood is currently used for carvings and furniture

the second field season (2006-07), in T. Osada and

making. Its presence on site can be linked with dyeing

A. Uesugi (eds.) Occasional Paper 5: Linguistics,

activities as its leaves produce a blue dye similar to

Archaeology and the Human Past. Research Institute

indigo (Orwa et al. 2009). The fact that these species

for Humanity and Nature, Kyoto. pp.25-43.

are found in fireplaces even though their primary use

Hather, J. G. (2000) The Identification of the Northern

was not as fuel might indicate the recycling of worn

European Woods. A guide for archaeologists and

out tools or other objects. Unfortunately, the absence

conservators. Archetype Publications, London.

of samples from later phases as well as the unclear

Howari, M.F., B.R. Jordan, M. Bouhoche and S. Wyllie-

stratigraphy at the site make it impossible to delineate

Echeverria (2009) Field and remote-sensing

a chronological sequence of wood exploitation at

assessment of mangrove forestsand seagrass beds

Kanmer at present, or to infer changes through time

in the Northwestern part of the United Arab

that can confirm the hypothesis of environmental

Emirates. Journal of Coastal Research 25: 48-56.

Inside Wood Database (2004-onwards) Inside Wood Database,

degradation.

http://insidewood.lib.ncsu.edu/search.
Kharakwal, J.S., Y.S. Rawat and T. Osada (2007) Kanmer:

References

a Harappan site in Kachchh, Gujarat, India, in

Asouti, E. (2003) Woodland vegetation and fuel exploitation

T. Osada (ed.) Occasional Paper 2: Linguistics,

at the prehistoric campsite of Pinarbas I, south-

Archaeology and the Human Past. Research Institute

central Anatolia, Turkey : the evidence from

for Humanity and Nature, Kyoto. pp.21-46.

Kharakwal, J.S., Y.S. Rawat and T. Osada (2008) Preliminary

the wood charcoal macro-remains. Journal of

obser vations on the excavation at Kanmer,

Archaeological Science 30: 1185-1201.

Asouti, E. and Fuller, D.Q (2007) Trees and Woodlands of

Kachchh, India 2006-2007, in T. Osada and

South India: Archaeological Perspectives. UCL,

A. Uesugi (eds.) Occasional Paper 5: Linguistics,

- 141 -

C. Lancelotti and M. Madella

Archaeology and the Human Past. Research Institute

Scheel-Ybert, R. (2002) Evaluation of sample reliability in

for Humanity and Nature, Kyoto. pp.5-23.

extant and fossil assemblages, in S. Thibault (ed.)

Mallon, D.P. (Access 2008) Tetracerus quadricornis, IUCN

Charcoal Analysis. Methodological Approaches,

2009. IUCN Red List of Threatened Species, www.

Palaeoecological Reconstruction and Wood Uses. BAR

iucnredlist.org.

International Series 1063. Archaeopress, Oxford.

Naidoo, G. (1987) Effects of Salinity and Nitrogen on

pp.9-16.

Growth and Water Relations in the Mangrove,

Schweingruber, F.H. (1990) Anatomie Europischer Hlzer/

Avicennia marina (Forsk.) Vierh. New Phytologist

Anatomy of European Woods. Verlag Paul Haupt,

107: 317-325.

Bern und Stuttgatrt.

Orwa, C., A. Mutua, R. Kindt, R. Jamnadass and S. Anthony

Selvam, V. (2007) Trees and Shrubs of the Maldives, Maldives

(2009) Agro Forestry Tree Database. http://www.

& Bangkok. Ministry of Fisheries, Agriculture and

worldagroforestrycentre.org/

Marine Resources & FAO.

Patel, R .I. (1971) Forest Flora of Gujarat State. Forest

Teng berg , M. (2002) Vegetation histor y and wood

Department of Gujarat State, Baroda.

exploitation in the Oman Peninsula from the

Rajaguru, S.N. and G.D. Shushama (2008) Preliminary

Bronze Age to the Classical period, in S. Thiebault

observations on Holocene fluvial sediments around

(ed.) Charcoal Analysis: Methodological Approaches,

Kanmer, Gujarat, India, in T. Osada and A. Uesugi

Palaeoecological Results and Wood Uses. BAR

(eds.) Occasional Paper 5: Linguistics, Archaeology

International Series 1063. Archaeopress, Oxford.

and the Human Past. Research Institute for

pp.151-157.

Humanity and Nature, Kyoto. pp.1-3.

- 142 -

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

(Field seasons 2006-2009)
Shahida Ansari
Lecturer cum Keeper in Museology
Deccan College Post-Graduate & Research Institute
Pune, India

Abstract
This paper is based on the ethnoarchaeological observations carried out at Kanmer and nearby villages during the field-seasons
2006-09. It attempts to brief the adaptation pattern through settlement and subsistence studies of the settled and ago-pastoral
communities (Rabari and Bharwad) of Kanmer. The various traditional knowledge systems, utility for storage facilities, hearth types,
pottery tradition, belief system, ornamentation, textile and tattoo are also discussed in this paper.

INTRODUCTION

Kachchh is land of few trees like nim (Melia

Azadrachta Linn.), pipal (Ficus religiosa), babul/bavar

Gujarat is a land of great contrasts; it stretches from

(Acacia sp.), tamarind, banyan (Ficus Bengalensis), and

the wet, fertile, rice-growing plains of the west coast,

mango (Mangifera indica); on the coast the coconut

north of Bombay city to the almost rainless salt

is difficult to rear as reported in Imperial Gazetteer

deserts of Kachchh in the Northwest. Kachchh (or the

of India 1908 (1908: 77). J.M. Campbell (1880: 30)

sea-cost land) comprising a single district is bounded

reports of large wild animals like lions and tigers in

on the south by the Gulf of Kachchh and on the north

Kachchh. The Imperial Gazetteer of India 1908 (1908:

and east is separated from Pakistan and the mainland

77) gives an account of large game, like leopards,

of India by the Rann of Kachchh, best described as

wild hog and smaller game like hare which occurred

a vast salt marsh covering about 18,000 sq. km. The

in large numbers; and the wild ass (Equus hemionus

Rann floods during the rainy season, slight though

Pallas) frequented the wastes of the Rann. G.D. Patel

the rains, and Kachchh is converted into an island; in

(1971: 36-44) in Gujarat State Gazetteers: Kutch

the dry season it is a sandy, salty plain plagued by dust

District, reports of panther (dipdo/Panthera Pardus

storms. To the south of Kachchh is the large peninsula

Linn.), Indian wolf (varu/Canis lupus Linn.), hayaena

of Kathiawar (Saurashtra), lying between the Gulf of

(taras/Hyaena hyaena Linn.), jackal (shial/ Canis

Kachchh and the Gulf of Khambhat (Cambay). The

aureus Linn.), fox (lonkadi/Vulpes bengalensis Shaw),

Little Rann of Kachchh extends northeast from the

wild boar (dukkar/Sus scorfa Linn.), blackbuck

Gulf of Kachchh and occupies about 5,100 sq. km in

(kaliar/Antelope cervicapra Linn.), wild ass (khar

Gujarat. Originally an extension of the Arabian Sea,

gaddha), Indian hare (saslo/Lepus nigricollis Guvier),

the Rann of Kachchh has been closed off by centuries

snakes and fishes. About 223 species of birds listed by

of silting. During the time of Alexander the Great

Captain C.D. Lester in the year 1898 and Salim Alis

it was a navigable lake, but it is now an extensive

list of 324 species of birds (both resident and migrant)

mudflat, inundated during monsoon seasons.

in the year 1945 in Kachchh are documented in this

gazetteer.

Settlement is limited to low isolated hills.

- 143 -

Figure 1 Map depicting villages where ethnographic survey was carried out

Shahida Ansari

- 144 -

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

Gujarat is one of the most intensively explored

(Mature Harappan) intensive burning activity is

areas (approximately 500 sites reported) in India,

reported inside the fortification wall, the garbage

of which more than sixty Harappan sites are from

thrown outside the fortification wall. The Late

Kachchh alone. The important excavated Harappan

Harappans (Period III) initially used the fortification

sites in Kachchh, Gujarat are ; Prabhas Patan

wall as defence wall, but subsequently its function

(excavated in 1955-57, 1971-72, 75-77), Desalpur (23

ceased to its purpose. They erected structures right

25 N.; 60 10 E.; excavated in 1964), Surkotada (23

on top of the fortification wall to make houses. The

27 N.; 70 50 E.; excavated in 1971-72), Pabhumath

Harappans moved out of Kanmer and after 1600

(23 40 N.; 70 44 E.; excavated for three seasons

years of gap, the Iron Age people occupied this

1977-80), Dholavira (23 52 N.; 70 13 E.; excavated

site. This historical deposit is marked by residential

for fourteen years beginning in the year 1989-90),

stone structures. A few medieval stone alignments

Shikarpur (23 16 N.; 70 41 E.; excavated in 1987-

belonging to medieval period are reported from the

90, 2006-09), and Junikuran (cited in Kharakwal et

site (Kharakwal et al. 2008).

al. 2008: 6).

R.S. Bisht and his colleagues of Archaeological

is supply of water and various other resources for

Survey of India were the first to locate the site of

sustenance. Interestingly, there are no permanent

Kanmer in the 1980s (IAR 1985-86: 15-19). The

rivers in Kachchh, but during the rainy season ( July-

excavations at Kanmer (23 23 N; 70 52 E), a

October) many streams of considerable size flow from

Harappan site in Kachchh, Gujarat, India, began

the central ranges of hills northwards to the Rann

in field season 2005-06 under the directions of

and southwards to the Gulf of Kachchh (Imperial

J.S. Kharakwal (Institute of Rajasthan Studies,

Gazetteer of India 1908: 74-75). For the rest of the year

JNR Rajasthan Vidyapeeth, Udaipur), Y.S. Rawat

the courses of these streams are marked by successions

(The Gujarat State Department of Archaeolog y,

of detached pools. Owing to the porous nature of

Gandhinagar) and Toshiki Osada (Research Institute

the upper soil, storage of water ponds and reservoirs

for Humanity and Nature, Kyoto, Japan). The site is

is difficult; but brackish water is readily found in

locally referred, as Bakarkot, and is situated about

rocks at no great depth from the surface and wells

200 m north of the present village and five and half

of this kind are fairly numerous. The site of Kanmer

kilometres from the Little Rann of Kachchh (Figure

is drained by a seasonal nallah known as Aludawaro

1). The mound covers an area of 115 x 105 m with a

Vokro and the village is surrounded by numerous

considerable height of eight metres from ground, and

wells, which serve as an important water source till

has impressive stonewall fortification, visible while

today.

The prerequisite of any successful settlement

nearing to the site (Figure 12b).

The site has yielded five-fold cultural sequence:

LAYOUT OF KANMER VILLAGE

Period I : Early Harappan, Period II : Mature

Harappan (Subdivided into II A and II B), Period III:
Late Harappan, Period IV: Historic and Period V:

The settlement pattern adopted at Kanmer village

Medieval. The site has impressive stone fortification

is according to the caste hierarchy. The central

wall (trapeze in shape, total thickness is measured

portion is occupied by those higher in the social

about 18 m) constructed during the Mature Harappan

scale, such as the landlords, money lenders, village

and later renovated/ strengthened during the Late

officials and higher caste families followed by those

Harappan period after a short gap. In Period II B

of the agriculturists, agro-pastorals, artisans and

- 145 -

Shahida Ansari

Figure 2 (a) Source of lelwa matti (b) Stone quarrying near Kanmer (c) Stone structure (d) Use of lelwa matti while construction
(e) Stone foundation for house (f ) Application of plaster to beautifying stone structures (g) Use of wattle and daub above
stonewalls (h) Use of sengtara above stonewalls (i) Use of dhauli matti and mirror for beautification of house walls
- 146 -

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

outskirts occupied by the lower caste group. Thus,

clay), dhauli matti (white shale procured from the

the occupancy within the settlement maintained

nearby hill section), lal leluwa (Reddish brown sandy

through generations are; Gadvi, Darbar (Rajput),

clay brought from nearby pond), pila leluwa (yellow

Rujput (Agriculturist), Bhar wad (Herdsmen),

sandy clay brought from the nearby pond), kali matti

Rabari (Herdsmen), Bavaji (Priest), Vania (Traders/

(black soil), mathadhowa ni matti (Fullers Earth), and

Jain), Musalman, Mali (Husbandmen), Koli 1) ,

rakhodi (ashy coloured clay/ located 3 km north along

Darva Maharaj, Luhar (Craftsmen- Ironworker or

the local river). These soils are used for construction,

Blacksmith), Nai (Barbers), and Harijan (Vankar/

flooring, beautification of wall, storage bins, pottery

Weavers). The Gadvi and Darbar are the most

making and also for medicinal purposes.

powerful group who occupies the central and the

best agricultural land in Kanmer. The communities in

stonewalls plastered with clay and cow-dung emulsion

Kanmer are mainly agriculturist, including Rabari and

(Figure 2f ). Stones are locally abundant around

Bharwad who are agro-pastorals.

Kanmer (Figure 2b). Those who are materially

The Gadulia Luhar's of Radhanpur are mobile

prosperous like the landowners and moneylenders

ironworkers who visit Kanmer once a year, i.e. on 15th

use burnt bricks for constructing houses. The binding

March, when they arrive at the village and stay for few

agent of stonewall is plastering clay called leluwa,

days. They are mainly invited by the villagers to repair

which they collect from the nearby area or by walking

agricultural tools or to make new iron objects. There

half a kilometre to get it (Figure 2a). Generally, in

are other groups of Gadulia Luhar, coming from

front of their houses small mounds of leluwa is kept

Rajasthan, who settle down at Adhoi and Anjar once

for repairing purposes.

in a year (Figure 8c). They make temporary shelters in

agricultural field in the month of January before they

sometimes by straw grass. Inclined roofs are mostly

move out at different places in Kachchh.

made. The slope in front is generally longer than the

At Kanmer, small, narrow roads or streets

one at the back, as the former encloses the verandah

branching off into lanes and bye-lanes usually separate

adjoining the main living room. In some of the

the rows of the houses. The linear arrangement

courtyard walls were also covered by thatched straw or

of houses is the most common pattern though

grass collected from nearby area popularly known as

isolated clusters are not uncommon. The houses are

sengtara (Figures 2g and 2h).

rectangular or square with single and double rooms

with common sidewalls and a common verandah in

2008) the Mature and Late Harappans floors at the

front surrounded by central courtyard. All houses

site were first levelled by packing base with yellow

have a verandah where members of family sit and

sandy clay (pila leluwa) mixed with fine clay (khotedo).

carry out household chores, serving as living cum

Above it was less than 5-cm thick bed of reddish sandy

meeting room. A separate kitchen is seen among the

clay (lal leluwa) sometimes even rammed. Finally a

Darbar, Rujputs, Bharwad, Rabaris and Harijans. A

thick plaster of white shale (dhauli matti) mixed with

courtyard encloses the open spaces in front of the

cow dung was applied (plaster sometimes were 15 cm

house of agriculturist where they keep cattle and the

thick). The Mature Harappan floors were thinner

artisans pile up their household industry.

compared to the Late Harappan floor.

Almost all villagers have wide knowledge of

Most of the houses have either mud walls or

Roof is generally covered with country-tiles or

According to the excavators (Kharakwal et al.

Presently, the villagers of Kanmer, Ghanithal,

soil of the region. At least seven type of soil is easily

Gagodar and Palanswa (surrounding villages) follow

recognised and known locally as: khotedo (pond

almost similar kind of flooring technique as employed

- 147 -

Shahida Ansari

Figure 3 (a,b,f ) Movable clay hearths (c,d,e,g,i) Immovable hearths (h) hearth made by potter

- 148 -

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

Figure 4 Dumping of ash, cow dung, other waste, outside the courtyard or lanes,
which is later collected to be spread over the agriculture field as manure

by the Harappans at Kanmer (Figure 1). Except the

storage bins (kothlas), walls and door frames of their

foundation or the base of the house is made of

houses, are decorated with embossed designs studded

m thick quarried stone (brought from the nearby

with abhlas (tiny pieces of mirrors), show the innate

hillocks), top of layered stone is three inch of fine

artistic taste of this agro-pastoral community. In some

clay (khotedo), above which 5 inch thick deposit of

of the houses the doorways have torans embroidered

yellow sandy clay (pila leluwa/sturdy as cement) is

or made of beads, while chaklas or square pieces of

put, before finally ramming it to get even floor base

cloth similarly worked in beautiful traditional designs

(Figures 2d and 2e). The floor above this is made of

are the most popular mode of wall decoration (Figure

1 inch mixture (thep) of black soil (kali matti) + cow

2i).

dung + water which is sealed with thick plaster (gar)

prepared of white shale (dhauli matti) + water. The

associated with the Mature Harappan floor, the

women of the house repeat the application of plaster

diameter of which was measured 30 cm (Kharakwal

of white shale + water regularly sometimes within

et al. 2008). It has survived up to a height of 30 cm.

three days or weekly (Figure 2i).

Presently; the cooking hearths in Kanmer are of

The artistic taste of the people is also visible in

two types, i.e. movable and immovable (Figure 3).

the interior decoration of their houses, which they

Movable hearths are made of clay (Figures 3a, 3b

try to beautify in variety of ways. The arrangement of

and 3f ) and other can is acquired/purchased from

wooden chest called pataras, household utensils, and

the potter (Figure 3h), whereas immovable ones are

other belongings in one of the interior room of the

made with stone and clay (Figures 3c, 3d, 3e and 3g).

houses is attractive. The clay work done by Rabaris on

Cooking area is separate from other rooms and is

- 149 -

A cooking hearth (U-shaped ) was found

Shahida Ansari

generally located at one corner of the entrance of the

of ash (sheli/from hearth) and then 6-inch layer

house (Figure 3i). Firewood and cow dung cakes are

of sand (reth/brought from nearby stream) was

used as fuel.

evenly spread over. Above this jowar grains were

The garbage dump comprising of cooking waste,

filled in. Before closing the pit 6 inch layer of sand

cow dung, hearth ash are generally dumped outside

was spread over, above which 1 foot high mound

the courtyard wall, or the lanes, which appears like

(tekada) was made of clay + straw plaster, which

a pile from a distance These piles are later carried to

covered the khann.

agriculture field as manure (Figure 4).

Type 2: At Palanswa, the khann, was located in
front courtyard of the house. It was circular in

PIT SILOS

shape, 1 m deep and 1.50 m wide. At the base of the

pit, 10-cm thick layer of ash (sheli) from hearth was

Staple foods, especially grain, are usually kept

evenly spread. Above the ashy layer, jowar stalks,

sometime in storage before being consumed, so that

(twisted and made into thick rope-puda), were

the preservation is a matter of importance everywhere.

placed, it also covered the pit wall. Above the stalk,

The underground grain storage facilities found in

old blanket (godadi) was placed and then jowar

archaeological sites are popularly referred as pit

grain was filled in the pit. On top of the grain,

silos, storage pits or corn pits. This seems one of the

old blanket was placed. Then 10-cm thick ashy

simplest of all that have ever been tried by agricultural

layer was put before closing it. A small mound, i.e.

people through time. A number of pits and pit silos

tekada made of clay + straw covered the khann.

are reported at the site in structural complex (St. no 1)

belonging to the Historic period in the northwestern

part of the mound (trenches R21, S21, T21 and U21)

in khann. In this storage facility the grains could be

at Kanmer (Figure 12a; Kharakwal et al. 2008: 11).

easily stored for two to three years. They were opened

Ethnographic sur vey in Kanmer and Palanswa

only if there was real need for grains. The grains in the

(nearby village) has envisaged our knowledge of

khann did not have any effect of rain, as it is scanty in

circular underground grain storage pit, i.e. khann.

this region. And if the rains did spoil the stored grain,

This technique was used to store millets, i.e. lal

the grain used to turn black and pungent smell used

jowar (Sorghum sp. Moench.). This technique is now

to emit from the khann. The use of ash and sand is for

abandoned due to frequent earthquake in this region.

dissuading insect or ant attack in the storage.

The traditional knowledge of khann, however still

2)

Only the surplus grains were kept for storage

A recent ethnographic study (2007-8) of

survives among the elderly people of the two villages,

underground grain storage technique, i.e. Khanni in

who inform that this technique used to safeguard

Orissa (eastern India) strengthens our understanding

their grain from thief. Two process of khann narrated

of similar knowledge system of storage of rice/paddy.

by the villagers are as follows:

Two types of underground storage facility, i.e. the

half-yearly underground storage popularly referred

Type 1: At Kanmer, the khann, was located in front

as Vedi Khanni (Figure 5a), rectangular in shape with

of the house. A circular pit, 1.5 m deep and 1.75 m

a raised platform above the ground) and the year-

wide was dug and plastered (gar) with three inch

round storage (circular in shape, slight platform above

thick layer (thar) of leluwa + cow dung was applied

the ground) popularly known as Kuan Khanni are

on the pit base and wall. Above this 10 cm layer

prevalent (Figure 5b). The soil type defines which type

- 150 -

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

Figure 5 Line drawing representing processes of Underground grain storage technique, i.e. khanni
(a) Veddi khanni (b) Kuan khanni

of khanni is to be employed in the area. Generally in

khanni is practised where the soil is sandy loam.

the coastal area, vedi khanni is practised where the

soil type is clayey loam and in the inland area, kuan

following (Figure 5).

- 151 -

The technique employed for khanni is as

Shahida Ansari

can be closed after reopening. There is no effect of

1) The grain used for storage is rice/paddy. The

rain on the grain. But care has to be taken as no water

grains stored in khanni cannot be used as seed for

logging occurs during rain in khanni area, which can

next cropping as it losses its germination quality.

lead to water seepage. This traditional technique of

2) For location of khanni the areas preferably used

underground storage continues in Ganjam district

is in front/rear of the house or in the farmyard. The

of Orissa, as the locals prefer this khanni rice for

choice of area for storing grains largely depends on

consumption. Besides, that the grain can be stored in

two factors, i.e. the amount of grain to be stored

bulk with 0.01% damage from insects and rodents,

and the elevation.

the khanni rice is easy to dehusk, non-sticky when

3) Firstly, a pit having depth of 3 to 4 feet in desired

cooked, and can be easily digested by the elderly

shape of square/circular is dug. The pit is left for

people.

drying for at least five to ten days. The depth of

khanni mainly varies according to the amount of

village also make bambaiyyu (bohera, in north

grains to be stored.

Gujarat), an underground storage chamber at the

4) A thick layer of rice straw (nada) is placed at the

basement, which has two-top window opening (bari/

base and walls of the pit. Men continuously make

jali) (Figure 6f ). This chamber is accessed through

rope (bara/bada) of rice straw, which is to be added

staircase. Thick layer (thar) of ash (shaeli) is kept at

at a regular interval. They place the straw rope at

the base, before the grains are spread over it evenly.

the centre and rotate it in a clockwise direction

While filling the grains in chamber sometimes these

touching the sidewalls till it reaches the top. Again

windows are also used, from where the grain is poured

a layer of straw is placed in the pit before grains are

from the top and it also works as ventilator. Six-inch

poured in it.

layer of sand/river gravel (reth brought from the

5) When it reaches the ground level a new round

seasonal river/nallah) covers the grains. Finally, the

of straw rope is added above ground for half or one

windows are sealed with a mixture of clay or mud, as

metre more. Straw is then added on the side before

the heat should not touch the grain in the storage.

pulling the soil for support for the raised rope wall.

6) Five pieces of dried cow dung cake is kept on

considered pure, is sieved (by chalni) when the grains

the grains, which serve as insecticide, after which

are removed for use. Year round storage is easily

a layer of straw is put. Above the straw cover,

possible in bambaiyyu. During rainy season no effect

straw rope is again laid down from side to top in

on the grain can be seen, as rains are insufficient to

clockwise fashion. Straw is again laid down on this

flood it. The scanty rainwater is generally absorbed by

and then lump of soil and soil is put on the khanni

the soil within minutes after rain, besides the houses

and left overnight for settling down.

are always located on elevated area. This technique is

7) This is then plastered by thick coat made of

also being abandoned slowly as many of the people

clay + cow dung + water, at repeated interval

have filled these storage chambers with soil and bricks

throughout the year.

after a series of earthquake in last ten years in this

The Rajput and Jain communities of Kanmer

The sand broug ht from seasona l river is

region. The grains preferably stored in bambaiyyu

In vedi khanni, the grains are removed from the

are jowar, mugh, and bajra. Cotton is usually sold

pit after the first rain in June (Figure 5a). But for kuan

immediately after the harvest and wheat cannot

khanni the storage lasts for years together (Figure 5b).

withstand the underground heat.

The grains can be removed at any part of the year and

- 152 -

The advantages of underground storage structures

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

Figure 6 Storage techniques (a, b, c, e) Kotha-kothi (d) Chaursi or Mujuguda (f ) bambaiyyu

- 153 -

Shahida Ansari

The Chaursi or Mujuguda is square shaped with

is that they are free from the seasonal variations in

temperature and humidity provided that adequate

four-legged base (Figure 6d). It has two opening; one

precautions are taken against the seepage of water in

from the top and the other is a wooden window in the

the structures, especially in areas where the water table

body of the bin. Kotha is rectangular or cylindrical

is high. Underground storage structures are also safer

in shape. Sometimes grains like math and mugh are

from threats from various external sources of damage,

mixed with sand before storage. This technique is

such as theft, rain or wind. However, they are not

employed especially by the Rabaris who believe that

good when the quantity available for storage is small,

grains can be stored for a longer period of time and

and there are a number of grain varieties to store. The

insect attack is less. When the grains are used they are

underground storage structures are easier to fill up

sieved.

owing to the factor of gravity. However, it becomes

cumbersome to take out the grains from these

kotha-kothi was observed at Lakshman Bhai Rujputs

structures. Whereas the advantage of surface storage

house, which was prepared by his great grandfather

structure is that they can be maintained in a hygienic

(Figure 6a). He continues to use it now, for which

condition by cleaning or plastering at regular interval.

they occasionally plaster it with dhauli matti. The

The danger of heating up of grain due to internal heat

interesting feature of this bin is that the two walls of

is lesser. Thus, the agricultural communities employ

the house support the long rectangular bin (earlier

both the techniques.

three walls had bin facility, one of which was removed

The pearl millet (bajra) with stalk is also kept

recently). It has openings from the top, from where

in open storage, i.e. kanduaa, a stack made of

the grains are poured in and also has wooden windows

bajra straw/stalk, at the centre of which they keep

from where the grains can be accessed. This storage

bajra. Storage of bajra can be for months together.

facility is also used for keeping blankets and clothes.

Sometimes a room is also converted for storing bajra.

All kind of grains can be stored in kotha-kothi.

At Palanswa a rectangular wall storage bin, i.e.

The aboveground clay storage facilities popularly

used by Rujputs and Rabari of Kanmer and Palanswa

CROPS GROWN AT KANMER

is known as Kotah-Kothi (Figures 6a, 6b, 6c and 6e),

Chaursi or Mujuguda (Figure 6d). A variety of grain
like bajra, mugh, math can be stored in them. These

In Kachchh, saline and alkaline soils cover the area

storage facilities are also used for keeping blankets,

to the north of Lat. 23 N. A narrow strip below

utensils, milk, cooked food and vegetables.

this consists of red and brown soils. In the rest of the

For ma king these clay bins a m ixture in

district, black soils prevail deep in the western half

proportion of 2 (khotedo/pond clay): 1 (leluwa/sandy

and medium elsewhere except for a narrow strip in

clay): 1 (sheli/hearth ash): 1 (laad/horse dung ): 1

the east with grey soils (NCA 1976: 6). The Kachchh

(gobar/cow dung ), with water and chaff of bajra/

falls in the lowest rainfall zone, which ranges from 26

wheat is prepared. Once the clay bin is prepared to

to 40 cm, in July its about 15 cm, followed by August

a desired shape, proportion made with mixture of 1

with 8 cm rainfall (NCA 1976: 6).

(dhauli matti/white shale): 1 (laad) is applied on the

bin. To keep the bin cool dhauli matti mixed with

fed crops, sown in June-July and harvest in October-

water is applied with the help of soft cloth; this gives

November. This is the main cropping pattern in the

a bright look to the bin. It is also decorated with

area as no important crops are grown in rabi season

appliqu designs and mirror work (Figure 6d).

in the absence of adequate irrigation facilities. The

- 154 -

The villagers of Kanmer grow kharif or the rain-

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

rabi or irrigated crops sown in September-October

morning.

and harvested in February and March is only a recent

phenomenon. The principal crops of the region are

circular kiln or bhatti to bake his pots (Figures 7a and

bajra, jowar (kharif), fodder, other pulses, cotton and

7b). The placement of the kiln is in his courtyard. In

groundnut occupying more then 90 % of the cropped

the kiln at the base he places dried acacia branches

area (NCA 1976: 9).

(35 cm thick layer) or hay, broken pots and the pots

At Kanmer the main crops are pearl millet (bajra/

to be baked. Above this another layer of dried babul

Pennisetum typhoides), jowar-millet (jaar/Sorghum

branches is placed, a cover of broken pots and tavdi

sp. Moench.), cotton (kapas), and wheat. The pulses

(pan) to be baked are kept. Above this layer of hay

like green gram (mugh/Vigna radiata) and mat bean

(covering half of the kiln), ash (sheli) is spread. A pot

(math/Vigna aconitifolia) are grown. Other food crops

with green leaves is lighted and the kiln is left to burn

are groundnut (Mung-phali/Arachis hypogaea), sesame

during the night. If the twigs have been well laid, the

(til/Sesamum indicum), castor (erunda/Ricinus

fire burns slowly and evenly throughout the entire

communis), cumin (jeeru/Cuminum cyminum),

period of baking. The baked pots are collected in the

red gram/pigeon pea (tuwar/adad/Cajanus cajan),

morning.

horse gram (kalthi/Dolichos biflorus), gram/chick

pea (chola/Cicer arietinum), and Indian butter bean

follows.

Khemji Bhai a potter of Gagodar use semi-

The important observations at potters yard are as

(val/Dolichos lablab), Indian oat (javawata/Avena

byzantia) and barley (Hordeum vulgare). They also

1) Three kind of soil is used for making pots- black

grow vegetables and fruits like onions (dungri),

soil, red soil and yellow soil, which are collected

spinach (taka), brinjal (rengana), tomatoes (tamata),

from three different locations around the village

chilli (mirch), garlic (lahsun), valor, galaka, turiya,

(Figure 7d). For making tavdi they use only black

givada, guvar, bhinda (ladys finger), kakoda, tinda

and yellow soil.

and watermelon (kariga).

2) The pots are made on wheel (chak), wooden

beater (thapa) and stone dabber (pinda) are used
for shaping pots (Figure 7g). The wheel is placed

POTTERY MAKING TRADITION

on a pivot stone, which has perforation at centre

(Figures 7e and 7f ), having smooth but pointed

At Kanmer wheel-turned wares having painting of

wood on which the wheel rotates. Stone is used to

leaves, fish, bands of wavy lines, diamond shapes,

sharpen this wood at regular intervals. The potter

executed with black or dusky red pigment are

use thread to cut the base of the pot on the wheel.

reported from all the cultural period. To understand

3) To decorate they use white, red, dark maroon

the pottery tradition in the region the potters

colours. Soft coloured stones having yellow and

(Kumbhars) of Gagodar and Palanswa villages were

red colour are collected from the seasonal nallahs,

visited, who are the local suppliers of variety pots to

and soaked in water before application. They make

the near by villages including Kanmer. At Palanswa

variety of painting , which includes geometric

they use open-fire in a circular depression in the

symbols, flowers, animal (scorpion, fish and crab)

ground, composed of cow dung cakes, stalk of bajra,

and birds. The girls and womenfolk generally do

babul branches, surrounded by a ring of broken pots

paintings (Figure 7i).

(Figures 7a, 7b, 7c). The kiln is lit in the afternoon,

4) They make variety of pots according to the

allowed to burn all-night and dismantled the next

requirement and season. For example pots (ghado)

- 155 -

Shahida Ansari

Figure 7 (a,b,c) Potters kiln (d) Soil types for making pottery (e) Potters wheel (f ) Bored stone (g) Beater and dabber (h)
Variety of pots (i) Painting the pots (j) Pot headrest made of coloured glass beads

- 156 -

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

with narrow neck (for keeping water), wide neck

the oppression of the Sumras. The Chavada Rajputs,

(cooking rice, keeping butter milk, setting curd),

who then ruled over Kachchh, granted the Sammas

pan (making chapati), and basin (kneading dough).

a tract of land; but in the time the latter subverted

During marriage season the potters make a variety

the rule of the Chavadas, and reigned in their stead

of pots (Figures 7h and 7i).

(1320). The sections of the Sammas forming the ruling

5) The villagers from as far as 15 to 20 km come to

family in Kachchh were known as Jadejas. Till 1540

collect various kinds of pots from Khemji bhai.

the Jadejas ruled over Kachchh in three branches;

He listed the name of the villages like Kidiyanagar,

until that year Khengar, the son of Jam Hamir, with

Nadiyadimb, Paragparthi, Thoriali, Sankapar,

assistance of the Muhammadan king of Ahmedabad,

Kanmer, Ganithal, Malampar, Vanthia, Chotapar

succeeded in making himself head of the tribe and

and so on.

master of the whole province. For six generations

after Khengar the Raos succeeded according to
primogeniture till 1762 when Ghulam Shah Kalhora,

HISTORY OF THE REGION

ruler of Sind, taking advantage of the disorders of

the State, twice invaded Kachchh with some success

The ethnographic account of the orig in and

in 1762-5. The disorder became intensified by the

movement of the Rabaris and Bharwads is important

insanity of Rao and the struggles of rival factions, one

in the light of understanding the early historic data

of which was headed by minister Fateh Muhammad,

available to us from Imperial Gazetteer of India (1908:

until in 1809 the help of the British Government was

77-79, 85). It gives account of the earliest historic

sought to restore order. In 1815, British force moved

notices of Kachchh, by the Greek writers. The waters

in Kachchh to restore order. The regency ( Jadejas

of the Rann were navigable lake in Alexanders time

nominated a minor to be successor, with the British

(325 B.C.) and a shallow lagoon at the date of Periplus

Resident and few chiefs as regency) was closed in

(third century AD), and there are local traditions of

the year 1834 when Rao Desalji became full king.

seaports on its borders. About 155-130 B.C., Kachchh

Followed by Pragmalji II (1860-75), Khengarji III

was part of the Menanders kingdom, and shortly

(1876-1942), Vijayrajji (1942-1948), and Madansinghji

afterwards passed to Parthians. Between AD 140 and

(1948). After independence Kachchh was placed

390 the Kshatrapas of Saurashtra ruled Kachchh. It

directly under the Central Government, Government

was included for a time in the Gupta Kingdom of

of India. On the bifurcation of the Bombay State on

Magadha and was ruled later by the Vallabhi kings.

1st May 1960, the Kachchh district became a part of

In the seventh century Kachchh formed part of the

the newly formed Gujarat State.

province of Sind. Hiuen Tsiang refers to it as Kiecha. Invasions of Charans, Kathis, and Chavadas

ETHNO-HISTORY OF
RABARIS AND BHARWADS

followed. In the 9th century the Arabs settled on the

coast. In 1023 Bhima Deo I of Anhilvada fled before
Mahmud of Ghazni to Kandhkot; and at the close of
that century the peninsula was overrun by Singhar,

The earliest reference on Rabaris is by Lieutenant-

the fourth Sumra ruler of Sind.

Colonel James Tod (1873: 289-293), who lists the

The modern history of Kachchh dates from its

Rebarris as the tribe inhabiting the desert and valley of

conquest by the Sind tribe of Samma Rajputs in the

Indus. The term Rebarris according to him is known

14th century. The Sammas fled to Kachchh to escape

throughout Hindustan to denote persons employed

- 157 -

Shahida Ansari

in rearing and tending camels, are always Moslems. In

Kachchh), Kathiawar, Palanpur and Ahmedabad.

Rajputana they are a distinct tribe employed entirely

They are identified as an immigrant tribes of

in rearing camels or in stealing them, in which they

herdsmen who were formerly resident in Marwar and

3)

evince a peculiar dexterity , competing with the

Sind, and perhaps at a remote date in Baluchistan.

Bhats in the practice. S.C. Dutt (1884: 53, 55-56)

They worship the g oddess Hinglaj and make

lists Rebarris as one of the principal wild tribe lying

pilgrimages to her shrine in Baluchistan.

between Rajwara and the Indus. He identifies them

as shepherds, goatherds and camel breeders, and has a

Tribes and Castes of Bombay takes account from the

name for stealing camels, in which they are said to be

notes provided by M.K. Mehta, states that the Rabaris

dextrous.

of Kathiawar and Kachchh rear cows, buffaloes,

B. Kirparam (1901: 287-289) in his compilation

goats and sheep, and camels. They were also known

work in Gazetteer of the Bombay Presidency, Gujarat

as Bhopa, Mogha, Raika, Vishotar and Sanai in

Population: Hindus, gives two origins of Rabaris

Cutch. Jyotrindra Jain and Mallika Sarabhai (1985:

firstly, Shiv while performing religious penance or tap

46) mention that the Rabari of Kutch (presently

created a five-footed camel and a man to graze it. This

Kachchh) is divided into three sub-castes, i.e. the

man had four daughters who married Rajputs of the

Kachhela of western Kutch and the Dhebaria and

Chohan, Ghambir, Parmar and Solanki tribes. These

Vagadia of eastern Kutch.

and their offspring were all camel grazers. Other

Rajputs joined them and they formed a separate caste.

Rabari of Gujarat believe that their forefathers were

Secondly, that they were Rajputs, who instead of

the inhabitants of Dwarka and Brindavan, the two

marrying Rajput women, married celestial damsels or

places associated with Lord Krishna, subsequently

apsaras that, is perhaps Charan women, devputris or

they migrated to Jaisalmer in Rajasthan. He provides

daughters of gods as they style themselves. According

us with their distribution in the districts of Mehsana,

to this account because they did not marry Rajput

Surat, Ahmedabad, Baroda and Sabarkantha. Their

wives they were called Rahabaharis, i.e. goers out of

habitat ranges between arid and semi-arid climate

the path. Their original home is North-West Provinces

zone. They are traditionally and primarily cowherds

from which they moved to Mar war and from

and milkmen. A few of them are small and marginal

Marwar to Gujarat Kathiawar and Kachh (presently

farmers and also work as agricultural labourers.

Kachchh). This story is supported by the fact that

the chief seat of Sikotra, the tribal goddess of some

word Rabari comes from the words rah and bari

Rabaris, is at Jodhpur. He identifies them as herdsmen

meaning whose ways are quite distinct from those

who rear cows, buffaloes, goats, sheep and camels in

of other people, i.e. those who take to outlawry and

Kathiawar and Kachh. With their droves of cows and

nomadic life. He traces their history to be at first

buffaloes they move about the country in search of

the inhabitants of Baluchistan, where they occupied

pasture. They make their living chiefly from the sale of

rocky, mountainous, barren and sandy places. They

clarified butter and the wool of sheep. A few among

kept number of camels, which were reared for long

them work as husbandmen and messengers.

journeys in barren tracts. They milked these camels

M.K. Mehta provides the other important

and made coarse cloth out of their hair, prepare

information on Rabaris in 1909 to the Superintendent

ghee from milk of the dromedaries (Arabian camel

of the Ethnoagraphic Survey (1909: 1-6) stating

with one hump) for use and sale. Due to population

that Rabaris are chiefly found in Cutch (presently

increase they migrated to Marwad and settled there.

- 158 -

R.E. Enthoven (1922: 252-258) in his book- The

According to K.S. Singh (1997: 1011-12) the

B.L. Mankad (1939: 31-32) mentions that the

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

Suitable climate was agreeable and profitable for

account of the term Bharwad as a modified form of

their business. They soon found that their camels

the word badawad, i.e. bada meaning sheep, and

and dromedaries were very useful for transport and

wada meaning a compound in Gujarat. The persons

speed message. They were recognised for skill for

who possessed compounds with sheep were referred

managing and training camels that they supplied

to as the Badawad, which in course of time came to be

to the States, for warfare activities. Marwad too

pronounced as Bharwad. Interestingly, their name is

proved a small country for Rabaris so they migrated

also related to the Bharade, a cattle-herder community

to Cutch. For some time they lived in Cutch before

of Madhya Pradesh. According to Singh (1997: 114)

migrating to Kathiawad. In Kathiawad they were in

their traditional occupation is goat and cattle herding,

different geographical feature having fertile soil and

the Bharwad also work as agricultural and non-

rich pastures led them in breeding buffaloes. In short

agricultural labourers. They sell milk in towns and a

period of time their clarified butter or ghee from

few are engaged in marginal cultivation.

Kathiawad went as far of India and Africa.

B. Kirparam (1901: 267-285) in his compilation

ORNAMENTS OF
RABARIS AND BHARWADS

works in Gazetteer of the Bombay Presidency, Gujarat

Population: Hindus, gives account of Bharvads as
shepherds who are found all over Gujarat. According
to one story, the Bharvads are the same caste as the

The Rabari and Bharwad men and women commonly

Mehers, to whom Krishnas foster-father Nand Meher

wear certain ornaments. These included kanthi

belonged. According to another story, they are the

(Figure 8a) around the neck, a silver kandora around

descendants of a Vaishya father and a Sudra mother.

the waist and a vinti (ring ) on one or more of the

Their original home is said to be Gokal Vrindavan

fingers. The use of ivory is extensively found among

near Mathura from where they are said to have moved

Bharwads and Rabaris in this region. The social and

to Mewar, to spread into Gujarat, Kathiawar and

economic status of the household or the class of

Cutch. They are closely related to the Rabaris with

society to which they belong mostly determines the

whom they eat but do not intermarry. The Bharvads

use of gold and silver ornaments they adorn.

are husbandmen and labourers, but the bulk are

shepherds or cattle-keepers. Their flocks of sheep

women are; heavy ivory (hansu) chuddla (thick

and goats are kept in the outskirts of villages and

hand bracelet, purchased from Ahmedabad and

are driven into the grass and bush lands by day and

Maniara) worn in forearm and is a marker ornament

brought back at night. Between November and June

of their caste (Figures 8e and 8g ). They also have

they move about the country in search of pasture, they

plastic imitation of chuddla, which are made and

sell goat, cow milk, weave, and sell woollen blankets.

sold at Bhachau. Other ornaments for hand include

They are also paid in grain or in cash for penning their

mandriyo (bracelet) and bangdi (bangles).

flocks in empty fields, as the manure is highly valued.

Instead of sheep and goats, many Bharvads keep

Superintendent of the Ethnoagraphic Survey (1909:

cattle, both cows and buffaloes, and make their living

2) provides information that in 1315 AD the Rabaris

chiefly by selling clarified butter as also seen among

took refuge with the Sammas (of Sind) and went

the present day Rabaris.

with them to Cutch, supplying them with milk from

R.E. Enthoven (1920: 118) has referred to the

their herds of camels and cows. But they never forgot

Bharvad as the Gadarias. K.S. Singh (1997: 113) gives

their former patrons, the Sumras, and to this day,

- 159 -

The traditional ornaments worn by Rabari

Interesting ly, M.K . Mehta in 1909 to the

Shahida Ansari

their women do not colour their ivory bangles but

stamped with the petals of a flower.

wear them white as a sign of mourning. Though the

present day Rabaris at Kanmer are unaware of this

girls of Saurashtra/Kathiawar wear a leaf or diamond-

information, they continue to adorn themselves with

shaped flat gold ornament with floral designs in

heavy ivory ornament, i.e. hansu which still remains

relief in their upper ear. Because of its leaf shape it

the most enigmatic ornament of Rabaris.

is called pandadi (leaf ). It is the prominent use of

It is interesting to note here that young Mer4)

The vedhala are typical silver earrings in the form

gold distinguishes Mer ornaments from those of

of a pile of four rhombi, worn in the upper ear, and

Rabaris ( Jain and Sarabhai 1985: 46). At Kanmer

kamp in earlobe (Figure 11e). The vari are rings worn

2008-09 a gold leaf is reported from the excavation.

in the upper ear by unmarried girls only. The kadla

The ornament study between the two communities

(Figure 8b) are worn in anklets, kanthi, twisted silver

has shown that ornaments worn are related to their

wire ornament worn in neck, necklace made of plastic

custom, belief and tradition. From childhood to

buttons, black beads and gold beads, kaieda is heavy

marriage they amass and adorn typical ornaments

silver finger rings and chametiya the toe rings.

for the occasion. Some ornaments represent as caste

The Bharwad women wear chudo/sudo (Figures

marker.

8i and 11b), a large cylindrical ornament covering

forearms, khoto balaiyaan are the false ivory imitation

made from silver. The most important centres of

ornament popularly worn by them in forearm

silver ornament making in the Gujarat are Rajkot and

(Figure 11c), lockit in upper arm. They wear silver

Ahmedabad. Bhuj, Anjar and Mandhavi in Kuchchh

ornament like vedhala in her upper ear, doiyana and

are famous for relief, filigree and enamel work. The

pokhadadi in earlobe, phul as nose pin. The Ramrami

castes that have worked over generations are the

and Sanpagalu (gold/silver/ivory plated with gold/

Soni and Luhar. The selected crafts of Gujarat of

silver strips) are the variety of gold necklace worn by

Census of India 1961 discuss that ivory bangles were

married women only (Figures 8f, 11a and 11b). The

manufactured in all-important towns by artisans of

kadala are worn as anklets, they also have variation in

Brahmakshatriya caste. The important centres were

anklets known as kambiyun, i.e. anklets with dropping

Jamnagar, Jamkhambhalia, Wadhwan-Surendranagar,

at each side, silver kaieda are heavy finger rings and

Dhrangadhra, Palitana, Mandvi, Anjar, Bhuj,

chametiya toe rings.

Ahmedabad, Surat, Kodinar, Palanpur, Patan and

The Bharwad and Rabari men predominantly

Baroda. There is special mention of Mahuva town of

wear silver ornaments. They usually put on a dokiyu

Bhavnagar district, which ranks first on account of

made of beads of coloured glass and gold around their

ivory carving workmanship (Trivedi 1967: 23).

neck (Figures 9a and 9b). In the upper ear they wear

gold or silver button called bhungri, through a hole

Kanmer, made of steatite, agate, carnelian, lapis lazuli,

in the centre of which is passed a gold ring called ver,

terracotta, shell, faience, and gold. The steatite beads

and just under is thoriyan (Figure 9c). Bharwad men

however outnumber the collection (Kharakwal et al.

will always wear handsome bracelet, i.e. kadla, of

2007). The nearest source of agate around Kanmer

heavy filigree work, with a flower and very small box-

is in the little Rann of Kachchh. This is an isolated

like cylinders welded on to it which is also the marker

piece of land, known as the Mardak Beyt/Mardakh

ornament of the caste (Figures 9d and 9e). Some

Bet, covering an area of 2,100 acres, is also an existing

men also wear silver finger rings. Some Bharvad boys

mining source for agate (akik), chalcedony and jasper

wear a silver necklace and a bracelet, whose end are

deposits (Figure 1).

- 160 -

The jewellery of rural Kachchh is predominantly

A number of beads are reported from the site of

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

Figure 8 Ornaments of womenfolk and tattoo (a) Bharwad and Rabari girls wearing kanthi (b) Kadla are
worn in anklets (c) A Gadulia Luhar women (d) A Bhil women (e, g) Ivory (hansu) made chuddla (f ) A Kumhar women with
gold ornaments (h) Golibar ( i, j, k) Tattoo on body parts
- 161 -

Shahida Ansari

R .K. Trivedi (1964: 10), gives an interesting

After the tattooing operation is over, turmeric powder

information regarding the Bhils (a hunting-gathering

is applied to the area tattooed for its antiseptic value.

community of Gujarat) who used to pick up the

For more than a week the tattooed area is avoided

Kapadvanj agate, chiefly found in the beds of the

with any contact of water (at times if tattoo is spoiled

river Majam and sold them to Mandva Bohras, who

the colour doesnt come on the tattooed area when

disposed them off to Cambay stone merchants. This

washed). This process is generally painful and has

piece of information gives a wonderful example

resulted to periodic death. This traditional technique

of how hunting-gathering community in 1960s

of tattooing is however banned by the Gujarat

interacted with traders (A Bhil women at Adesar

State Government now. At Kanmer the Rabari and

Figure 8d). Such interaction in the region around

Bharwad women though still continue to tattoo

Kanmer will be looked into in future studies. The

through traditional as well as by modern technique

other sources of agates in Gujarat are Rajpipla,

(Figures 8i, 8j and 8k). Traditional technique of

Ranpur, Mahedpur, Tankara and the river Majam,

tattooing is done by oneself and sometimes to each

near Kapadvanj in Kaira district (Trivedi 1964: 9). At

other.

Adesar, gypsum and agate stones are found.

The tattooing symbols have variations and

represents meaning (Figure 13). The older women have

retained the knowledge of these varied symbols and

TATTOOING (TRAZAWA)

can identify its significance by mentioning it as the

highest form of adornment, which can be carried even

The formally frequent practice of tattooing is now

after death. The tattooing also represents a marker

fast dying out. Examples of tattooing were to be

for endurance. Only married women are allowed to

found among women of all age group. The name of

tattoo their legs. Among the elderly Rabari women

god and goddesses, personal names and traditional

a variety of symbols are seen which they recognise

symbols are tattooed on cheeks, chin, forehead,

when asked for, whereas among the younger girls

between eyebrows, neck, chest arms, wrist, legs and

identifying symbol is less and sometimes even nil.

calves. The tattoo marks reveal two-colour preference

green and black. According to information provided

around Kanmer: kheri jar/pilu (Salvodora persica

by M.K. Mehta in 1909 to the Superintendent of the

Linn; English known as Mustard tree/Saltbush/

Ethnographic Survey (1909: 3) states that both men

Tooth-brush tree), mithi jar/mithe pilu (Salvodora

and women were fond of tattooing their bodies, the

oleoides Decne), thor (Euphorbia tirucalli Linn./

men their arms, and the women their cheeks, chin,

Indian Tree Spurge/Milk Bush), avar (Cassia

lower lip, breast and arms.

auriculata Linn.), laso bavar (Leucoena glauca Benth.

Tattooing is achieved by a simple manual

or White Popinac/Lead tree), bavar (Acacia arabica

operation constituting of dipping the sharp end of a

Willd.), khijdo/khejra (Prosopis spicigera Linn.) and

needle or thorn in the chosen pigment and pricking

gando bavar (Prosopis juliflora DC.) which are also

the skin of the area chosen for tattooing with it. The

represented in symbolic tattoo form.

Following trees were observed by the author

materials used for preparing the dye are indigenous,

consisting of lamp-soot5) (made out of kerosene), juice

TEXTILE

collected by soaking of bark of beeoo6) and the bright

green juice of leaves of the creeper-golibad/golibar

7)

(Figure 8h). The mixture of all this is known as pilari.

The Rabari of Kachchh can be easily recognised due

- 162 -

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

Figure 9 Ornaments and attire of men folk (a, b) Traditional dress and ornaments (dokiyu) of Bharwad (c) Ornament like
Bhungri and thoriyan worn by Bharwad man (d, e) Kadla worn by Bharwad man and a young boy (f ) A villager at Ghanithal
- 163 -

Shahida Ansari

Figure 10 (a) Kiln used for making colour for tie and dye textile at Adesar (b) A Harijan weaver working at handloom at
Ghanithal (c) A Rabari women displaying the items made from sheep and camel wool (d) Bundles of sheep and camel wool (e)
Woollen rope used for making cot and pillow by Rabari at Kanmer (f ) Abandoned rectangular loom pit filled with clay next to
the in use clay lined loom pit (g) Embroidery done on dhabra (h) Fagal pal (i) Raw material for making Bandhani (j, l, m) Textile
made at Ghanithal (k) Bharwad women doing embroidery on Bandhani
- 164 -

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

Figure 11 Dress worn by women folk (a) Kumbhar women at Palanswa (b) Bharwad women at Kanmer
(c) Bharwad women at Anjar (d) Rabari women at Kanmer (e) Vedhala, a typical silver earrings
in the form of a pile of four rhombi worn by old Rabari women at Kanmer

- 165 -

Shahida Ansari

to black woollen veil clothes, ivory bangle and profuse

at times, plastic mirrors and plastic buttons are used.

silver ornament (Figures 8, 9 and 10). Bharwad and

The colour of the tie-dyed design is orange for an

Rabari women put on petticoats embroidered with

unmarried girl and red for married women. The

glass pieces (abhlas). Instead of sari they put on a red

motifs made by them are flowers, medallions, stylised

or dark coloured cloth called odhani.

horse and rider, dear, peacock and scorpion. Bharwad

Bandhani is very popular in Kachchh where it

women wear a long back less tight fitting bodice either

is customary to present it to the bride at the time

of silk or of cotton (Figures 11b and 11c). The Harijans

of marriage. Sarees, odhani (shawl), ghaghra (skirt)

are master weavers and the main producers of (Figures

and rumal (handkerchief ) or bandhani, popularly

10j, 10l and 10m).

known as tie and dye, is the process of resist printing,

is one of the most important crafts of the Kachchh,

1) Black and red woollen head cloth (charmalia)

especially noticed at Adesar, about 56 km from Rapar.

for women (Figure 11b),

The raw materials required are cloth, colours and

2) Shawl woven in narrow width loom in two

thread procured from the local market. The process of

pieces and put together by the women; joint is

bandhani production is simple yet laborious and time

embroidered with multicoloured threads (Figures

consuming. The cloth which may be silk, georgette,

10g and 10k),

voile or mulmul is spread on a table or wooden plank

3) And white blanket (dhabra) for men has series

and designs are marked on it with the help of wooden

of borders woven at both ends, an architectural

blocks or by hand (Figure 10i). It is then tied in tiny

pattern woven into the body. The typical Bharvad

knots with thread, where after the cloth is dyed in

dress for men is three blankets of undyed wool one

appropriate colours. On completion of this process,

wound in broad bans round the head, a second

when the ties are removed, the finished product

tied round the waist reaching the knee, and a

presents a repeat design in dots covering a variety

third thrown across the shoulder. These are made

of motifs. The colours, which are generally used in

in large variety of decorative pieces for use in the

Kachchh, are red green, blue and black. The only tools

commercial urban markets (Palanswa, Bhimasar

and equipment needed are utensils for dyeing (Figure

and Rapar).

10a) and wooden blocks for printing designs.

The Harijans of Ghanithal village (20 km north

These are made in large variety of decorative

of Kanmer) are the main weavers in this region

pieces for use in the commercial urban market

(Figures 10b). The Rabaris and Bharvads wear clothes

(Palanswa, Bhimasar and Rapar).

those made by these Harijans. The Harijans, Rabaris

and Bharvads produce skilled embroidery, which is

Harijans and in return they used to make blankets/

bold and vigorous in style in this region. The Bharvad

shawls. In earlier days the Rabaris wore only black

decorate their ghaghras and woollen head cloths with

coloured dress. As they were barred to wear any

wide borders. Thick tie-dyed fabric is made heavier

colourful dress, the reason being they are of lower

with large geometric, animal and floral motifs in

caste. Some 25 years back they used to barter wool

brilliant yellow, blue, purple, green, red and orange

in exchange for grain, pottery, and iron objects with

colours (Figures 10g and 10k). Plastic and glass are

other communities. Those residing in the Kanmer

now used for embellishment. The Rabari bride must

village continue to make wool out of sheep/goat and

own three heavily worked blouse and a woollen tie-

camel hair. It is while grazing their animals in the

dyed shawl embroidered with borders and medallions,

field or far off places for pasture, that they carry a bag
- 166 -

Traditionally, the Rabaris used to supply wool to

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

having hair of the goat and camels. While walking

earlier days he moved to Saurashtra, Ahmedabad,

they make (thin or thick) rope by adding bunch of

Panchmahal and Varodara with his sheep and

hair and simultaneously twisting it, this is then made

camel. To reach Saurashtra they crossed Little

into a woollen ball (Figure 10d). It is of two colours

Rann of Kachchh, which approximately takes one

black and cream/off white colour. The items made

and half day. Movement generally begins in night;

are: woollen rope (used while making wooden cots),

they travel for first 12 hrs and then take 7-hour

pillows (though rough in texture but are still preferred

rest before moving again. While travelling they

for its sturdiness and warmth), bags, straps of different

carry cooked food or items, which can be eaten

shape and size to be put on camel (use of thick large

on way. Across the little Rann is village Teekar,

iron bangles/rings are used, and galon used to through

where they take rest before moving to different

stone on wild animals and birds (Figures 10c and 10e)

areas in Saurashtra like Mahoa, Paragarh, Thowai,

Daboi, and Junagadh. The whole family moves in
this temporary movement except the old parents

MOVEMENT PATTERN OF RABARIS

and pregnant ladies, who stay back at the village.

He also informed that many families also moved

A shepherd, Vajju Bhai (55 yr. old), a Rabari was

towards central India (Malwa region) for grazing

interviewed and he provided following information.

sheep and camel.

6) They also used to carry salt for two specific

1) He was with his grandson, aged 10 yr. Only male

reasons. Firstly, to be given to camels so that they

child at the age of two, start moving along with

drink enough water before start of the journey and

elders, while grazing khinta/ghinta or sheep (200

secondly, to be sold to Maldhari (Herdsmen or

in number) in nearby agriculture field.

grazers) and Khatik (Butcher men) for cash.

2) Every day they graze their sheep from 7 AM

7) When asked about Bhils, he sighted their

to 8 PM. Sometime he also stays out for one to

traditional enmity with them, and doesnt like

three days with sheep, generally in open, when the

any association with them. Bhil according to him

pasture is not enough.

use bow and arrow for hunting and stealing their

3) Has good knowledge of their surrounding

sheep, whereas they use galon for protection.

ecosystem mainly due to their movement per day is

four to six kilometres. The area of movement varies

SALT MAKING TECHNIQUE

during rainy season.

4) While moving he carries a bag having many
items like : food (rotala- jowar roti, cooked

Being on the seacoast, Kachchh abounds in salt,

vegetable, chilli paste), aluminium utensils (plate,

which is recovered at present from sea brine either

bowls), wool, two blanket, kerosene lamp wick

in artificial pans or in open natural pans. There are

bottle, matchbox, torch, utensil for making tea, tea

thirteen salt works in Kachchh, of which twelve are

leaves, sugar, rope, medicines for sheep, syringes,

major, six located at Kandla, two at Chirai, one each

colour, galon, stick and two water bottles. He had

at Jakhau, Mundra, Padana and Shikarpur. At Adesar

collected discarded iron wire (Figure 12f ). They

the salt is manufactured from the wells near the

whistle to call their animals.

border of the Rann (Patel 1971: 234, 581).

5) Since last ten years he has stopped travelling too

far of places due to new government policies. In

manufacturing site just a kilometre away from the

- 167 -

The author got an opportunity to visit the salt

Shahida Ansari

Figure 12 (a) Storage pits of Period IV (Historic) at Kanmer (b) Stone Fortification wall, Kanmer (c) Hero stones (pariya)
at Kanmer (d) Uri mata (e) Koli women at salt manufacturing site, Shikarpur (f ) Items carried by Rabari man (g) Temporary
settlement of Koli at salt manufacturing site, Shikarpur (h,i) Salt manufacturing at Shikarpur

- 168 -

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

Figure 13 Tattoo symbols made by Rabari and Bharwad women in Kanmer, Kachchh district, Gujarat

- 169 -

Shahida Ansari

Harappan site of Shikarpur (Figures 12h, 12i). The salt

continue to make salt.

workers belong to the Koli communities, who come

with their families from surrounding villages. The

of Kachchh. On the first day of Chaitra (April)

reason for hardship is that with no rain, sustenance

and on the Shravan (August) full moon all traders

through agricultural activities is hindered, which

worship the sea by throwing flowers and a coconut

leads them to undertake this occupation. Salt making

into it. Salt, called mithu, i.e. sweetener, and sabras,

has always played important role in this region. For

i.e. the essence of taste, is the luckiest of all purchase

the convenience the manufacturer/owner provides

on the New Year day in Kartik (November). To be

the Kolis with basic amenities like house, electricity

freed from the throes of death a dying person makes

and water facility. But a few Koli also make temporary

a gift of salt to Brahmans. For this reason most high

shelter, where they keep buffaloes and sell milk and its

caste Hindus are careful not to pass salt from hand to

product to the villagers at Shikarpur. In the temporary

hand in everyday use. And salt is also used in all spirit-

settlement a small hut erected with tree branches,

scaring rites, and on the Great Spirit day-Kalichaudas

which is used for keeping their belongings (bedding

(dark fourteenth in October-November), high caste

and clothes), however utensils, cooking pots are

Hindu women make salt marks of various shapes at

placed outside in open, where they carryout cooking

crossroads (Kirparam 1901: 349).

Salt plays important role in the life of the people

and make buttermilk (Figure 12g). A cattle pen is also

seen in thorny courtyard with stacking of cow dung

RELIGIOUS BELIEF

cakes and fuel sticks/branches. Thorny branches of

bavar enclose the hut and the open area. The Koli
women wear beautiful silver ornaments (Figure 12e).

The villagers of Kanmer worship mother goddess

However, the children and women were seen with

(mata) R ajbai, whose temple is located at the

skin disease as well as deform eye ailment. A small

hillock at Kanmer. As reported by B. Kirparam

shrine of devi ma was seen near this settlement. The

(1901: 288) and also by M.K. Mehta in 1909 to

Koli dont do fishing, it is the Muslim communities of

the Superintendent of the Ethnographic Survey

this area (Surajbari village) who are engaged into sea

(1909: 5), the Rabaris are generally worshippers

fishing. They employ boats (navdi) made by the local

of mothers or matas, though a few among them

carpenters (suttar) for fishing.

belong to the Bijmargi, Ramanandi, and Shaiva sect.

Most of the artificial brines8) are obtained by

In Kachchh many of them serve as priest in mata

pumping seawater to flow through a series of channels

temples, commonest forms under which the mata is

constructed of sand into a series of rectangular

worshipped are a peacocks feather and a conch-shell.

shallow ponds/pan separated by sand mounds (Figure

At Kanmer the Rabaris continue the same tradition

9)

12h and 12i). The salt in each crystallising pan is

till date. They also continue to worship cobra on the

raked into rows, where it is allowed to drain for

Kalichaudas (dark fourteenth in October-November)

several days (at least 20 days). After that it is collected

and offer the cobra milk and cocoa kernel.

into heaps, drained again, lifted from the pans, and

finally dried. Wooden and steel furrows are used

of her is often figured on silver and copper plates, a

during this process. Both men and women work in

few also worship Ramamnandis. The most sacred is

night, as it is comparatively cooler. The salt making

the Navratri or nine nights in September-October,

is done throughout the year except in two months

Kalichaudas and Akatrij in April-May is their sheep-

( July and August) of rain. But if it is a dry spell they

shearing day (Kirparam 1901: 268).

- 170 -

The Bharvads are also followers of matas, image

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

The aboriginal tribes of Gujarat as reported by

The fortification at Kanmer and the residential

B. Kirparam (1901: 292, 305), worship two classes of

structures are mainly fashioned in stone, reflecting the

beings, local or tribal deities and the spirits of their

skills of its builder and his technology. The technical

ancestors. Of the local or tribal divinities, Khatri

knowledge is also reflected in the range of materials

Dev, a pillar of wood, is worshipped by most of the

that were available and used for construction during

aboriginal tribes except by the Bhils and Gamtas;

the ancient times. It is the skill on stone that permits

Khetrapal, the god of boundaries, a stone carved with

the erection of larger and more elaborate structures

the figure of a horse; Baba Dev, a clay image of a horse

suggesting building is more than an adaptation

and rider, worshipped chiefly by Bhils; Magaria and

to climate. This being an earthquake zone, stone

Vagh Dev, or the alligator and tiger deities chiefly

structure is not advisable for construction. Despite

worshiped by Gamtas. The animal held in most

this knowledge the villagers continue to construct

veneration by the Bhils is the horse. If a prayer has

their house, courtyard, and boundary wall in stone

been granted, they often make small clay horses and

blocks quarried from nearby hill. The Harappan

arrange them around an idol, or in the spirit-yard.

workman who were adept to the local environment

At Ghanithal image of urimata were seen in

and possible tectonic disturbances at the region,

those houses where the children were suffering from

continued to build such impressive structure with

chicken pox (Figure 12d). The mother goddess is

care. A section at Kanmer during Harappan time was

worshiped for curing properties; once the child is

probably involved in masonry, as the fortification

cured the image of mata is immersed in water. At

wall appears to have been plastered with white shale

Kanmer and Thoriali the local and those of distant

(dhauli matti) brought from nearby area, suggesting

relatives worship Hero stones, i.e. pariya and sati

the upkeep of the structure at regular interval.

stones. The hero stones have beautiful representation

of attire, ornaments, weapons, animas, cart, and sun/

complex societies, subsistence factors, in narrow

moon symbol (Figure 12c). The date and details of the

sense, are less important as determinants of the size

issuers/hero is inscribed in devnagari script on the

and location of communities. Trade may provide a

stone pillars.

source of wealth and stimulate the growth of large

According to B.G. Trigger (1968: 63) in more

cities in remote regions, and the wealth amassed

from trade may in turn serve either to finance the

COMMENTS

developments of novel agricultural systems in areas

where they would otherwise be impractical, or to

The ethnographic obser vations in and around

effect the importation of food from distant regions.

Kanmer brought to light many aspects of settlement

Settlements may also spring up in wastelands where

and subsistence pattern. The villages at a radius of 15-

rare and valuable minerals are discovered but lacking

20 km are known for their traditional craft and people

to agricultural potentials. Kanmer is located very near

who supply raw material in the region, i.e. Gagodhar

to Mardak Beyt, a source for agate, raw material used

and Palanswa for potters, Ghanithal for textile, Adesar

for making beads, a popular and in demand during

for textile and jewellery, Kanmer for agro-pastoral

the Harappan times. A number of artefacts (tools)

communities like Rabaris and Bharwads (highest

and agate beads are found at the site, which suggests

in the region). The villagers are interdependent and

that it was a local craft at the site. To acquire this

interwoven for the craft in the modern times as a

raw material a section of the community within or

source of subsistence and survival.

outside were involved in extracting the raw material

- 171 -

Shahida Ansari

and supplying at the site. The site of Kanmer seems

dependence on other food sources available in the

to be one of the centres, which supplied raw agate or

region, suggested in the faunal assemblage and

finished artefacts to other Harappan towns. According

other archaeological evidences like structures on

Trigger (1968: 64) a specialized local centre can

fortification wall.

develop when a number of villages are linked together

in a trade network. They are better able to transcend

of Rabari/Bharwad in Little Rann of Kachchh

local limitations in natural resources; and a trading

will be carried out. The ethnoarchaeological study

network also fosters greater specialization and better

will gather information concerning production

products than does an autonomous village economy.

relationship through study of the layout of workshops

Thus, a careful investigation of trade and of the nature

(for example-bead manufacturing agate/steatite or

of production is prerequisite to estimates of the size

pottery making), will note whether places of work

and the social cultural complexity of communities.

and residence are together or separated, whether

Thus, it is important for us to determine what goods

the workshops of one trade are in the same part

were present and produced at the site of Kanmer.

of the community or at scattered points. Data of

this sort will permit us for reconstruction of social

Within any region, people tend to establish

their settlements in places that are close to drinking

Further, research on the movement pattern

organization at the site of Kanmer.

water, sources of food and places, which is safe.

Agricultural groups generally seek locations where

Acknowledgements

the soil and weather are favourable to their crops

I am grateful to the directors of Kanmer excavation

and their method of cultivation. Presently, at

Dr. Jeewan Sing h Kharakwal, Dr. Y.S. Rawat

Kanmer agricultural activities are possible only due

and Prof. T. Osada for allowing me to carry out

to underground sources like wells, seasonal nallah

ethnoarchaeological studies at the site of Kanmer. I

and recent introduction of Narmada river water

thank Katariyaji and Sameer for accompanying me

supply. The villagers mainly rely on kharif crops. The

during ethnographic survey in Kanmer and nearby

grassland provides pastoral communities like Rabaris

villages; to Gababhai, Udubaa and Shiva, enthusiastic

and Bharwads to have grazing land for domesticated

villagers at Kanmer who readily parted with their

animals like sheep, goat, buffaloes, cow and camel.

immense knowledge about culture and tradition of

From the Archaeological evidences from Kanmer

Kanmer; and to Shri D.D. Phule of Deccan College,

it seems to be a craft specialized centre, but some

Pune for helping me with the line drawings.

section of the society was also carrying out partial

agricultural, hunting and gathering activities for
sustenance.

Notes

The dependence on the wild animal (large

1) The name Koli meaning clansman, clubman, or boatman,

bovid, wild pig, antelopes, deer, carnivores and small

applied to the middle classes of the military or predatory

Hindus of Gujarat, includes tribes that differ widely from each

mammals like hare and rodents) in the Late Harappan

others (B. Kirparam in Gazetteer of the Bombay Presidency:

phase was more than any cultural phase (Goyal and

237).

Joglekar 2008: 39). This also suggests that possibly

2) At Kanmer Dudio Kenga Jaga, the oldest Rujput man

the Late Harappans at Kanmer had some pressure

(aged- 110 years) and Vinod Bhai (aged-55 years) a Jain and

from other regional craft specialised villages who were

at Palanswa village Lakshman Bhai (aged-65 year) a Rujput,

also engaged in supplying of raw/finished material.

provided with the information regarding the traditional

The gradual pressure on the economy resulted in

underground storage technique of khann.

- 172 -

Ethnoarchaeological study at the Harappan site of Kanmer, Kachchh, Gujarat

3) When Rebarris come upon a herd of camel grazing, the

(eds.) Occasional Paper 5: Linguistics, Archaeology and

boldest and most experienced strikes his lance into the first

the Human Past. Indus Project Research Institute for

animal he reaches, then dips a cloth in the blood, which at the

Humanity and Nature, Kyoto. pp.25-43.

end of his lance he thrusts close to the nose of the next, and

Kharakwal, J.S., Y.S.Rawat and Toshiki Osada (2008)

wheeling about, sets off at speed, followed by the whole herd,

Preliminary Observations on the Excavation at Kanmer,

lured by the scent of the blood, and is followed by the rest.

Kachchh, India 2006-2007, in T. Osada and A. Uesugi

4) Identified as herdsmen by B. Kirparam in Gazetteer of the

(eds.) Occasional Paper 5: Linguistics, Archaeology and

Bombay Presidency: 285-86, chiefly found in Kathiawar. They

the Human Past. Indus Project Research Institute for

call themselves Rajputs, claiming descent from one Ranhirji,

Humanity and Nature, Kyoto. pp.5-23.

a Jethva. The census of 1891 includes them under Ahirs or

Jain, J and M. Sarabhai (1985) Ornaments, in J. Dhamija

Jethvas.

(ed.) Living Traditions of India-Crafts of Gujarat. Mapin

5) Duhando or the lamp soot is collected from the i.e. kerosene

International, New York. pp.41-64.

Lieutanant- Colonel James Todd (1873) Annals and

oil kept overnight, partially covered by inverted vessel.

6) After four days the beeoo wood is soaked in water turns

Antiquities of Rajasthan or the Central and Western Rajpoot

Neelu (green colour).

States of India. Higginbotham and Company, Madras.

7) The juice of golibari is sieved in thin cloth. Golibari grows

Mankad, B.L. (1939) Rabaris of Kathiawar (A Social Study).

in rainy season on pilu or Zare tree.

Journal of the University of Bombay 7(4): 31-79.

National Commission on Agriculture (NCA) (1976) Rainfall

8) Brine is water containing a high concentration of salt.

9) Salt is produced by solar evaporation i.e. when an aqueous

and Cropping Patterns, vol.IV. Government of India,

solution of several salts (seawater, for example) is evaporated;

Ministry of Agriculture and Irrigation, New Delhi.

each of the salts precipitates as it reaches its point of saturation

Patel, G.D. (1971) Gujarat State Gazetteers: Kutch District.

in the solution. Thus, the different salts in seawater will

Government Printing , Stationary and Publications,

precipitate at different times, forming layers on the bottom of

Ahmedabad.
Singh, K.S. (1997) The Scheduled Tribes. People of India,

the evaporating pond.

National Series, volume III. Oxford University Press

References

(Oxford India Paperbacks), New Delhi.

Dutt, S.C. (1884) The Wild Tribes of India. Gilbert and

Superintendent of the Ethnographic Survey of Bombay

(1909) The Ethnographic Survey of Bombay Monograph no.

Rivington Limited, London.

Enthoven, R.E. (1920) The Tribes and Castes of Bombay, vol. I:

137- Rabari. The Government Central Press, Bombay.

118-123. Government Central Press, Bombay.

Trigger, B.G. (1968) The Determinants of Settlement

Enthoven, R.E. (1922) The Tribes and Castes of Bombay, vol.

Patterns, in K.C. Chang (ed.) Settlement Archaeology.

III: 252-258. Government Central Press, Bombay.

National Press Books, California. pp.53-78

Kirparam, B. (compiled) (1901) Gazetteer of the Bombay

Trivedi, R.K. (1964) Agate Industry of Cambay, in Census

Presidency, Gujarat Population: Hindus, volume IX, part 1.

of India 1961, volume V, part VII-A (1), Selected Crafts of

The Government Central Press, Bombay.

Gujarat. Manager of Publication, Delhi.

Goyal, P. and P.P. Joglekar (2008) Report on the faunal

Trivedi, R.K. (1967) Census of India 1961- Selected Crafts of

remains recovered from Kanmer, Gujarat, during the

Gujarat. volume V, part VII-A. Manager of Publication,

second field season (2006-07), in T. Osada and A. Uesugi

Delhi.

- 173 -

Shahida Ansari

- 174 -

Memorial stones at Kanmer, Gujarat, India

Memorial stones at Kanmer, Gujarat, India

Tilok Thakuria
Department of Archaeology, Deccan College, Pune-6

J.S. Kharakwal
Rajasthan Vidyapeeth, Udaipur, Rajasthan

Toshiki Osada
Research Institute for Humanity and Nature

Y.S. Rawat
State Department of Archaeology, Gujarat

Abstract
Memorial stones, like other parts of India, are commonly found in parts of the state of Gujarat. The present paper is an attempt to
document and understand various aspects like location, iconography, ritualistic and socio-economic factors of the memorial stones
at Kanmer. Kanmer is a small village in Kachchh district of Gujarat.

Kanmer is a small village at Rapar taluka of Kachchh

district, Gujarat. The village is known in archaeologi-

cal data for a society or community to know about the

cal literature for a Harappan settlement reported by

socio-economic, socio-religious and socio-political

Bisht (IAR 1985-86: 15-19, Kharakwal et al. 2007).

events. The inscription engraved on the memorial

The present village of Kanmer is at the foot of a hill-

stone is also used as historical data. The present paper

ock. Pliyas are found in the surrounding areas of

deals with the documentation of Pliyas found in the

the village in a well defined distance from the village

village of Kanmer. This documentation is fashioned

boundary. There are seven clusters of Pliyas and four

to create a dataset on the location, construction, ico-

isolated Pliyas are seen in different locations.

nography and folk stories related with the Pliyas of

The practice of erecting memorials to commemo-

Kanmer. All these aspects have been put together in

rate heroic death is widespread in many parts of India,

this paper to avail them for further comparative re-

and Gujarat is no exception .Memorial stones known

search on Pliyas of Gujarat in general and Saurashtra

by various local names are commonly found in dif-

and Kachchh in particular. They are located either on

ferent parts of Gujarat and it is still a practice seen in

asmall hillock or near water bodies which are a little

many villages of Gujarat (Doshi 1982, Jain 1982, Shah

far from the village (Figure 1). Cluster 1 and Cluster

1982). In Saurashtra and Kachchh memorial stones

4 are situated on a small hillock while Cluster 2 is

are locally known as Pliya. They are immense in

located near a nala. On the other hand, Clusters 5, 6

number in villages of Saurashtra and Kachchh. They

and 7 all are outside the village boundary. Conversely,

are found in great numbers in the villages of Saurash-

Cluster 3 falls inside the village. But it seems that it

tra and Kachch. Fascinated by the Pliyas found in

was originally at the margin of the village. After the

the villages of Saurashtra, James Todd remarked we

recent earthquake the village was reestablished with

find them planted thick as grave stone in an English

new construction. The boundary of the village might

church (Todd 1971: 301).

have been altered by the new constructions and as

- 175 -

Memorial stones are considered as Anthropologi-

T. Thakuria et al.

Table 1 Distribution of Pliyas found in Kanmer

Cluster no.
1

Location

Total number of Pliyas

In the cremation ground belonging to Darbari community,

30

towards the Northwest corner of Harappan settlement

In the Harijans graveyard

13

Near the Kanmer village panchayat office

20

Towards the east of Harappan mound at Kanmer

32

Near the camp of the Kanmer Project, 2007

6
7
8
9

GPS coordinates

Housed in the Jivnisati temple located on the entrance of Village

Kanmer
On the entrance of the village, around 50 m east from Jivni Sati
temple
Located around 1km towards east of Kanmer village. Housed in
temple known as Surdhar Dada
Near the Harappan mound

4
3
1
1

2325'12"
7051'34"
2324'54.9"
7051'33.3"
2324'46"
7051'42"
2324'48.2"
7051'58"
2324'53.1"
7051'42.2"
2324'55.3"
7051'47.3"
2324'52.7"
7051'49.4"
2324'54.2"
7052'31.7"
2315'02"
7051'52.7"

cluster 3 was in the margin of the village boundary

the cluster 1. Similarly, 4 are housed under a Chatri

came under the newly expanded village area. Cluster 7

in cluster 3 and others have a recently made platform.

and 6 are on the village entrance. There are 4 isolated

Pliyas are associated with the ancestor and it is cus-

Pliyas located at the margin and outside the village

tomary to offer worship. Hence, the platforms, stone

boundary. The one known as SurdharDada which is

enclosures or Chatris are made by the family members

housed inside a temple, is around one and half km

to facilitate worship or to show honor to their ances-

from the village towards east located on a small hill-

tor. Such provisions are made to only those Pliyas

ock. Pliyas at Kanmer represent both Hero Pliya

whose family members can afford it or identify their

and Sati Pliya.

ancestors Pliyas after a long period.

All the Pliyas located in the village are of sand-

A Pliya has two basic parts- the carving and the

stone which is available in Kachchh. A Pliya is

inscription below. The arch or angular top portion

actually an upright flat stone. The top portion of it

depicts the sun and moon symbols accompanied with

is either arch shaped or angular. It is raised in a pit

floral motif. The general iconography of the hero

dug into the earth surface. To give support at the base

Pliya shows the hero either riding a horse or camel

stone boulders and soil are put inside the pit. Some-

or standing in warrior dress. Sati Pliya is depicted as

times a stone platform or enclosure is seen surround-

an upright female arm or with a dead body on her lap.

ing a single Pliya or a row of Pliyas at Kanmer. In

However, there are many detailed differences in the

cluster 1 two Pliyas have rectangular stone enclosure.

iconographic appearance in both the Hero and Sati

In cluster 4 a stone platform is made in front of 3

Pliyas at Kanmer.

Pliyas (Figure 2). On the other hand, Chatris are

constructed to house Pliyas. 4 Pliyas are housed

a) Pliya with a horse rider: This iconographic feature

inside two adjoined Chatris with a high raised plat-

is most commonly seen among Pliyas at Kanmer.

form and one single Pliyas with a separate Chatri in

In this the hero is shown in warrior attire riding a

- 176 -

Memorial stones at Kanmer, Gujarat, India

horse. He holds a spear either in his right hand or in

bow and shield in the other hand (Figure 9). The

left hand. The spear is shown to be held either hori-

arrow container is shown hanging from the waist.

zontally over the head or behind the body. The other

Examples of two warrior figures holding sword and

hand holds bow and arrows. In most cases, a shield is

shield are also available.

shown to be held in the hand in a protective manner

f ) Pliya with a bullock cart: A man riding a bullock

(Figure 3). Sometimes, instead of spear and bow the

cart and holding a sword in his right hand is shown.

figure is shown with sword and shield. There are also

Beside him, a milk-churn pitcher is also shown (Figure

examples of two horse rider figures in the same Pliya

10). It is said that such Pliyas are erected in memory

stone (Figure 4).

of the brides who die in plunder during the bridal

b) Pliya with a horse rider holding a gun: There are

procession.

two examples of such type at Kanmer. In one case the

g) Sati Pliya with a dead husband on lap: The sati

figure is shown holding a spear in his left hand over

figure is shown standing with her right arm raised

the head and in the right hand a gun. He is shown rid-

with open fingers and the dead body placed vertically

ing a horse. In the other example the figure is shown

on her lap (Figure 11).

holding the gun in shooting posture (Figure 5). He is

h) Sati Pliya with a raised arm: The complete figure

also shown riding a horse.

of the sati is not shown, but a raised arm with open

c) Pliya with a horse rider and a standing figure: The

fingers is shown (Figure 12).

main figure is shown riding a horse holding spear in

i) Sati Pliya and horse rider: The male figure is

left hand and bow in the right. Sword and dagger are

shown in warrior attire riding a horse and the raised

shown hanging from the waist. Behind the horse-

arm of sati with open fingers is shown either in front

riding figure, a human figure has shown standing. It is

or behind the horse rider. (Figure 13).

difficult to make out the characteristics of the stand-

j) Standing Sati Pliya: The sati figure is shown stand-

ing figure as it has worn out because of weathering. In

ing with her right arm raised with open fingers and

another example, the standing figure is shown in front

holding a rosary in her left hand (Figure 14).

of the main figure. The main figure is shown riding

a horse and holding a spear in his right hand and in

the left hand probably a rosary. The figure in front is

istics in the Hero Pliyas. They all are shown having

shown standing on a pedestal or similar arrangement

water container hanging from the waist. They also

(Figure 6).

wear turban or headgear and shoes.

d) Pliya with camel rider: Instead of a horse, the

male figure is shown riding a camel, dressed in warrior

it just below the carving. In some cases the inscription

attire, and holding bow and arrow with sword and

is not traceable because of erosion or damage of the

dagger hanging from the waist. The left hand holds

stone. It is also because of construction of platform,

the reins of the camel (Figure 7). In another case, two

stone enclosure etc. The inscription either gets buried

male figures are shown on the camel. Both of them

or mutilates because of such construction. Inscriptions

hold a spear over the head and bow and arrow in the

are written in Gujarati script and Gujarati language.

right hand. One of the Pliya housed in the Jivni Sati

They contain the name of the person, astrological de-

temple shows a child in between tow figures (Figure

tails of the dead and period of erection of the memo-

8).

rial stone. Information about the cause of the death is

e) Pliya with a standing posture: A warrior is shown

not available in the inscriptions in Pliyas at Kanmer.

standing holding a sword in attacking posture and

- 177 -

There are some general iconographic character-

Each Pliya is having an inscription engraved on

The stories about the Pliyas are more or less

T. Thakuria et al.

Figure 1 Cluster of Pliyas on a small hillock (Cluster 1)

Figure 2 Pliyas with a stone platform in front

- 178 -

Memorial stones at Kanmer, Gujarat, India

Figure 3 Pliya with a horse rider

Figure 4 Pliya with two horse riders

- 179 -

T. Thakuria et al.

Figure 5 Pliya with a horse rider holding a gun

Figure 6 Pliya with a horse rider and a standing figure

- 180 -

Memorial stones at Kanmer, Gujarat, India

Figure 7 Pliya with a camel rider

Figure 8 Pliya with two camel riders and a child in between

- 181 -

T. Thakuria et al.

Figure 9 Pliya with a standing figure

Figure 10 Pliya with a bullock cart rider

- 182 -

Memorial stones at Kanmer, Gujarat, India

Figure 11 Sati Pliya with a dead husband on lap

Figure 12 Sati Pliya with a raised arm

- 183 -

T. Thakuria et al.

Figure 13 Sati Pliya and a horse rider

Figure 14 Standing Sati Pliya

- 184 -

Memorial stones at Kanmer, Gujarat, India

Figure 15 The ruin of the Darbars house in the village

Figure 16 Blouse as offering on the tree located in the Jivni Sati temple area
- 185 -

T. Thakuria et al.

same as available in other villages in Saurashtra and

the Kanmer village and nearby village worship Jivni

Kachchh. These stories narrate the courage and brave-

as symbol of motherhood. Women folk worship her

ness. The most famous folk story among the people of

to get enough breastfeeding to their child and offer

Kanmer is that of Jivni Sati. According to the story,

blouse to get blessings. They hang the blouse as an

Jivni was the youngest daughter of a Bharvad. The

offering on the tree located in the temple area (Figure

Bharvad had a herd of goats. Among the herd Jivni

16). Local people of the village say that nobody even

used to like one kid. One day, when the herd returned

dares to pass her temple on horse back. It is because

after grazing she did not find the kid she liked. She in-

the curse she gave to the Darbar that the father will

quired about the kid going missing Her father replied

die when the son become able to ride horse.

that it must be nearby somewhere and will appear

soon. Jivni was not convinced by the reply and said

courage and bravery of the individuals who died in

it can not be some where nearby because everyday it

war against invaders, fight with robber and other

comes to her first at the time the herd arrived home.

courageous activities. About the Pliyas in Cluster 4

She realized that her father was lying to her. So she

story says that a bridal procession was plundered and

enquired out the missing of the kid. Finally, she came

killed after the loot. So, in memory of them Pliyas

to know that a Darbar managed to have the kid for

were erected. There is a Pliyas in Cluster 1 showing

a feast with the help of few Koli. It gave her sorrow

a horse rider with gun in hand. The story about this

and anguish. She rushed to the Darbars house to

Pliya is that there was a raid by robbers in the village

get justice. The Darbar did not show any interest in

armed with guns. Villagers were afraid of them as they

Jivnis grief. He did not even open the main door of

possessed gun. So, nobody dared to fight with them.

his house. She became angry because of the Darbars

When the robbers entered and tried to loot his house,

ignorance on the matter and got inside the Darbars

he came out with his gun and fought a gun fire battle

house by breaking the door. The story says that she

with them. Unfortunately, he died in the battle. These

broke the door by a kick. Getting inside the house

stories are a product of oral tradition and it is difficult

she killed herself with a dagger hanging on the wall

to get background story for each of the Pliyas. There

and cursed the family. She cursed that Darbar will

must be some truth in the stories in general, but the

die when his son will become able to ride a horse and

possibility of exaggeration can not be denied.

they can never have more than one kitchen among

brothers. After the curse, as the story says, the social

auspicious occasions and festivals like Diwali. Newly

and economic condition of the Darbar declined. The

married couple and new born pay visit to their ances-

ruin of the Darbars house is still present in the vil-

tor Pliya for blessing and protection from the evil.

lage (Figure 15). Villagers presently worship the place

People believe that Pliyas protect the village from all

where Jivni sacrificed her life. To commemorate Jivnis

kind of evils and enemies.

sacrifice a Pliya was erected outside the village which

is now housed in a temple. It is said that after the

at village Kanmer considering the custom of Pliya

tragic event of Jivnis sacrifice, all the Bharvad people

erection particular to Kanmer or may be Gujarat in

left the village to protest and respect Jivnis sacrifice.

general can be put as (i) They are located outside

They come yearly or occasionally to offer worship to

the village boundary. (ii)They are placed either on a

her Pliya. They bring each and every thing necessary

small hillock or near water sources. (iii) Pliyas are

for worship. They dont even take fire and water from

not necessarily erected at the very spot where the

the people at Kanmer village. At present people of

death occurred. Jivni Sati story suggests the same.

- 186 -

The other stories about the Pliyas narrate the

Villagers offer ancestor-worship to the Pliyas on

Altogether, the general characteristics of Pliyas

Memorial stones at Kanmer, Gujarat, India

Pliya for Jivni was erected outside the village bound-

Acknowledgment

ary though she sacrificed her life inside the village in

I am indebted to members of Kanmer Archaeological

Darbars house. (iv) Sati Pliya is not only for those

Research Project and villager of Kanmer team for help

who immolate themselves on ones husbands fu-

and support.

neral pyre. It can be for both - unmarried or married

women who sacrificed their lives for a noble cause or

References

against unethical deeds. Pliya for Jivni was erected

Doshi, Saryu (1982) Pliyas of Saurashtra, in S. Settar and

to commemorate her sacrifice against injustice. (v)

Gunther D. Sontheimer (eds.) Memorial Stones. Institute

All hero Pliyas face eastward and the figure of the

of Indian Art, Dharwad. pp.157-173.

hero faces north (vi) The most common theme of the

IAR: Indian Archaeology - A Review. Archaeological Survey of

folk stories regarding Pliyas is courage, bravery and

India, New Delhi.

fight against plunder. (vii) A Pliyas can be erected

Jain, Jyotindra (1982) Ethnic Background of some Hero-

after a year or more of ones death. Inscription in One

stones of Gujarat, in S. Settar and Gunther D.

Sontheimer (eds.) Memorial Stones. Institute of Indian

of the Pliyas from Cluster 1 says that the hero died

Art, Dharwad. pp.79-82.

somewhere in Kachchh and the Pliyas was erected

Kharakwal, J.S., Y.S. Rawat and Toshiki Osada (2007)

by his family after one year of his death. (Viii) It is

Kanmer: A Harappan site in Kachchh, Gujarat,

difficult to make out socio-economic position of

India, in T. Osada (ed.) Occasional Paper 2: Linguistics,

the dead from the iconographic feature. It is because

Archaeology and the Human Past. Research Institute for

iconographic feature is not much different from each

Humanity and Nature, Kyoto. pp.21-46.

other. (ix) It was customary to show Pliya either on

Kharakwal, J.S., Y.S. Rawat and Toshiki Osada (2009)

horse or Camel. Local people say horse or camel sug-

Excavations at Kanmer: A Harappan Site in Kachchh,

gest that the person was able to keep the same animal

Gujarat. Purtattva 39: 147-164.

Shah, Haku (1982) Tribal Memorials in Gujarat, in S. Settar

as vehicle. Another view suggests that Pliya with

and Gunther D. Sontheimer (eds.) Memorial Stones.

Camel riding belongs to Bharvad community as only

Institute of Indian Art, Dharwad. pp.101-116.

they keep camels. Otherwise, it is difficult to make

Todd, James (1839) Travels in Western India. Reprinted by

out which communities they belong from only the

Oriental Publishers, Delhi in 1971.

iconographic features. (x) Offer worship to their family Pliya on the day after Diwali. A new born infant
or a newlywed bride pays visit to the family Pliya to
get blessings.

- 187 -

T. Thakuria et al.

- 188 -