r

CHAPTER 2

Kecaikhari, Eater of Raw Flesh:

A Profile of the Multifaceted
Goddess in the North-East

JAE-EUN SHIN

One of the most famous goddesses worshipped in the north-east

and one of the first to receive attention from British colonial officer­
scholars in the 1800s, is the Goddess Kecaikhatl, eater of raw flesh,
and her history should have been well-chronicled. This is, however,
not the case. She has been mentioned only in relation to human
sacrifice or represented as a rypical bloodthirsry tribal goddess in
many works on the north-east. 1 This article will look at the hitherto
little known history of Kecaikhari in the period between the eleventh
and nineteenth centuries and will then attempt to contribute to­
wards a better understanding of her multiple identities created by
different social groups with their own intentions.

THE GODDESS OF THE EASTERN LIMIT

OF KAMARUPA

The shrine of Kecaikhatl has always been referred to as the Tamre­

svarl Temple in later times because it was covered with a roof of beaten
sheets of copper. The temple formerly stood at lat. 27 ° 56', long;
96° 21' near Paya in Arunachal Pradesh, about 8 km away from
Tebangkhunti on the Sadiya-Teju route. 2 On the basis of the in­
scription of Saka 1364 (AD 1442) found in the temple, which was
first noticed in The Annual Report on the Indian Epigraphy, 1957-8,
p. 56 (No. B 386), the goddess was identified as Digaravasinl. The
62 Jae-Eun Shin
name is, for Sircar, no doubt the same as or mistaken for that of
Dikkaravasini, residing on Dikkara, the eastern frontier of early
medieval Brahmaputra Valley where the kingdom Kamarupa flour­
ished.3 According to the eleventh-century Jvi,fikapura1J4 (51.76-7),
Kamarupa is triangular in shape and also one hundred yojanas in
length from the Karatoya to the Dikkara and thirty yojanas in breadth
from the north to the south. It is black in colour and interspersed
with innumerable hills and hundreds of rivers.4 Here, the dwelling
place of Dikkaravasini is deemed as the eastern limit of Kamarupa.
This supposition is further supported by a reference in the sixteenth­
century Yogi-nztantra ( 1.11.16-7), describing the eastern end of
Kamarupa as the abode of Dikkaravasini.5 However, her precise
location, Dikkara, is still uncertain: some scholars locate it at
Sadiya,6 while others locate it at modern Dikrang near Sadiya and
sometimes beside the Di�unadi which is identified with the mod­
ern Dikhu falling in the Brahmaputra near Sivasagar in Assam.7
In any case, the area in which Dikkaravasini resided seems to
have remained on the fringes of sedentary society, and was scarcely
absorbed in the Brahmanical social set up by the thirteenth century.
This view is corroborated by the distribution of material evidence:
while a growing number of inscriptions between the fifth and
twelfth centuries in the lower Brahmaputra Valley indicate steady
progress of the Brahmanical culture. Their woeful paucity in the
upper valley, especially in the two contiguous districts of Sivasagar
and Dibrugarh, shows that this process progressed at a much slower
pace. The extant ruins of the pre-thirteenth-century temple structure
are conspicuous by their absence in these districts. As Guha rightly
argued, the social conditions of early Assam are assumed to have re­
mained uneven in two parts of the Brahmaputra Valley. The Nagajari­
Khanikargaon, Negheriting and Deopani finds only suggest that
the Indo-Aryan thrust reached eastward up to Jorhat district south
of the Brahmaputra.8 Such different social conditions are well re­
flected in the spatial perception of Kamarupa represented in the
Naraka story of the KiilikapurarJ,a. Naraka, the legendary progenitor
of Kamarupa, drove away the Kiratas, or forest tribes of the region,
up to the abode of Dikkaravasini, when he established his kingdom
following the instructions of Vigrn.9 It indicates that the worship
.,
Kecaikhat i, Eater of Raw Flesh 63
of Dikkaravasinl was already in practice among local communities
before the conquest of Naraka, and her place belonged to the realm
of the Kiratas, a vast area of the upper Brahmaputra Valley. That
was in sharp contrast to the lower valley, especially the area of the
present city of Guwahati and its environs, which was the dwell­
ing-place of Kamakhya (kamakhya-nilaya), many Brahmins well­
versed in the Vedas and Sastras, and people in the vanya order.10
Despite being recognized as one of the seven goddesses of renowned
Sak.ta pithas, her trajectory seemed different from that of Kamakhya
due to the far-off location.11
In view of the above, Dikkaravasini the eastern frontier of Kama­
rii.pa was, in all likelihood, a tribal goddess though her name had
a Sanskrit tinge. The Kalikapurti'JJ,a (80.64b-5a) claims that the
goddess is called Dikkaravasini because she resides on the sun and
V igrn, both are youthfull (dikkara), but it is a preposterous philo­
logical interpretation. Her tribal character is well represented in
her terrible form called Tik�1;akanta or Fiery Mistress, having a
black complexion, a pot belly, a braid of matted hair (ekajata), and
ferociousness. In this form, she is famed as the Goddess U gratara.
Dikkaravasini has her other form called Lalitakanta or Charming
Mistress, but she is always worshipped in the former attribute by
votaries. 12 Moreover, she is worshipped with a sacred circle (ma'IJ,4ala),
spells (mantras), eight door-keepers (dvarapalas), and six yoginis.
The sacrificial vessels, ritual object and others, and also the place
and seat prescribed in the Uttaratantra, should all be used for
worshipping the Goddess Dikkaravasini in both her Tik�1;akanta
and Lalitakanta forms. 13
It is important to note that Dikkaravasini in the form ofTI�1;a­
kanta was compared to Ugratara or Fierce Tara, one of the most
powerful goddesses in the Vajrayana pantheon. It leads some
scholars to the strong suspicion that Dikkaravasini was of Buddhist
origin, 14 but that is not very convincing. As has been argued else­
where, a number of ferocious goddesses were gradually adopted
from Vajrayana to Sak.ta Pura1;ic and Tantric traditions of eastern
India in the eleventh to thirteenth centuries. The terrifying forms
of Tara were identified with Kali or other fearsome goddesses of
the Brahmanical pantheon. For example, Ugratara, found at the
64 Jae-Eun Shin
Ugratara Temple in Uzan Bazar, Guwahati, is quite different from
any of her known Buddhist images. It rather represents the fierce
Goddess CamUl).q.a, who has an emaciated belly, sunken eyes and
a corpse as per vehicle (vahana). This adaptation inevitably in­
volved a conscious modification through which Buddhist attributes
of the goddess were removed or reinterpreted, and different features
were superimposed upon her. 15 Ugratara depicted in the Kalika­
pura1'}a (61.636-8) is not associated with Ak�obhya, one of the five
Dhyani-Buddhas, but has Saiva traits such as a wreath of shaven
human heads, a snake necklace, a tiger skin, bone ornaments, etc.
A close affinity between Siva and Ugratara is also emphasized in
another story of the same text (81.1-23), in which she is said to be
Vama, the goddess worshipped in the left-hand path (vamacara),
among mlecchas. The term 'mkccha' has been used to describe the
people, who do not conform to the usual Brahmanical norm such
as foreigners or people living beyond a sedentary society. Similarly,
vamacara has been regarded as unconventional, revolting or inde­
cent practices by Brahmanical traditions because their rites usually
contain five esoteric elements (paiicamakaras), viz., liquor, meat,
fish, a gesture or seal, sexual intercourse. These practices were often
associated with the lower social strata, including tribal communi­
ties. Considering that, the goddess Tikg1akanta, the ferocious form
of Dikkaravasini, was not of Buddhist origin, though compared to
Ugratara. Rather she had a strong local affiliation. It is laid down
in the Yogi,nztantra (2.9.10) that the local rites and practices and
also rules of drink are not to be transgressed.
The offerings made to the Goddess Dikkaravasini demonstrate
her non-Brahmanical character even more clearly. According to the
Kalikapura1'}a (80.496-51), she in the form of Tikg1akanta specially
relished strong spirituous liquor among drinks, human flesh among
all sacrifices, sweetmeat (modaka), coconut, meat curry and sugar
cane. It talks about sacrifices including those of human beings
who were generally offered to U�I].akanta. The aim of sacrifices was
to accomplish worldly goals, such as killing one's enemy, bring
prosperity to one, having long life, wealth, fame and destroying
the innumerable obstacles in one's way. This transgressive practice
is given canonical sanction in the Kalikapura'fJ,a. Chapter 67 in the
,r

Kecaikhati, Eater of Raw Flesh 65

book, titled Rudhiradhyaya or Chapter of Blood, gives a long de­
scriptive list of sacrificial objects for the Goddesses Ca:r:ic;lika and
Bhirava, including birds, tortoises, alligators, fish, nine species
of deer, buffaloes, big lizards, bulls, he-goats, mongoose, boars,
rhinoceros, antelopes, Sarabhas (a fabulous beast with eight feet),
lions, panthers, men and blood drawn from one's own body. 17
Among them the best sacrifice was that of human beings. T herefore,
the goddess would be pleased for full one thousand years when a
man was sacrificed, and for one hundred thousand years when
three men were sacrificed. 18
As Urban pointed out, this rather motley list of victims was
drawn less from a traditional Vedic rite than from a wide array of
local traditions and the many jun gle animals sacrificed by diverse
indigenous people of the hills in the north-east. 19 Tribal rituals
and customs exerted a strong influence on the framing of rules of
0

sacrifice, and brahmins of the region admitted it by saying that

Siva himself created animals for the sake of being sacrificed. Kill­
ing for sacrifice is not murder. 20 Such open recognition to sacrifice
notwithstanding, brahmins themselves tried to keep a strict dis­
tance from its actual execution. T hey were supposed to follow the
injunction that a brahmin should never offer a lion or tiger or mail
and also the blood from his body and spirituous liquor to the
Goddess Durga. If a brahmin sacrificed either a lion or tiger or
man, he would go to hell and live for a short period in this world
suffering misery and misfortune. If a brahmin offered blood drawn
from his own body, he became guilty of killing another brahmin.21
Needless to say, brahmins should not be the victims of sacrifice,
and those who have already been given to brahmins should not be
sacrificed. 22 These references lead one to assume that the sacrificial
ritual for Dikkaravasini in the form of Tikg1akanta was presided
over by non-Brahmanical priests such as the Deoris among the
Chutiyas in later times. Moreover, no inscriptional evidence of royal
donation to the Dikkaravasini Temple by any Kamarupa rulers casts
strong doubt on the presence of brahmaqas on that sacred place in
the pre-thirteenth century. Considering the description of her seat
given by the Kii.likapurar;a, even the existence of a permanent temple
structure is uncertain. It is said that the holy stream of Sitaganga,
66 Jae-Eun Shin
which is identified as the River Dikhu by Sircar, flowed by the
side of Dikkaravasini, and she lived in the seat located on the ground
and was occasionally submerged under the water.23

THE TUTELARY GODDESS OF

THE CHUTIYAS

The period from the thirteenth to the sixteenth centuries saw a new
phase in the history of the north-east, characterized by the arrival,
growth and consolidation of a new power, the Ahoms, and the
establishment of many local powers-the Chutiyas, Kamatas,
Dimsa-Kacharis, Koches, Jayantias, etc. Among them a great deal
of importance is given to the Chutiyas in connection with worship
of the Goddess Kecaikhati, mentioned as Digaravasini (i.e. Dikkara­
vasini) in their inscriptions and later known as Tamresvari. The
Chutiyas belonged to the Bodos, a linguistic group of the Brahma­
putra Valley, speaking Tibeto-Burman languages and having dif­
ferent cognate groups within them.24 They seem to have assumed
political power in the Sadiya area in upper Assam and the conti­
guous area falling within Arunachal Pradesh at some point of time
before the entry of the Ahoms there-in. 25 However, the first con­
frontation between the Ahoms and the Chutiyas as a political power
was recorded in some Assamese chronicles such as the Deodhai
Assam Buranji only during the reign of King Sutupha (1369-76),
about hundred years after the death of the first Ahom ruler, King
Sukapha (1228-68). It is more likely that, if there was any Chutiya
state, it was of little significance till the second half of the fourteenth
century.26 Besides, the earliest known inscription of the Chutiyas
comes from that period. The Dhenukhana copper plate inscrip­
tion of 1314 saka (AD 1392) records, for instance, that King Nandin
(or Nandi), a great hero of many virtues, was the lord of Sadhayapuri
(sadhayapurisa), and his son was Satyanaraya1,1a. Sadhayapuri is
probably the same as Sadhiya or Sadiya of later times. 27 It was the
political centre of the Chutiyas.
The aforementioned inscription on the Tamresvari Temple near
Paya, about 40 km east of Sadiya, is dated to 1364 Saka (AD 1442)
and written in five lines in the Bengali-Assamese character of
 Kecaikhati, Eater of Raw Flesh 67
Sanskrit prose. It records that through the grace of Siva's feet, the
doubly illustrious Muktadharmanarayai;ia, who was the son of the
vrddharajan, effected the construction of the wall of the doubly
illustrious Digaravasini (i.e. Dikkaravasini) in bricks, etc., on the
date quoted above. Here vrddharajan, the old king, mentioned with­
out disclosing his personal name, as the father of Muktadharma­
narayai;ia, seems to suggest that the prince was administrating the
state during the old age of his father. 28 Neog considers that Mukta­
dharmanarayai:ia was Lak�minarayai:ia, the son of Satyanarayai:ia,
on the basis of the records of another inscription. In this case, the
old king was perhaps Satyanarayai;ia, the son of Nandin, the lord
of Sadiya. According to the Barmurtiyabil copper plate inscription
dated to 1313 S aka (AD 1392), Muktadharmanarayai:ia was a val­
orous king and was much given to the worship of the goddess.29
Given these two inscriptional records, the temple of Dikkara­
vasini seemed to be built, or rebuilt if there had been any earlier
construction on the site, in the period between the end of the
fourteenth century and the former half of the fifteenth centuries,
possibly in the reign of King Muktadharmanarayai;ia of the Chutiyas.
This temple is now completely in ruins, and we can-not help de­
pending on the accounts of modern explorers for a picture of the
temple and its features. Dalton mentions in his short note on the
Chutiyas of Upper Assam in 1848 that the Tamar Char or copper
temple was lately visited by Captain Vetch, probably in 1841-2.
He says:
It is described [by Vetch] as a small stone building, nearly square, built without
cement, the stones joined by iron pins not clamped. The roof was of copper, but
it has fallen in and now lies there. The interior is 8 feet square. The whole is
enclosed within a brick wall 130 feet by 200. Near the grand entrance in the
western wall is a small stone tripod.30

Major (later Col.) S.F. Hanny of Bengal Army visited the shrine
at a slightly later time, around 1848, which he calls Tamaseree
Mai or copper temple, on the right bank of the little stream, Dol
or Dewul panee. By citing the report of Francis Hamilton, he calls
the goddess of the temple 'the eastern Kamakhya. Further details
of its structure are given as follows:
68 Jae-Eun Shin
The dimensions of the interior is a square of 8 feet, the walls being about
4.5 feet thick, excepting in front, where there are two recesses on each side of
the door, which is formed of three entire blocks of scone.The outer line of
wall therefore encloses a square of about 17 feet.... [The temple was] covered
over with sheets of beaten copper, laced together through copper loops
fastened on the edges of the different sheets; as the groins, however, are
not above 5 or 6 feet long, the roof must have been rather flat; a carved
vase-shaped block, now lying in the river, in all probability formed the centre
of the dome.31
Hanny adds the comment that the style of architecture is an­
cient, but the present temple was rebuilt in the middle of the
fifteenth century with the materials of an earlier construction.32
Judging from these two accounts, Tamresvari, another popular
name of the Goddess Kecaikhati, is probably derived from a unique
feature of her temple roofed with copper (Skt. tamra; local pro­
nunciation tamar). The reason why they covered the temple with
copper sheets remains obscure; it might have been chosen for pro­
tecting the building from heavy rains which often cause damages,
or showing the bloodthirsty nature of the goddess by red colour of
the metal. Bloch, on the other hand, points to the fact that very
often the first part of similar compound names of Indian deities
enshrined in a temple contains the name of the person who puts
up such particular deities. He further suggests that the name of
Tamresvari in all probability originally meant simply 'the Isvari,
or Durga, put up by Tamra'.33 Neog supports this view by identi­
fying Tamra as one of the seven sons of Naraka on the basis of a
reference in the Bhagavatapura1Ja. For him, the Tamresvari Temple
may have been set up by this very Tamra, and this was the point
up to which the aboriginal Kiratas were expelled by Naraka.34 I
find both the arguments less convincing on two points. First, it
is rare to find goddesses named after their donors, though it is not
uncommon for Siva and Vi�i:iu in early medieval South and South­
east Asia.35 In many cases, goddesses are named after localities or
specific geographical features like mountains, rivers, caves, etc., be­
cause the locale of worship and its attributes are the most important
aspects of goddess cults, especially on a popular level. Second, Tamra
is not the son of Naraka but that of Mura, a demon (asura) having
 Kecaikhati, Eater ofRaw Flesh 69
five heads. His heads were cut-off by Kr�.l).a, and his seven sons
including Tamra were put to death by Garu9a in charge of the
outskirts of the city of Pragjyoti�a. 36 The connection between the
Tamresvari Temple and the son of Naraka, therefore, cannot be
established. Notwithstanding that, the opinion expressed by Neog
leads one to consider an important aspect of goddess cults in north­
east India which had long been associated with demonic kings.
For instance, the Goddess Kamakhya is well-known for her close
association with Narakasura or demon Naraka. As a son of the
Earth and Vigrn in his Varaha incarnation, he had been the central
figure in the fabricated genealogy of Kamarupa and the constant
source of political authority of three ruling families, the Varmans,
Mlecchas, and Palas, from the seventh to the twelfth century. 37
The royal genealogy beginning with Naraka is not found in the
inscriptional records of the post-twelfth century. Nevertheless, it
does not mean that the tradition of a demon ancestor of ruling
families faded away in the north-east. According to the Dhenukhana
inscription (AD 1392), Satyanaraya.l).a, a son of King Nandin who
was the lord of Sadhayapurl (i.e. Sadiya), had his origin in the
womb of Daivaki, Nandin's wife, forming part of the lineage of the
enemy of the gods (suraripu-Val'flsii:f(lsa-bhuto). 38 As mentioned ear­
lier, Mukta-dharmanarayal).a, who constructed the wall of Tamres­
varl (alias Kecaikhatl, Dikkaravasini) Temple or the temple itself
in AD 1442, was probably a son of Satyanaraya.l).a belonging to the
lineage of the enemy of the gods. Neog interprets this lineage as
the asura dynasty. 39
The reason for his demonic lineage is not explained explicitly in
the inscription. Also, it is not yet known whether this lineage had
some connection with demon Naraka of the previous tradition.40
It is, however, plausible that ruling powers of indigenous origin,
often represented as descendants of demons in the pre-modern
north-east, achieved political legitimacy and influence through the
royal devotion to tutelary goddesses and patronage to their temples.
The Kamariipa rulers initiated this process between the ninth
and eleventh centuries, which proceeded in the fifteenth century
under the Chutiya kings and still later under the Dimasa-Kachari
kings around the eighteenth century. The Goddesses Kamakhya,
70 Jae-Eun Shin

Tamresvari and RaI,1acary9-I attained their exalted positions in dif­

ferent places and time through this process. It can be seen clearly
that the sphere of influence of the goddess cults associated with
political powers extended from the nuclear area of the early state of
the lower Brahmaputra Valley to its periphery such as eastern Sadiya
and southern Cachar Hill where so-called secondary states emerged.
It was a continual, ongoing process. But it is by no means a repeti­
tion of the same exercise because the agents for each goddess cult
consisted of a specific social group and the way they interacted
with ruling powers differed in each case.

THE GODDESS SERVED BY THE

DEORIS, TRIBAL PRIESTS
In the case of Tamresvari Temple, the goddess was served by the
Deoris, the representatives of the priestly class among the Chutiyas.
They were permitted to continue their religious service to her in­
cluding human sacrifice even after their subjugation by the Ahems
in 1523. 41 In Brown's study of the Deoris, they were presumed to
have preserved the language, religion, and customs which had come
down to them with comparatively little change from a period an­
terior to the Ahom period. They drank strong liquor and ate all
kinds of flesh except bee£ There were four different kinds of priests
of whom each was attached to a particular division (khel), viz., the
Bar Deori (Deori Dema) and the Saru Deori (Deori Surba); the
Bar Bharali and the Saru Bharali. It was the duty of the two Bharalis
to collect the dues of the temple and to provide animals for sacri­
fice. The two Deoris performed the sacrifice; they alone entered
the temple and sung hymns, which were scarcely understood by
the common people. The Deoris had a great reputation among
ordinary Assamese for working black magic. They were believed to
be able to make their enemies die mysteriously of a wasting disease,
or resorted to in cases of loss of cattle and undetected robberies.42
Though the Deoris were well-known for their sacrificial ritual
for the Goddess Tamresvari, they also served as priests at the other
two temples. The first one was Girasi-gira or the Old Ones, called
Bura-buri in Assamese, always spoken of as a wedded pair and
 Kecaikhati, Eater of Raw Flesh 71
worshipped by the Dibongia khel. Their original temple was on
the Kundil River. The second one was Pishadema or the Elder Son,
called Boliya-hemata in Assamese, and worshipped by the Tenga­
pania khel. His temple was on the Tengapani River. The Goddess
Tamresvari was called Pishasi or the Daughter and also known as
Tameshari Mai or mother of copper temple and Kechakhati or
eater of raw flesh. Her temple was somewhere around Chunpura
on the Brahmaputra. She was worshipped by the Borgaya khel. 43
It is interesting to note that Tamresvari was considered to be a
member of the family of deities. She was the daughter of Girasi­
gira and the younger sister of Pishadema, even while renowned as
an independent goddess with her own identity. Pishasi, another
name of Tamresvari, may have been derived from Pi.sad (female
Pisaca), a flesh eating demon in Hindu mythology. This name
retains the original features of Tamresvari such as demanding hu­
man sacrifice and eating raw flesh, and is no other than Kecaikhati.
According to the myths on the subject, human sacrifice was
originally a propitiation for the introduction of sin into the world,
offered by the daughter to the old ones, viz., Pishasi to Girasi-gira.
However, in effect, it become a sacrifice to her on behalf of the
whole nation. 44 Such a shift added a political dimension to the
religious practice, closely tied to power, warfare, and royalty.45 As
Beane pointed out, 'the sacrificial cults had royal patronage, and
sacrifices were demanded of the most royal officials . ...[T]he occa­
sion tended to coincide with calamities such as war or for obtaining
wealth.'46 The KalikapurarJa already made it clear that royal mem­
bers and courtiers could offer human sacrifice, but it needed the
permission of the king.
Princes, ministers, counsellors and sauptikas (persons engaged in nocturnal
adventure or combat) may offer human sacrifice for the purpose of attaining
prosperity and wealth. If one offers a human being without permission of the
reigning king, he will find great misfortune. During an invasion or war, one
may perform human sacrifice at will, but only a royal person [may do so], and
no one else.47

A similar injunction is also found in second chapter of the Tikha

Kalpa, a manuscript found in Manipur State Library. It is said that
72 Jae-Eun Shin
human sacrifices [to the goddess living on the golden mountain]
were to be made, after the royal consent had been obtained, on the
occurrence of public calamities, such as war, or for the purpose of
obtaining great wealth. 48 The Ahem kings, too, are said to have
the exclusive right to perform human sacrifice. The four chief Deoris
thus had to make an annual visit to the Ahem court in order to
demand and obtain the sanction of the king for their annual sacri­
fice of human beings. They were treated with much respect by the
Ahem kings on that occasion. It is said that when any Hindu
Gosain (the leaders of neo-Vaig1ava order) appeared before the
Ahem .king, the latter remained seated, while the Gosain put the
garland on his neck and gave blessing; but when the four chief
Deoris made an annual visit to the court, the king rose and stood
before them to receive their salutation.49 It shows the distinctive
identity and important position of the Deoris as the only priests
who actually carried out such a dangerous ritual for the king. They
were even exempted from compulsory labour service and provided
with guards and slaves in the same way as the leaders of neo-Vai�l).ava
institutions (satras) or brahmins of established Hindu temples.
The copper plate inscriptions mentioning land grants to some
Deoris were still extant at the end of the nineteenth century, al­
though their whereabouts are unknown to us.so
The Deoris offered human sacrifices to the goddess on certain
special occasions including their annual performance, and also to
avert special calamities such as cholera, small pox, and drought.st
According to some Buranjis, people even offered sacrifices of one
hundred men to Ai (Mother Goddess) of Sadiya for extending the
life of King Pratapa Sim.ha (alias Susenghpa, 1603-41) in the last
moments of his life.s2 Here, Ai of Sadiya is probably the Goddess
Kecaikhatl of the Tamresvarl Temple. But it is not easy to know
precisely who the victims of that sacrifice were. In Dalton's report,
the victims to be immolated should be of pure caste and perfect
form, the slightest blemish or mutation rendering them unfit to be
offered to the goddess. Brahmins and members of the royal family
were exempted as a privilege. Dams, Haris, Musalmans, and women
were excluded as unfit.S3 The Tikha Kalpa takes a similar view on
the selection of victims:
 Kecaikhati, Eater of Raw Flesh 73
A brahmin or a woman should never be sacrificed. Neither should one sacrifice
his own body, as then he will be guilty of the sin of suicide. A brother, a father,
a son, a wife's brother, a sister's son, a maternal uncle, none of these should
be sacrificed; nor anyone who is acquainted with the Vedas, or has renounced
the world, or is a student, or belongs to the royal family. An enemy, a sick person,
a eunuch, one who is infirm or has defects or scars should not be offered. 54
These instructions, though canonical in character, show that
the selection of victims for human sacrifice was carefully made on
the basis of caste hierarchy, gender, religion, physical feature, etc.
They seemed to be far more elaborate than the ones given by the
Kalikapurar_ta of the eleventh century. 55 In the latter part of the
Ahom period, victims were usually chosen from among criminals
who were to be executed. When none such were obtainable, a
particular section of population or khel, known as the Baruali khel,
was called upon to produce one, in return for which certain privi­
leges were granted to the members of the khel, such as being ex­
empted from payment of ferry and market duties. 56 Gait provides
further details of this custom.
On a woman of the khelbecoming pregnant, the astrologers were called on to
say whether the child would be a boy or a girl, and if they predicted that it
would be a boy, the mother was carefully looked after, and the child was
anointed, as soon as it was born, with a paste made of tamarind and matikalai
(Phaseolus radiatus, mung bean). When a sacrifice was thought necessary,
volunteers were called for, and some person usually came forward of his own
accord; if not, a victim was taken by force. In either case, the victim was
shaved, anointed with the paste of tamarind and matika/,ai, and decked
with gold and silver ornaments. He was then conducted before the image of
goddess, when he prostrated himself, and was promptly decapitated by
the Bar Deori or the high priest. The body was left on the spot until the
following morning, when it was removed. V ictims were taken only from
amongst the juvenile male members of the clan; on attaining full age, they
became exempt, and were allowed to marry and settle down. 57
These rituals continued till the late eighteenth century. But in 1794
in the reign of King Gaurlnatha Simha (alias Suhitpangpha, 1789-
95), Sadiya was taken by the Khamtis who had migrated from
Upper Burma some fifty or sixry years back. Their chief arrogated
to himself the title of Sadiya-khowa Gohain, the officer appointed
74 Jae-Eun Shin
by the Ahoms since the early sixteenth century to administer the
area ruled by the Chutiyas. Human sacrifices in the temple of
Kecaikhati seem to have ceased to be performed from this time. 58
For the Deoris, this institution was abolished by King Gaurinatha
Sim.ha himself; who also, being unable to protect them from the
Mishmis and other tribes, removed them to the Majuli. Despite
that, their strong beliefs in the effect of human sacrifice remained
unchanged. The Deoris have remarked to Brown that 'from the
abolition of this sacrifice, the Ahom kingdom began to go from
bad to worse'. 59

THE GODDESS AND NEW IDENTITY

CONSTRUCTION OF THE CHUTIYAS

Meanwhile, there has been an interesting change in perception of

the Goddess Kecaikhati, especially in connection with a new iden­
tity of the Chutiyas. According to the chronicle found in 1850,
the Chutiyas traced their descent from Birpal (Skt. Vlrapala),
a chief belonging to the lineage of Bhl�maka.60 Bhl�maka was a
legendary king whose daughter, RukmitJ.i, was married to Kr�tJ.a.
Though he was known as the ruler of Vidarbha kingdom located
in the south of the Vindhya range in the epics and Purcu;as, he has
gained much popularity in the areas around Sadiya. The name of
KutJ.4ina, the capital of Bhi�maka, is supposed to survive in that of
the Kundil River at Sadiya.61 The ruins of two forts in Lohit dis­
trict of Arunachal Pradesh are said to be the remains of his city,
viz., Bhi�makanagara.62 Some scholars, thus, claimed the ancient
presence of Bhi�maka's kingdom in the Sadiya area,63 but there is
no conclusive evidence to support that. The earliest one found on
the site is an inscribed brick with the name of Srisri-�minaraya[J.a
of the early fifteenth century, who constructed the wall of the
Tamresvari Temple in AD 1442. 64
The prominence of Bhi�maka can be attributed to the Rukmi,:tz­
hara ,:t a, the well-known poem composed by Sankaradeva. He
interwove the two versions of story of the abduction of Rukmi[).l
by Kr�tJ.a, as given in the Harivamsa II, chapters 59-60 and the
Bhagavatapura ,:t a, 10, chapters 52-4, and incorporated distinct
 Kecaikhati, Eater of Raw Flesh 75
regional features in the poem. To the original tale he added scenes
of common domestic life, which transformed the Purar:iic story
into a narrative of popular experience. Rukmir:iI's family was repre­
sented as a prototype of Assamese Hindu family in the poem, and
King Bhi�maka was recast as an ideal father and a devout Vai�r:iava. 65
Considering the wide popularity of the poem among the people,
it is quite probable that the legendary places associated with
Bhi�maka were reproduced in the local landscape of the Sadiya
area at a later time, possibly in the seventeenth century. The neo­
Vai�r:iava movement seemed to play a significant role in transplan­
tation of the Bhi�maka legend into the area around Sadiya. Further,
this reproduction led to conscious identity construction of some
tribal groups including the Chutiyas and the Idu Mishmis of the
Dibang Valley. They claimd to be the descendants of Bhi�maka
and considerd Bhi�makanagara as the sacred heritage symbolizing
their past glory. 66
Eventually, Bhi�maka was represented as the king who autho­
rized the right of the Chutiyas' to serve the Goddess Kecaikhati as
priests in her temple. According to a popular local story, Kecai­
khati assumed the form of a stone image and lay in the Dibang
River in order to reveal her identity to the people. King Bhi�maka
wanted to lift this image and install it in a temple, but his men
failed to do it. Thereupon, the king announced that one who could
lift the image, individually or collectively, would be given the right
to officiate as priests in the Tamresvarl Temple. Many people at­
tempted to do it, but on their failure four Chutiyas having purified
mind and soul were able to lift the image and install it in the
Tamresvarl Temple. 67 This story illuatrates an important change
in the Kecaikhati cult of the Chutiyas, from aniconic worship to
iconic adoration. It was a part of the gradual process of Brahman­
ization of a tribal group, which replaces indigenous practices pro­
gressively with mainstream traditions. In fact, the expansion of
brahmin settlements and diffusion of their influence were seen in
the areas ruled by the Chutiyas from the end of the fourteenth
century, albeit on a very limited scale. 68 Dalton's report said that
some brahmins of the areas moved the Ahom kings to send one of
their members to the temple of Kecaikhati to superintend the
76 Jae-Eun Shin
rites. The Deoris would not admit for a long time the brahmin or
alter their ceremonies according to their instructions. Finally, they
consented to introduce some innovations, but continued with their
own customs and rites, declining the brahmin's assistance or inter­
ference; he was, however, allowed to perform a puja of his own
during the season of the sacrifice. 69
This shows how the Deoris tried to keep their religious hege­
mony over other social groups, probably including brahmins of
the areas, while accepting adoration of sacred images that was of the
upmost importance in Brahmanical mode of worship. Interest­
ingly, the person who authorized the priesthood of the Deoris of
the Tamresvad Temple in the aforementioned story was not the
Ahom ruler, the overlord of the Chutiyas, but Bhi�maka, the king
of the ancient kingdom of Vidarbha, renowned as one of the cul­
tural centres since the Mahabharata. He was moreover the father­
in-law of Kw.1-a, who became one of the most significant deities
in the north-east after the sixteenth century. The close tie forged
between King Bhi�maka and the Deoris is not based on any histori­
cal facts, but nevertheless it discreetly connects the wider world of
the Epic and the local centre of the goddes worship.
This is further emphasized by another popular story about con­
struction of the temple by a Chutiya king. It is said that the
Goddess Kecaikhati revealed herself before a king who was the
descendant of Bhi�maka. She told him she would be his devotee
only on the condition that he could erect a temple with a copper
roof in one single night before the first cock's crow in the morning.
In the case of the king's failure to do it, their role would be re­
versed and the king would have to be her devotee. The king worked
all night, building a temple with a copper roof, and then, just be­
fore its completion, the goddess caused a magical cock to crow.
The king had to leave the work unfinished and become the devotee
of the goddess. 70 A similar story is also told of Naraka, except in
this case the king wanted the Goddess Kamakhya as his bride and
constructed the incomplete staircase of the Kamakhya Temple. 71
The Chuciyas seemed to remake this well-known story of Naraka
and Kamakhya for explaining the origin of the Tamresvari Temple.
In their version, the temple was built by a king descended from
 Kecaikhat1, Eater of Raw Flesh 77

Bh1�maka. This king, whose name was not mentioned in the story,
was possibly Birpal, the supposed progenitor of Chutiya kingdom,
belonging to the lineage of Bh1�maka. Scholars ascribed the Chutiya
chronicle beginning with Birpal to the early nineteenth century,
perhaps written by some members of the Chutiya aristocracy as an
attempt to legitimize their claim over a part of Assam during the
establishment of the Matak kingdom in 1805, or after 1826 when
the East India Company annexed the region. 72 The lineage narra­
tive of Birpal links the peripheral power with established ones en­
dowed with all appropriate royal qualities and locates it within the
ambit of the Epic-Purat).ic world of the mainstream tradition. Look­
ing at this, the above story appears to be made at a much later time
despite its antiquarian taste. The nexus between the goddess and
the king with sacred genealogy and its antiquity are not a given
fact but an imagined reality. A significant point to remember is
that it is part of a long religious process which has been ongoing
since the twelfth century when the image of Goddess Kamakhya
and King Naraka was made in the lower Brahmaputra Valley.

THE AFTERLIFE OF THE GODDESS

Kecaikhad still enjoyed her fame in the nineteenth century al­
though the human sacrifice to her ceased during the reign of King
Gaurlnatha Siinha (1780-95). Francis Hamilton in his report on
the survey of eastern India in 1808-14, refers to three important
pilgrimage centres of Assam, including the Kamakhya Temple at
Nilachala, a temple (its name not defined) at Kaliyabar now in
Nagaon, and the Dikkaravasinl Temple at Sadiya. 73 Dalton also
said in 1848 that the copper temple of the goddess was respected
by and had its votaries among all the hill tribes in the vicinity, as
well as by the Ahom and Hindu population of Assam. It was also
visited by pilgrims, who brought offerings from regions far more
remote, perhaps from Tibet and China, as well as from other parts of
India. 74
When Bloch visited the site in 1905, however, the temple was
nothing more than a heap of broken stones past all repairs with
only a part of the wall standing; a huge tree growing over the
78 Jae-Eun Shin
debris. He identified the two figures on the lintels on the door in
the northern wall as images of Mahadeva posing as Dvarapala. On
the east of the boundary wall he found some ruins of brick and
stone buildings, amongst which he could make out three statues
of Hindu deities: Surya, KalI dancing on Siva and the third prob­
ably represented SarasvatI, if the indistinct object, which she held
in her right hand was a musical instrument. Among other inter­
esting archaeological finds on the site was a line of tiles, about
three feet apart from each other, carved on the inner sides of the
walls, except the eastern one. These carvings represented figures of
men, animals, birds, flowers, and geometrical figures without any
religious bearing. The style was that of semi-tribal kind, as in the
carvings at Dimapur and other places of Assam. Remarkable among
these were the Durga image with the maneless lion, often found in
Assamese painting. 75 It has not yet been possible to date these
images accurately, since none of stylistic features were scrutinized.
But in view of the fact that Durga puja with the earthen image of
the goddess was first introduced to the Ahoms by King Pratapa
Sim.ha in the middle of the seventeenth century,76 these images
including Durga can be assigned to the eighteenth century.
In the 1950s, Debala Mitra explored the place and wrote a short
report suggesting that the temple was originally a caturayatana
having four shrines or cells. 77 However, the temple site and structure
are now reported to be untraceable, probably because of the great
earthquake of 1950. 78 In my observation in 2012, it was found
that the Tamresvarl Temple located in Sadiya town at present is a
recent construction of not more than fifty years old. The local
people still consider the Goddess KecaikhatI as a daughter of the
Old Ones called Bura-buri in Assamese or Girasi-gira in Deori
language. They offer buffalo sacrifice to her at this new Tamresvad
Temple every three years, and a Deori who comes from a specific
family performs the sacrificial ritual.
Some other goddesses with similar names and attributes are still
worshiped by different communities in different parts of the re­
gion. 79 For instance, KachakantI (or Kancakanti) has been a
renowned goddess of the Kacharis whose shrine is now at Udhar­
bond near Silchar in Cachar district. According to a local legend,
 Kecaikh�tI, Eater of Raw Flesh 79
King K rg1acandra (1773-1813) had a dream in 1806 and then
constructed a temple for worshipping Kachakantl, one of the terri­
fying forms of Sakti. The royal preceptor pandit Sonaram Sarma on
hearing the description of Kachakanti from the king made a four­
armed golden image and installed it in a temple inside a deep forest
near present Udharbond. 80 Some scholars identify Kachakanti of
Cachar with Kecaikhati of Sadiya, based on the assumption that
when the Kacharis were in the upper Brahmaputra Valley Kecaikhatl
was their tutelary deity. 81 The Kachakanti Temple of Cachar was
also one of the principal places where human sacrifices were offered.
The ceremony was performed by a family of brahmin residents at
Udharbond, who were known as Deshmukhiyas. The practice is
said to have been put a stop to during the reign of King Krg1acandra
around 1818, in consequence of a brahmin having narrowly es­
caped being sacrificed by mistake. 82 The possible association be­
tween Kecaikhati of Sadiya and Kachakantl of Cachar, though based
on very limited historical evidence, throws light on a proliferation
of goddess cults in the process of tribal state formation of the north­
east. It does not mean a unilinear development but a continuous
adoption from the past and retrospective redefinition from the
present.
To sum up, among a number of powerful goddesses worshipped
in the north-east, Kecaikhati was infamous for eating raw flesh as
the name shows. Her cult spot, later known as the Tamresvari
Temple near Sadiya, was almost always described as the most no­
torious place of human sacrifice in the reports of British colonial
officer-scholars in the nineteenth century. They repeatedly identi­
fied the practice of human sacrifice associated with Kecaikhati as
one of the clearest signs of 'degenerate Hinduism' or of the bar­
baric characters of 'impure tribes' in the region. Sometimes, they
claimed that a wily and bigoted brahmin may have made it a price
for the liberty of proselytism to his creed. 83 Even some Indian
scholars tend to internalize such a negative image of the goddess
coming down from colonial sources. However, the history of Kecai­
khati over a prolonged period from the early medieval times to the
present rather shows her multiple identities represented in various
names such as Dikkaravasini, Tik�i:iakanta, Ugratara, Lalitakanta,
 80 Jae-Eun Shin
Tamresvari, Pishasi (Pisad?), Kachakanti and so on. Her attributes
and nature have neither been static nor monolithic, since different
social groups with their own interests, including tribal priests,
kings, brahmin and local people, participated in the making of
the goddess. Brahmanization of tribal custom and resistance to
Brahmanical dominance over tribal worship appear to go against
each other, yet the one does not necessarily negate the other. The
Kecaikhati cult is indeed remarkable as a site of constant and subtle
negotiation between those two processes which has characterized
the religious traditions of the north-east.

NOTES

1. Neog's article may serve as the most useful introduction to the Goddess
Kecaikhati , see M. Neog, 'Goddess T amresvari and Blood Sacrifices', in
idem, Religion ofthe North-East: Studies in the Formal Religiom ofNorth­
Eastern India, Delhi: Munshiram Manoharlal, 1984, pp. 109-72. For a
brief reference to the Goddess, see K.L. Barua, Early History ofKamarupa:
From the Earliest Times to the End ofthe Sixteenth Century, Gauhati: Lawyers
11 Book Stall, 1933 (1966 repr.), p. 183; B. Kakati, The Mother Goddess
1
Kamakhya, Guwahati: Publication Board Assam, 1948 (2003 repr.),
pp. 55-63; S.N. Sharma, A Socio-Economic and Cultural History ofMedieval
Assam, AD 1200-1800, Guwahati: Bina Library, 1989, p. 97; N.N.
Bhattacharyya, Religious Culture ofNorth-Eastern India, New Delhi: Manohar,
1995, pp. 102-7; N.R. Mishra, Kamakhya: A Socio-Cultural Study, New
Delhi: D.K. Printworld, 2004, pp. 5, 18; Hugh B. Urban, The Power of
Tantra: Religion, Sexuality and the Politics ofSouthAsian Studies, New York:
LB. Tauris, 2010, p. 97; Mary Storm, Head and Heart: Va/,our and Self
Sacrifice in the Art ofIndia, New Delhi: Routledge, 2013, p. 80.
2. D.C. Sircar, 'Pragjyotisha-Kamariipa', in H.K. Barpujari (ed.), The Compre­
hensive History ofAssam, vol. 1, Guwahati: Publication Board Assam, 2007
(3rd edn.), p. 64.
3. D.C. Sircar, 'Paya T amresvari (Dikkaravasini) Temple Inscription of
Muktadharmanarayar:ia, Saka 1364 (AD 1442)', in idem, Some Epigraphical
Records ofthe Medieval Periodftom Eastern India, New Delhi: Abhinav, 1979,
P· 3.
4. Biswanarayan Shastri (ed. and tr.), Kalikapurar;a, Delhi: Nag Publisher,
1991-2.
 Kecaikhat1, Eater ofRaw Flesh 81

5. Biswanarayan Shastri (ed.), Yoginitantra, Delhi: Bharatiya Vidya Prakashan,

1982.
6. P.C. Choudhury, The History ofthe Civilization ofthe People ofAssam to the
Twelfth Century, Guwahati and Delhi: Spectrum, 1959 (1987 repr.), p. 420.
7. D.C. Sircar, TheSaktapj,;has, Delhi:Motilal Banarsidass, 1948 (1998, 2nd
edn.), p. 17, fn. 3; D.C. Sircar, Studies in the Geography ofAncient and
Medievallndia, Delhi: Motilal Banarsidass, 1971 (2nd edn.), p. 163.
8. Amalendu Guha, 'Pre-Ahom Roots and the Medieval State in Assam: A
Reply', Social Scientist, vol. 12, no. 6, 1984,June, pp. 71-2.
9. Kalikapurar;,a, 38. 113-27.
10. Kalikapurar;,a, 38. 128-30. As to the spatial extent and perception
ofKamarupa, see Jae-Eun Shin, 'Region Formed and Imagined: Reconsid­
ering Temporal, Spatial and Social Context ofKamarupa', in Lipokmar
Dzuvichu and Manjeet Baruah (eds.), Modem Practices in North East
India: History, Culture, Representation, London and Routledge, 2018,
pp. 40-1.
11. For the seven Sakta pj,thas, see Kalikapurar;,a, 18. 41-50; for the trajectory
ofthe Goddess Kamakhya, see Jae-Eun Shin, 'Yoni, Yoginis andMahavidyas:
Feminine Divinities from Early Medieval Kamarupa to Medieval Koch
Behar', Studies in History, vol. 26, issue 1, 2010, pp. 1-29.
12. Kalikapurar;,a, 80. 37-9a.
13. Ibid., 80. 40-5.
14. B. Kakati, The Mother Goddess Kamakhya, p. 60; M. Neog, 'Goddess
T amresvari and Blood Sacrifices', p. 110.
15. Jae-Eun Shin, Change, Continuity and Complexity: The Mahavidyas in East
Indian Sakta Traditiom, London: Routledge; New Delhi: Manohar, 2018,
pp. 123-5.
16. C. Chakravarti, Tantras: Studies on Their Religion and Literature, Calcutta:
Punthi Pustak, 1972 (repr.), p. 55; N.N. Bhattacharyya, History of the
Tantric Religion, New Delhi: Manohar, 1999 (2nd edn.), p. 316.
17. Kalikapurar;,a, 67.3-5a.
18. Ibid., 67.18.
19. Hugh B. Urban, The Power ofTantra, p. 63.
2.0. Kalikapurar;,a, 67.39.
21. Ibid., 67.486-50.
22. Ibid., 67.101-2.
23. Ibid., 80.32b-4a.
24. R.C. Buragohain, 'A Note oftheMorans, the Borahis and the Chutiyas', in
H.K. Barpujari (ed.), The Comprehensive History ofAssam, vol. 2, Guwahati:
Publication Board Assam, 2003 (2nd. edn.), p. 61.
82 Jae-Eun Shin
25. M. Momin, 'Socio-Economic Linkages in Decline ofPragjyoti�a-Kamariipa',
in Fozail Ahmad Qadri (ed.), Society and Economy in North-East India,
vol. 2, New Delhi: Regency, p. 44. Ney Elias, on the basis ofan old Assamese
chronicle found in the possession of a Burmese king, stated that when
the Ahoms migrated to the Brahmaputra Valley, it was inhabited by three
tribes, the Chutiyas, Mo rans and Borahis, of which the Chutiyas was the
ruling group and thirty three Chutiya kings had ruled in succession before
the foundation ofthe Ahom kingdom in the early thirteenth century un­
der the leadership of Sukapha (1228-68). See Ney Elias, Introductory
Sketch ofthe History ofthe Shans in Upper Burma and Western Yunnan,
Calcutta: Foreign Department Press, 1876, p. 61. Based on this, some
scholars tried to trace the history ofthe Chutiyas back to the middle ofthe
seventh century. See S.L. Baruah, A Comprehemive History ofAssam, New
Delhi: Munshiram Manoharlal, 2002, p. 184. Yet no convincing evidence,
both archaeological and textual, asserts such an early presence ofthe Chutiya
kingdom. The inscriptions ofKamarupa between the seventh and twelfth
centuries provide no reference to Sadiya and the names ofpeople inhabit­
ing the area.
26. D. Nath, 'State Formation in the Peripheral Areas: A Study ofthe Chutiya
Kingdom of the Brahmaputra Valley', in J.B. Bhattacharjee and D.R.
Syiemlieh (eds.), Early State in North East India, New Delhi: Regency
Publications, 2013, pp. 24-5.
27. M. Neog, 'Lights on a Ruling Dynasty ofArunachal Pradesh in the Four­
teenth and Fifteenth Centuries', in idem (ed.), Prachya-sasanavali: An Antho­
logy ofRoyal Charters, etc. Imcribed onStone, Copper, etc., of.Ka,marupa,Assam
(Saumara), Koch-Behar, etc., .from 1205 AD to 1847AD, Guwahati: Publica­
tion Board, Assam, 1974 (2008 repr.), p. 94, 11.8-10; p. 218.
28. D.C. Sircar, 'Paya Tamresvarl (Dikkaravasim)Temple Inscription', p. 2.
29. M. Neog, 'Lights on a Ruling Dynasty ofArunachal Pradesh', p. 216.
30. E.T. Dalton, 'Notes on the Chutiyas of Upper Assam', 1848, manuscript
printed in WB. Brown, An Outline Grammar ofthe Deori Chutiya Lan­
guage (Spoken in UpperAssam), Shillong: The Assam Secretariat Printing
Office, 1895, p. 76.
31. S.F. Hanny, 'Notes on Ancient Temples and Other Remains in the Vicinity
of Suddyah, Upper Assam', journal ofthe Asiatic Society ofBenga4 vol. 17,
pt. 1, 1848, pp. 465-6.
32. Ibid., pp. 466-7.
33. T. Bloch, Report ofanArchaeohgical Tour inAssam inJanuary and February,
Shillong: Assam Secretariat Office, 1905, p. 25.
34. M. Neog, 'GoddessTamresvari and Blood Sacrifices', p. 115.
 Kecaikhati, Eater of Raw Flesh 83
3 5. For some examples ofSaiva and Vain,lJ.ava temples in which images oftheir
chosen god were installed,animated,and named after the donor kings,see
A. Sanderson,'The Saiva Age: The Rise and Dominance ofSaivism during
the Early Medieval Period',in S. Einoo (ed.), Genesis and Development of
Tantrism, Tokyo: Institute ofOriental Culture,2009,pp. 60,274.
36. C.L.Goswami and MA. Shastri(ed. and tr.),Bhagavatapuril:1Ja, Gorakhpur:
Gita Press, 1971, 10.59. 6-19.
3 7. & to N araka and his relationship with Kamakhya,the tutelary Goddess,
see Jae-Eun Shin,Change, Continuity and Complexity, pp. 256-9. For Naraka
in the political genealogy of Kamarupa, see Jae-Eun Shin, 'Changing
Dynasties,Enduring Genealogy: A Critical Study on the Political Legitima­
tion in Early Medieval Kamarupa', Journal ofAncient Indian History,
vol.27,2011,pp.173-87.
38. M. Neog(ed.),Pracya-sasanavatz, p.94,11.10-12.
39. M. Neog,'Lights on a Ruling Dynasty ofArunachal Pradesh',p. 212.
40. The socio-historical implication of this issue is beyond the scope of
this article.For further discussion, see Jae-Eun Shin, 'Descending from
Demons, Ascending to Kshatriyas: Genealogical Claims and Political
Process in Pre-modern Northeast India, the Chutiyas and the Dimasas',
The Indian Economic & Social History Review, vol. 57, issue 1, 2020,
pp.49-75.
41. In 1523, Ahom King Suhungmung, alias Dihingia Raja (1497-1539),
conquered the Chutiyas and annexed their kingdom to his state. A new
officer of state, known as the Sadiya Khowa Gohain, was appointed to
administer the area ruled by the Chutiyas. See EA. Gait,A History ofAssam,
Guwahati: Spectrum, 1905(2011 repr.),p. 88.
42. WB. Brown,An Outline Grammar ofthe Deori Chutiyalanguage, pp. iii-v.
43. Ibid., p. iv.
44. Ibid., p. vii. On the other hand, Dalton reported in 1848 that a yearly
human sacrifice was offered at these three temples. See E.T Dalton,'Notes
on the Chutiyas ofUpper Assam',p. 75.
45. For more details on human sacrifice and kingship in the region,see Hugh
B. Urban, The Power ofTantra, pp. 88-98.
46. Wendell C. Beane,Myths, Cult andSymbols in Sakta Hinduism:AStudy of
the Indian Mother Goddess, Leiden: E.J.Brill, 1977,p.59.
47. Kalikapura17a67. 116-17.
48. Cited in E.A. Gait, 'Human Sacrifices in Ancient Assam',journal of the
AsiaticSocietyofBengal, vol.67,pt. 3,1898,p. 62. The TikhaKalpadeals
with the mode ofworshipping Kall or Tara including human and other
sacrifices. For a briefintroduction ofthe TikhaKalpa, E.A. Gait,Report on
 �

84 Jae-Eun Shin
the Progress ofHistorical Research inAssam, Shillong: the Assam Secretariat
Press, 1897, p. 25.
49. W.B. Brown,An Outline Grammar ofthe Deori ChutiyaLanguage, pp. vi-vii.
50. For the privileges of the Deoris received from the Ahom kings, see ibid.,
p. vi.
51. E.A. Gait, 'Human Sacrifices in Ancient Assam', p. 58.
52. Kasinath Tamuli-Phukan, Asam Buran}� p. 35; Gunabhiram Barua, Asam
Buranji, p. 105, cited in M. Neog, 'Goddess Tamresvari and Blood
Sacrifices', p. 112.
53. E.T. Dalton, 'Notes on the Chutiyas of Upper Assam', p. 76.
54. Cited in E.A. Gait, 'Human Sacrifices in Ancient Assam', pp. 62-3.
55. According to the Kalikapurib;a (67.101-2), brahmaqas, caqqalas, and
princes should not be sacrificed. But the sons ofthe enemy kings, who are
captured in battle, could be offered.
56. E.A. Gait, A History ofAssam, p. 88; E.A. Gait, 'Human Sacrifices in
Ancient Assam', p. 58; E.T. Dalton, 'Notes on the Chutiyas of Upper
Assam', p. 76 .
57. E.A. Gait, 'Human Sacrifices in Ancient Assam', p. 58
58. Assam District Gazetteers, vol. 11: The Sadiya Frontier Tract Gazetteer, pt. 1,
Shillong: Assam Secretariat Press, 1928, p. 17; E.A. Gait, A History of
Assam, p. 214.
59. W.B. Brown,An Outline Grammar of the Deori ChutiyaLanguage, p. vii.
60. It was found in the possession ofsome Chutiyas by William Robinson and
published in Assamese in the Orunodoi, December, 1850. It has been re­
produced in E.A. Gait's Report on the Progress of Historical Research inAssam,
pp. 18-19 and in the DeodhaiAsam Buran}� 1932 and the SatsariAsam
Buranji, 1960, both compiled and edited by S.K. Bhuyan.
61. E.A. Gait, A History ofAssam, pp.15-16.
62. For more details on the sites, see Sukanya Sharma,A Sourcebook ofArchaeo­
logy of the Himalayan Region: Arunachal Pradesh, Kolkata: A Centre for
Archaeological Studies & Training, Eastern India, 2014, pp. 77-80.
63. See P.C. Choudhury, The History ofthe Civilization ofth;People ofAssam to
the Twelfth Century, p. 236; D. Sarma (ed.), Kamarupafasanavali, Gauhati:
Publication Board Assam, 1981, p. 73.
64. M. Neog, 'Lights on a Ruling Dynasty of Arunachal Pradesh', p. 216.
65. For the Rukmi17z-hara1Jt1, see M. N eog, Sankaradeva and His Times: Early
History of the Vai117ava Faith and Movement in Assam: Delhi: Motilal
Banarsidass, 1985, pp. 177-8; B.K. Barua, 'Sarikaradeva: His Poetical Works',
in Banikanta Kakati (ed.), Aspects of Early Assamese Literature, Gauhati:
Gauhati University, 1953, pp. 88-90.
 Kecaikhati, Eater of Raw Flesh 85

66. For the Bhl�maka legend and tribal groups, see Amrendra Kr. Thakur,
'Pre-modern Accommodation ofDifferences and Modern Innovations:
Religion and Society ofArunachal Pradesh', inTripathy and S. Dutta (eds.),
Religious History ofArunachalPradesh, New Delhi: Gyan Publishing House,
2008, pp. 345-6.
67. The story is reported in N.N. Bhattacharyya, Religious Culture ofNorth­
Eastern India, p. 103.
68. There are only two inscriptional records on this matter: the Dhenukhana
plate (AD 1392) mentions that Satyanarayar:ia gifted 600 putis ofland in
the village Ludumimari to the son of the brahmin Narayar:ia; and
Pratyak�anarayar:ia made another such gift of 600 putis in the village
Vyaghramari, and Yasanarayar:ia (or Yamanarayar:ia) still another gift of
200 putis to the son ofBhargava among brahmins. See M. Neog, Pracya­
siisaniivali, p. 94, II. 13-6; p. 96, II. 34-6. The Ghilamara plate of
L�minarayar:ia (AD 1401) records his donation of200 putis ofland in
the village Bakhana to the brahmin Ravideva, the son ofHari who was a
devotee ofVasudeva. See Ibid., p. 97, 11. 8-12.
69. E.T. Dalton, 'Notes on the Chutiyas ofUpper Assam', p. 76.
70. The story is reported in N.N. Bhattacharyya, Religious Culture ofNorth­
Eastern India, p. 104.
71. For details ofthis story, see N.R Mishra, Kamakhya: A Soci(}-Cultural Study,
p. 147.
72. D. Nath, 'State Formation in the Peripheral Areas: A Study ofthe Chutiya
Kingdom ofthe Brahmaputra Valley', p. 29.
73. S.K. Bhuyan, An Account ofAssam: First Compiled in 1807-1814 by Francis
Hamilton, Gauhati: Department ofHistorical and Antiquarian Studies,
1963, p. 57.
74. E.T. Dalton, 'Notes on the Chutiyas ofUpper Assam', p. 77.
75. T. Bloch, Report of an Archaeological Tour in Assam inJanuary and February,
Shillong: Assam Secretariat Office, 1905, pp. 92-4, cited in M. Neog,
'Goddess Tamresvarl and Blood Sacrifices', pp. 113-14. See also Annual
Report ofArchaeological Survey ofIndia 1904-05, Calcutta: Superintendent
Government Printing, India, 1908, pp. 7-8.
76. For the early Ahom kings who adopted Brahmanical traditions, see L. Gogoi,
The Buranjis: Historical Literature ofAssam, New Delhi: Omsons, 1986,
pp. 256-61.
77. D.C. Sircar, 'PayaTamresvari (Dikkaravasini) Temple Inscription', p. 4.
78. Census oflndia 2011: Arunachal Pradesh, Series 13, Part XII-A, Arunachal
Pradesh: Directorate ofCensus Operations, 2011, p. 52.
79. It is reported that there was a temple ofKecaikhati in North Lakhimpur,
86 Jae-Eun Shin
though its historical connection with the Keciikhati temple inSadiya has
not yet been examined.See N.N. Bhattacharyya, Religious Culture ofNorth­
Eastern India, p. 103.
80. GouriSen, 'Life in the Kachari Kingdom at Khaspur', unpublished PhD
thesis submitted to the University ofGauhati, 1994, p. 117.
81. S.K. Bhuyan (ed.), Kachari Buranji,Gauhati: TheGovernment of Assam
in Department of Historical and Antiquarian Studies, 1951, p. vi; J.B.
Bhattacharjee, 'The Kachari (Dimasa)State Formation', in H.K. Barpujari
(ed.), The Comprehensive History ofAssam, vol. 2,Guwahati, Publication
Board Assam, 2003 (2nd edn.), p. 393.
82. E.AGait, 'HumanSacrifices in Ancient Assam', p. 57;GouriSen, 'Life in
the Kachari Kingdom at Khaspur', p. 120.
83. S.F. Hanny, 'Notes on Ancient Temples and Other Remains in the Vicinity
ofSuddyah, Upper Assam', p. 463.
T
i

Black Magic, Witchcraft

and Occultism
Secret Cultural Practices in India

Edited by
SAJAL NAG

MANOHAR
2023
 lj

First published 2023

© Individual Contributors, 2023
All rights reserved. No part of this publication may be
reproduced or transmitted, in any form or by any means,
without prior permission of the editor and the publisher.
This edition is for sale in India, Sri Lanka, Nepal,
Bangladesh, Afghanistan, Pakistan and Bhutan.
ISBN 978-93-91928-03-2
Published by
Ajay Kumar Jain for
Manohar Publishers & Distributors
4753/23 Ansari Road, Daryaganj
New Delhi 110 002
Typeset by
Kohli Print
Delhi 110 051
Printed at
Replika Press Pvt. Ltd.