Hinduized Tribal Priests in the Cult of Jagannatha

Dr. Gaganendra Nath Dash

Abstract Tribal origin of Daita and Sudha Suara priests
Introduction Conclusions
Hinduization of Tribal Deity References


Some of the present day priests of Lord Jagannatha are descendants of the tribal priests of a tribal deity, which, on Hinduization, came to be known as Lord Jagannatha.

1. Introduction

This study is based on the observations of a number of scholars, who have tried to trace and establish the tribal origin of the cult of Jagannath. According to these scholars, the deity, whom the tribal chief, as per the Nila Madhava-Jagannatha tradition, was worshiping, was not an Aryan deity. It was a tribal deity, which in the course of time was Hinduized and was transformed to be known as Lord Jagannatha. This deity was also known as Nila Madhava as per the said tradition ostensibly as a matter of strategy to cause its smooth transition from a tribal deity to a Hindu deity and to neutralize any future opposition to its legitimization (Geib, 1975; Mishra, 1971; Padhi, 1975;).

It is irrelevant here to raise the issue as to whether the Hinduization started from below i.e., by the tribals worshipping their tribal deity or from the above i.e., by the brahmans and/or kings or chieftains. It could be simultaneous both from above and below. According to Geib (1975, p.131 and p.181), though a section of orthodox Brahmins opposed the recognition and acceptance of this tribal deity, another section of brahmans accorded it acceptance and contributed towards its legitimization and Hinduization.

Such transformation released certain forces, initiated certain processes and unleashed certain actions and reactions, the prominent one being the tribal absorption into Hindu Society.

2. Hinduization of Tribal Deity

Hinduization of this tribal deity, while facilitated the tribal worshippers or priests of this Hinduized deity to get absorbed into Hindu society, it also allowed gradual entry of Brahminic rites and consequently the engagement of brahman priests in the worship of this deity as per the Hindu scriptures. But with such brahminic influence spreading, it is interesting to observe that neither the original tribal worshippers or priests just could be completely ignored or thrown out nor their descendants' rights and privileges to worship this Hinduized tribal deity could be out rightly rejected. They had to be recognized though gradually restricted in course of time. It is probable that the original tribal priests and their descendants were accorded priestly or semi- priestly recognition or at least very high status in the newly Hinduized cult, side by side with newly engaged brahmana priests. Kulke (1978a, pp.130-31),citing analogous cases, has shown why or how the worshippers / priests of other tribal deities of Orissa were accepted as priests in the Hinduized cults. The present author’s independent fieldwork studies have only confirmed Kulke’s findings.

3. Tribal origin of Daita and Sudha Suara priests

3.1 Textual Evidence

It may be mentioned here that according to the Record-of-Rights (1956a, pp.7-10) there are 1 18 categories of the temple servitors (sebaka performing different religious and ritual duties / services in the temple of the Lord Jagannatha at Puri, besides the services performed by different monasteries (matha) and by certain individuals. Out of these 118 categories of the temple servitors/ servants, only a few perform the priestly services while the others perform some other kind of services such as keeping the store, watching the temple premises and beating the drums etc and belong to different castes or jatis. It seems that from among the temple priests, performing the priestly services ‘the Daitas' and ‘the Sudha Suaras’ (sometimes written as Suddha / Sudu / Suda Suara and which includes the category Anasara Sudha/ Suddha/Sudu / Suda Suaras), who enjoy priestly or semi—priestly recognition and status (or at least very high status) in the hierarchy of the temple servants, are almost certainly of tribal origin, the descendants of the original tribal worshippers or priests.

The reasons behind believing the Daitas and the Sudha Suaras to be of tribal origin are the following.

Firstly the temple records like the Sevakarmani as well as a paper manuscript found in the Gopala Tirtha Matha of Bhubaneswar, which is closely related to the former, mention that the Daitas are of tribal origin (See Dash, 2010, p.83 and Gopala Tirtha Math MSS,ORP the descendants of the original tribal (Savara) worshipper or priest (, p.178). Further, the Karmangi of the temple of Jagannatha as well as the Record—of-Rights recognize the Sudha Suaras as a sub-category of the Daita category of the temple servants (Karmangi, 1983, p.128; Record of Rights, Part III (iii), 1956b, p.19).

Secondly in the text of Sarala Mahabharata, Deula Tola texts by (Sisu) Krsna Das and Magunia Das, the tribal origin of the Daitas as well as the Sudha Suaras has been mentioned. Sarala Mahabharata recognizes only two broad categories of the temple servants (or really the priests) of the temple of Jagannatha:

(i) The brahmanas, the descendants of the brahmana explorer Basu (thecorresponding Sanskrit form being Vasu or Visva Vasu) (Sarala Mahabharat, Musali Parba, (12/153), p.117)

(ii) The Daitas or the non—brahmanas, the descendants of the original Savara worshipper / priest Jara alias Basu(the corresponding Sanskrit form for Basu being Vasu or Visva Vasu) ( Sarala Mahabharat, Musali Parba, (12/151), p.117).

Deula Tola by (Sisu) Krsna Das, however, mentions three major categories of the temple servants (or again the priests) on the basis of their origin:

(i) The real sebakas (servants), i.e., the brahmanas, the descendants of brahmana explorer Bidyapati (the corresponding Sanskrit form being Vidyapati) by his brahmana wife (Deula Tola, 1950, p.25; 1953, p.20)

(ii) the Daitas or the non-brahmanas, the descendants of the original worshipper-cum-Savara chief Basu (the corresponding Sanskrit form being Vasu or Visva Vasu) (Deula Tola, 1950, p.25; 1953, pp.19-20)


(iii) the Sudha Suaras, the half—brahmanas or semi-brahmanas, the descendants of brahmana Bidyapati by his Savara wife Lalita, the daughter of the Savara chief Basu. (Deula Tola, 1950, p.26; 1953, p.20)

On the other hand in Deula Tola by Magunia Das one finds the following three categories of temple servants:

(i) the Daitapatis,

(ii) the Suaras, i.e., the cooks (of Balabhadra gotra or clan),

(iii) the Pasupalakas (Deula Tola 1932, p. 37. ).

According to this text the Daitapatis belong to a single category of temple servants instead of two, i.e., the Daitas and the Patis and are supposed to be descendants of the Savara chief who was worshipping the Nila Madhava in the forest (Deula Tola by Magunia Das) But this text is silent regarding the progenitors of the other two categories of the temple servants, i.e., the Suaras and the Pasupalakas. It may be mentioned that Patis (or Pati Mahapatras) are believed to be another section of temple servants who are supposedly Brahminic counterpart of the tribal or non-Brahminic Daitas.

3.2 Linguistic Evidence

Besides looking for references in different records and texts as mentioned above to establish the tribal origin of the Daitās and Sudha Suāra, attempts were also made to determine the linguistic source of the words Daitās and Sudha Suāras. Strangely one finds in literature that most of the scholars while accepting the tribal origin of the Daitās have invariably tried to derive the name from Sanskritic or Aryan sources (Padhi, 1975: p.25; Geib, 1975:p.191; Tripathi, 1978:p.224). Instead of trying to trace the origin of the word Daitā from one of such Sanskrit sources which is far—fetched it may be accepted as a word of Austroasiatic or Munda origin as the Daitās are supposed to be of Savara origin who may be identified with the Savara hill tribe of the Ganjam and Vizagapatam tract speaking an Austroasiatic tongue. Most probably this word Daitā or the original Austroasiatic word from which it has been derived (which is presently not known)originally denoted a particular class or sub-class of the Savara or Sora tribals who used to worship the pre—Hinduized Lord Jagannatha.

Similarly reasons given to show that the word Sudha Suāra is derived from Sanskritic or Aryan sources have been found untenable. Because when one assumes Sudha to have been derived from the Sanskrit word suddho., meaning pure or holy, and Suāra from sūpakāra (from sūpa meaning sauce / broth / soup) meaning cook, then follows the conclusion that Sudha Suāras are the holy temple cooks, which in reality is not the case as they do not perform any cooking service in the temple. Rather in the background of the facts that (i) Sudho was one of the eight sub-classes among the Savaras or Saurār Sora hill tribe of Ganjam and Vizagapatam tract as observed by Thruston (1975), (ii) Lord Jagannatha is a Hinduized tribal deity whom Sudha Suāras, enjoying priestly or semi priestly status, serve as temple servant, (iii) the word Sudho might have been derived from one of the three homonym morphemes viz., sUd-( to transfer grain etc. into basket with one`s own hand, to give), sUd-(to drench, to make wet), sUd·(to spread all around) or the word: sUda—(large, big, great) and (iv) ‘Suara’ might have been derived from the word ‘Saurā’ that currently denote the Savara or Sora hill tribe as per the principle of metathesis, it appears quite reasonable to assume that the Sudha Suāras are of Savara, particularly of Sudho subclass, origin.

3.3 Evidence from Census

Against the present scenario, it needs to be stated that this coastal Puri once had a tribal population worshipping a deity, which later Hinuized to be known as Lord Jagannatha; and this tribal population got absorbed into Hindu Society. In fact Sudhas, who are of tribal origin or real aborigines (O’ Malley, 1908, p.82), were, as per the 1901 census data, the fifth largest populous jati amongst jatis that inhabited Puri district; they were 41,080 out of a total population of 1,017,284 (O’ Malley, 1908, p.79). According to the 1911 census the number of the Sudhas in the district of Puri was 42,95039 out of a total population of 1,023,402 (O’ Malley, 1929, p.87), while

the total population of the Sudhas in Orissa was 79,334 (O’ Malley, 1913, p.79). It may be relevant to mention here that nowadays many Sudhas of the district of Puri claim to be of ksatriya or Khandayata and some to be of Karana caste or jati. Thus in census records they are not mentioned under the Scheduled Tribe category. Besides the Sudhas, the Savaras and the Sahars, who are clearly of Saura or Sora origin, are also found in the District of Puri (O’ Malley, 1908, p.71-72; 1929, p.78). According to O’Malley, "A large proportion of the population still consists of aboriginal races or semi-Hinduized castes, such as Bauris who number 84,188, Sahars (18,095), Pans (7,064),Savars (3,260), and Khonds (2,560) .... Every village in Khurda where these aborigines dwell has its Bauri and Savar quarter.”( O’ Malley, 1908, p.71-72). As per the 1971 and 1981 censuses, the scheduled tribes form 3.70 and 3.45 per cent respectively of the total population of the district of Puri.” (Swain, 1989, p.122; Tripathy, 1973, p.307). Besides that the scheduled castes of the district, whose percentage to total population stood at 13.53 and 12.93 respectively in 1971 and 1981 censuses are mainly of tribal origin.

Thus there exists a sound basis to infer that once upon a time the area comprising the formerly undivided district of Puri including areas in and around the Puri town was largely inhabited by the aborigines or the autochthonous tribes, which include a sizeable number of the Sauras or Savaras, especially belonging to its the Sudha section, and speaking an Austroasiatic language. Later, with the advent of the Aryans, most probably some of them retreated and / or were pushed into the interior hills or forests inhabited by their fellow tribes, both in the west and in the south, whereas others continued to inhabit the area either as aborigines or got gradually absorbed into the Hindu society. Generally the Sudhas of the south continued to retain their Austroasiatic tongue as well as their Saura or Savara identity/affiliation, perhaps because of the concentration and dominant position of the Sauras there. But the Sudhas of the west and of the district of Puri in course of time, with their gradual absorption into Hindu Society, forgot their own Austroasiatic language, substituting it by another especially of Aryan derivation, and also their Saura affiliation. This explains why most of the speakers of the Austroasiatic Saura or Savara language, in fact more than 95 per cent of them, are to be found only in south Orissa (i.e., the Ganjam and Koraput districts) and north Andhra Pradesh.

It appears, therefore, plausible that the Sudhas of the district of Puri can be connected on the one hand with the Sudhas of the south, i.e., one of the eight sub-classes of the Savara or Saura hill tribe found in the Ganjam and Vizagapatam tract as well as with the Sudhas of the west, found in the western part and even beyond the western border of Orissa, and on the other with the Sudha Suaras, a category of temple—servants of the Jagannatha temple,

Puri with either the priestly or semi-priestly status, providing a link between the former and the latter.

3.4. Tutelary Deity

That the Sudha Suāras, a category of the temple-servants of the Jagannatha temple at Puri, are of Sudha Saurā origin gets further strengthened by the fact that the tutelary deity of the Sudhas of the district of Puri was goddess Khambesvari(O’ Malley, 1908, p.82; Kuanr. 1977, p.137) and this Hinduized, tribal deity, ‘the lady of the post’, also called Stambhesvari, ‘the lady of the pillar’, perhaps a Sanskritized form of the former, has a strong iconographical correspondence with the Subhadra and the Sudarsana images of the J agannatha triad.(Kulke, 1978b, p.142) According to Eschmann et al (1978, pp177-78), "Both, the figures of Khambesvari and of J agannatha developed from the same tribal substrata. It could be shown that these types of tribal cults must have in pre-Yayati times been present on the seashore”.

4. Conclusions

(i) Thus search for Sanskritic source for the derivation of the names — either Daitā or Sudha Suāra – yields no satisfactory result. Instead, a highly satisfactory alternative tribal (i.e., Austroasiatic) source for the derivation of the name Sudha Suāra is discovered. Therefore, the tribal, i.e.,Saurā or Savara origin of the Sudha Suāras (as well as of the Daitās, who are supposed to have been very closely related to the Sudha Suāras) may be inferred which is in agreement with the already mentioned evidence in the temple records like the Sevakarmani section of Madala Panji, the Karmangi and the Record—of-Rights, as well as with the descriptions in Sarala Mahabharata and the Deula Tola texts by Sisu Krsna Das and Magunia Das.

(ii) That Daitas, another category of temple priests, and Sudha Suaras share the same origin of Saura or Savara is sufficiently indicated in temple records like the Karmangi and Record of Rights as well as in the texts Deula Tola and Sarala Mahabharat. Rather such records and texts suggest that at a later period, a sub- category of the Daita category of the priest- functionaries emerged as a separate category and in order to establish and emphasize their separate and independent identity assumed the name Sudha Suara. A sub-category of Daitas, if assumes the name Sudha Suara, which is most probably derived from the term Sudha Saura, then it can be concluded that Daitas originally belonged to the Sudha subclass of the Sauras or Savaras, and this sub-category for its independent identity preferred to assume the original tribal group name as the name of their group or category. lt appears that they had certainly forgotten their Saura/Savara affiliation, but not completely by that time (i.e., the time of their emergence as a separate major category of priest—functionaries). At least the changed name Sudha Suara (derived from the original name Sudha Saura) was there in the collective memory of the group or class however indistinct or faint that might have become over the years.

(iii) The term Savara occurring in the different versions of the Nila Madhava-Jagannatha tradition could not have been used as a generic term meaning just any aboriginal tribe as some scholars have assumed. Rather it is a specific term to denote the Sudha sub-class of the Saura or Savara hill tribe speaking an Austroasiatic language whereas Daitas belong to Sudha sub-class . So the Sudha sub-class of the Saura or Savara hill tribe were the original worshippers of a tribal deity which in course of time got Hinduized and ultimately became Lord Jagannatha of today.

To phrase it differently the pre-Hinduized or tribal substrata of Lord Jagannatha was the tutelary deity (ista devata) of the Sudha sub—class of the Savara or Saura or Sora hill tribe, speakers of an Austroasiatic language, which inhabited the area in and around Puri town in distant past. With the advent of the Aryans to the region, some of them retreated southward and some westward but others remained in that area. Again, some of them who remained retained their tribal / aboriginal way of life but many got gradually Hinduized. Of the latter some got Hinduized through the Hinduization of the tribal deity they worshipped because of their very close association with the same but others otherwise. Thus the Daitas and the Sudha Suaras are the Hinduized descendants of some of the original Sudha priests of the pre—Hinduized Lord J agannatha who got Hinduized through their close association with that deity.

(iv) The deity Khambesvari or ‘the lady of the post’ was not brought to Puri from western Orissa, as assumed, her only and real homeland, in order to be united with another local tribal deity — resulting in the emergence of the Hindu goddess Subhadra when her companion deity was Hinduized and became Lord J agannatha.. On the contrary Puri was also very much the homeland of ‘the lady of the post`, a tutelary deity of the Sudha sub—section of the Suara or Savara hill tribe inhabiting the area. The Hinduization and consequent transformation of this deity took several different directions. While one such direction led to the emergence of the semi—Hinduized deity Khambesvari, a different direction united ‘the lady of the post’ with another local tribal deity which ultimately resulted in the emergence of the Hindu goddess Subhadra, when her companion deity was also Hinduized and became Lord Jagannatha of the famous Puri triad.


  • Dash, Gaganendra Nath, 1982. Janasruti O’ Samajika Itihasa. The Jhankar, 34/4, pp.337 - 46.
  • Dash, Gaganendra Nath, 1998. Hindus and Tribals. – Quest for Co-existence - . (Social Dynamics in medieval Orissa). New Delhi, Decent Books.
  • Dash, Gaganendra Natha (in collaboration with Ranjan Kumar Das) 2010. Jagannatha and the Gajapati Kings of Orissa. A Compendium of Late Medieval Texts (Rajabhog, Sevakarmani, Deshakhanja and Other Minor Texts). New Delhi, Manohar
  • .Deula Tola by (Sisu) Krsna Das,
    (a) 1950, Cuttack: Dasarathi Pustakalaya
    (b) 1953, Cuttack: Kohinoor Press,
  • Deula Tola by Sri Kshetra Mahatmya by Maguni Jyotisha/Magunia Das, 1932, Arunodaya Pustakalaya, Cuttack. Oriya.
  • Eschmann,A., Kulke, H., and Tripathi, G.C., 1978. The Formation of the Jagannatha Triad, in The Cult of Jagannatha and the Regional Tradition of Orissa, by A. Eschmann, H.Kulke and G.C. Tripathy (eds.) 1978, pp.169-96...
  • Geib, Ruprecht, 1975. Indradyumna Legende, Ein beitrag zur Geschichtedes Jagannatha Kultes, Wiesbaden: Steiner Verlag.
  • Gopala Tirtha Math MSS,ORP. (It is an unpublished manuscript preserved in Gopala Tirtha Matha, Bhubaneswar.)
  • J.Hastings et al. (eds.) 1971, Edinburgh: T & T Clark. (First Published 1911)
  • Karmangi, the complete title is “ Sri Sri Purusottama Jagannatha Mandira Sebaka Khatani Nijog Karmangi”, i.e., the record of the services of the temple servants of the temple of Purushottama Jagannatha, edited by Sidheswar Mahapatra in Anama Sankalana-19, 1983, Cuttack: G.B.Mohanty. Oriya.
  • Kuanr.D.C., (ed),1977, Orissa District Gazetteers: Puri, 1977, Bhubaneswar: Gazetteer Unit, Department of Revenue, Govt. of Orissa.
  • Kulke,Hermann, 1978a. Royal Temple Policy and the Structure of Medieval Hindu Kingdom, in The Cult of Jagannatha and the Regional Tradition of Orissa, by A. Eschmann, H.Kulke and G.C. Tripathy (eds.) 1978, pp.125-137.
  • Kulke,Hermann, 1978b. Early Royal Patronage of the Jagannatha Cult, in The Cult of Jagannatha and the Regional Tradition of Orissa, by A. Eschmann, H.Kulke and G.C. Tripathy (eds.) 1978, pp.139-155..
  • Mishra,K.C., 1971. The Cult of Jagannath, Calcutta: Firma K.L.Mukhopadhyay.
  • Monier-Williams, Sir Monier, 1976. A Sanskrit-English Dictionary, Delhi: Motilal Banarasidass. (First Published 1899)
  • O’ Malley, L.S.S.,1908, Puri, Bengal District Gazetteers, Calcutta.
  • O’ Malley, L.S.S.,1913, Census of India 1911, vol.5, Bihar and Orissa, Part III, Tables, Calcutta.
  • O’ Malley, L.S.S.,1929, Puri (Bihar and Orissa District Gazetteers). Revised edition by P.T.Mansfield, Patna: Superintendent Government Printing Press.
  • Padhi,B.M., 1975. Daru Devata, Cuttack: Cuttack Students Store, (First Published in 1964). Oriya
  • Record of Rights, 1956a, Part III(iii), Vol. 1, Gazette No.293.
  • Record of Rights, 1956b, Part III(iii), Vol. ,4 Gazette No.203.
  • Sarala Mahabharat (Asvamedha,Musali O Svargarohana Parba,) by Sarala Das edited by A.B.Mohanty, 1970, Bhubaneswar: Department of Cultural Affairs, Government of Orissa.
  • Swain, S.K. 1989, A Portrait of Population, Orissa (Census of India 1981), New Delhi: The Controller of Publications.
  • Tripathy,B., 1973, A Portrait of Population, Orissa (Census of India 1971), New Delhi: The Controller of Publications.
  • Tripathi, G.C., 1978. Navakalevar: The Unique Ceremony of the ‘birth’ and ‘death’ of the Lord of the World, in The Cult of Jagannatha and the Regional Tradition of Orissa, by A. Eschmann, H.Kulke and G.C. Tripathy (eds.) 1978, pp.223-64