ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Ex-Criminal Tribes of Punjab

Ex-Criminal Tribes of Punjab

Numerous tribes in Punjab that were proclaimed "criminal" by the British were declared as scheduled castes in 1952. These groups, living in miserable conditions in an otherwise prosperous state, have been asking for tribal status as they are placed at a disadvantage in the present caste-based reservation system. A number of suggestions have been made to the government to consider the anomalies in identifying and characterising these groups. Their condition is worsening, occasionally breaking out into open protest, the Meena-Gujjar conflict in Rajasthan being an example. The 27 per cent reservation to the Other Backward Classes is likely to intensify the demands of these communities.


Ex-Criminal Tribes of Punjab

Birinder Pal Singh

Numerous tribes in Punjab that were proclaimed “criminal” by the British were declared as scheduled castes in 1952. These groups, living in miserable conditions in an otherwise prosperous state, have been asking for tribal status as they are placed at a disadvantage in the present caste-based reservation system. A number of suggestions have been made to the government to consider the anomalies in identifying and characterising these groups. Their condition is worsening, occasionally breaking out into open protest, the Meena-Gujjar conflict in Rajasthan being an example. The 27 per cent reservation to the Other Backward Classes is likely to intensify the demands of these communities.

Thanks are due to Suresh Sharma, CSDS, Delhi and B L Abbi, CRRID, Chandigarh for their comments.

This study is a part of the project “An Ethnographic Study of the Denotified and Nomadic Tribes of Punjab” granted by the Punjab government to the department of sociology and social anthropology of Punjabi University, Patiala to ascertain the tribal character of 14 communities at the behest of the central government.

Birinder Pal Singh ( is with the Department of Sociology and Social Anthropology, Punjabi University, Patiala.

As white man’s rat has driven away the native rat so the European fly drives away our own, and the clover kills our fern, so will the Maoris disappear before the white man himself.

– A Maori Proverb

esides many projects that the imperial government launched, one of the most important was to prepare a d etailed demographic profile of the Indian population, their subjects.


The Census of India operations were launched in the last quarter of the 19th century throughout the length and breadth of the country. The British administrators could not understand the cultural pluralism of the Indian people and their sense of accommodation and adjustment with different religions and communities. For instance, when the respondents returned the information that they are Hindu-Muslim or Hindu-Sikh the British were at a loss. How could one be both? They were looking though the glasses of m odernity where, as Michel Foucault says the episteme of difference has taken precedence over the episteme of resemblance.

The imperial rulers also could not understand the meaning of nomadism, for instance, which was a way of life with a large number of communities in India and outside. It was not only a life style but also a philosophy of life. Why stick on to a place? Why change nature and i ntervene in its practices, etc?

Thus such classes of people and communities who lived with and in nature, and practised nomadism were misunderstood by the imperial rulers and administrators. They considered them not only uncivilised and uncultured but “savages and barbarians” who posed a threat to the law and order of the society. Since they were poor and lacked definite means of production to earn their living, they were assumed to be thieves and dacoits. Thus they were dubbed as criminals. According to David Arnold (1985: 85) the Criminal Tribes Act (CTA) was used against “wandering groups, nomadic petty traders and pastoralists, gypsy types, hilland forest-dwelling tribals, in short, against a wide variety of marginals who did not conform to the colonial pattern of settled agricultural and wage labour”.

The Criminal Tribes Act

The prescriptions of W H Sleeman and others connected with the “thuggee and dacoity” operations in central India had proved this point to the rulers that such people without a permanent hearth and home were a potential threat to the society. This led to the enactment of the CTA of 1871 and the Punjab administration then readily asked for its implementation in the state. Consequently

december 20, 2008

numerous communities in Punjab chiefly the bauria, bazigar, barad, bangala, gandhila, nat and sansi were d eclared criminals. These nomadic communities were forced to settle on government lands on the peripheries of villages and cities. They were issued identity cards that were mandatory for them to carry whenever they moved out of their settlements. For instance, the bauria were made to carry “chittha”, an identity card c ontained in a m etallic pipe. Its absence was an unbailable offence and one could be arrested without any warrants. Bauria males were r ollcalled thrice a day by the local chowkidar, lambardar or at the police post.

The CTA of 1871 was modified in 1911 making certain draconian recommendations of getting finger prints, etc, and also registering these communities properly. It was noted during the fieldwork among the bauria who mentioned: “It was only their men who were made to give thumb impressions by rolling the whole thumb from left to right. It was no simple thumb impression as taken from other communities.” About the amendments in the CTA from 1871 to 1911, Meena Radhakrishna (2001: 6) writes: “The needs of practical governance led to a search for a ‘social scientific’ explanation of crime in India, connecting Indian criminality to the introduction of the railways, the new forest policy, repeated famines and so on.”

Sandra Freitag (1991: 260) also concludes:

The language of the revised Criminal Tribes Act of 1911 reflected the imperial refocus from the countryside to the urban landscape. Though the fictional label of ‘tribe’ was retained, the groups who could be p roclaimed and imprisoned without establishing individual guilt now included any collection of persons – that is, ‘gang’ – who had engaged

over time in premeditated collective crime.

The imperial authority’s criteria of characterising a community as criminal were lopsided and methodologically flawed. Anand Yang (1985: 116) argues:

Official descriptions of criminal tribes were also static and ignored the historical dimension. Banjaras, for example, were treated as a criminal tribe without any comprehension of their past. But as recent research indicates, most Banjaras were not involved in illegal activities, and those that were, did as a result of nineteenth century develop

ments which made their regular means of livelihood redundant.

Andrew Major (1999: 662) too notes:

In 1835 Sleeman reached the conclusion that the Thugs and the criminal tribes were one and the same people; in 1852 H Brereton, the Superintendent Thuggee Investigations, Punjab, reaffirmed this linkage, observing that Thugs in the Punjab, although mostly Mazhabis (the Chuhra section within the Sikh community), were recruited from the general criminal class: ‘Chuhra thieves, Sainsee burglars, and Child Stealers, and Jat dacoits’. But the linkage, in terms of a common origin, was erroneous for the simple reason that the Thugs were a professional organisation of individuals, recruited from the whole spectrum of society, not a tribe or community.

Major also cites the proceeding of the legislative department:

The Judicial Commissioner of the Punjab then wrote to the seven Division Commissioners to obtain their views. Not surprisingly, the seven were unanimous in agreeing that a system of surveillance, which had since annexation been in operation against the Mazhabi ‘Thugs’, should be applied to tribes like the Baurias and Sansis: in the words of one Commissioner, ‘the existence of a criminal class, living notoriously on robbery, is an outrage on civilised society, and their suppression is

Economic & Political Weekly

december 20, 2008

urgently called for.’ Consequently, in 1856 the Judicial Commissioner issued Book Circular No 18 which provided for the system of surveillance and control that would take effect in the rest of India only after 1860. Under the authority of this Circular all Sansis, Harnis and B aurias were to be registered at the local thanas (police stations), the lambardars of the villages in which these tribes nominally resided were to be answerable for their conduct and movements, registered tribesmen were not to be allowed to slip away from their villages w ithout a ticket of leave from the thanadar, and any tribesman found absent without leave was to furnish security for good behaviour or, failing that, be sent to jail (ibid: 665).

Exploiting Marginal Communities

One may note the promptness with which the administrators of Punjab had agreed to such a proposal that would empower them to exploit, harass or intimidate all marginal communities. During the field work we encountered numerous stories and personalised anecdotes from the members of such communities who had been harassed especially by the police. And, they quote this as an evidence of the police terror in their minds till date.1 An old b auria man quipped: ‘Bauriye hale vi police nu dekh ke seham jande ne’ (a bauria person still gets scared at the very sight of a policeman).

The CTA remained effective even after 1947. They got “independence” on August 31, 1952 when the tag of criminality was removed. Now they are called denotified tribes or ‘vimukt jatis’. It is ironical that they celebrate this day as the day of their independence and not August 15. These tribes since then had been struggling to get themselves included in the list of scheduled tribes (STs). The issue becomes grim as there is no uniformity in the characterisation of STs. One community in some states has the ST status whereas in another one it has the scheduled caste (SC) status and vice versa. Milind Bokil (2002) cites the case of phanse pardhis of Maharashtra who are STs there but their counterparts, the haran shikaris or gaon pardhis are listed as vimukt jatis. Similarly kaikadis in the Vidarbha region are called SCs but in other parts of the same state they are vimukt jatis. And, the same kaikadis are STs in Andhra Pradesh. The banjaras too have similar problem. They are vimukt jatis in Maharashtra but SCs in Karnataka.

Roy Burman (1993: 176) tries to explain this anomaly:

Scheduled Tribes is an administrative political category created under Article 342 of Constitution of India. This Article enables the President (and the Parliament in modification of the initial notification issued by the President in consultation with the Governor of the concerned state) to specify not only tribes and tribal communities but even parts of groups within any tribe or tribal community, as belonging to the category of Scheduled Tribes (emphasis added).

He continues that if this Article is seen along with Articles 342(2) and 341 of the Constitution

it becomes obvious that listing of Scheduled Tribes or Scheduled Castes is an instrument of policy (ibid: 176).


it is possible that some communities, who are treated as tribals by the anthropologists do not find mention in the list of Scheduled Tribes: on the other hand the list may include many communities whose status as tribals is debatable… The Constitution has not laid down any criteria for specification of communities as Scheduled Tribes (ibid: 176).

Virginius Xaxa (2003: 376) also confirms: “Many studies have shown that there is little scientific basis on which the present cate gorisation of STs may be defended. But since the list is linked to the extension of administrative and political concessions to the group concerned, the exclusion or inclusion of a particular group reflects political mobilisation rather than a neutral application of criteria.” Andre Beteille (1999: 59) too gives his explanation for the perpetuation of this anomaly: “The p roblem in India was to identify rather than define tribes, and scientific and theoretical considerations were never allowed to displace administrative or political ones” (emphasis added).

It was this administrative expediency and political opportunism in independent India that created more mess than resolve the issue of tribal definition and their identification. Consequently, the vimukt jatis established an All India Denotified Tribes (Vimukt Jatis) Sewak Sangh in 1982 to articulate the demands of the member communities. Its Punjab unit has also been raising the voice of seven denotified tribes (DNTs) at various levels. A ccording to the Annual Report (2003-2004) of the ministry of tribal affairs, government of India (GoI), there is no tribal population in the state.

These communities resent their SC characterisation as they b elieve themselves to be kshatriya rajputs. Hence their complaint is at two levels: One, denying them their original and traditional social status and degrading them by clubbing with the untouchables with whom they keep distance. Two, denying them their right to separate quotas as prescribed in the reservation policy of the country. They lament that the dominant balmiks and chamars take all the benefits and hardly any benefit reaches them.


The Census of India operations did not have a separate count for all the communities under study. The latest count for all excriminal tribes or vimukt jatis – bauria, bazigar, banjara, bangala, barad, gandhila, nat and sansi – in 2001 stands at 4,36,809 persons. The bazigars2 have the highest population (2,08,442) followed by sansis (1,05,337) and bauria (1,02,232). All of these with the exception of bauria who have no person in Amritsar district, are distributed all over Punjab. The nat (1,071) and gandhila (3,283) have the lowest population and they are also not distributed all over the state. Gandhilas are completely missing from Gurdaspur, Kapurthala, Nawanshahr, Fatehgarh Sahib and Muktsar. They are mainly concentrated in Jalandhar and Ludhiana.3

These erstwhile nomadic communities have been relevant to other communities in one way or another, directly or indirectly, but they neither adopted the culture of Punjabi society nor mixed with them till they were settled forcibly by the colonial government. Since these communities had no permanent abode, they were suspected as being a threat to the law and order, not only as dissenting voice against the imperialists but also as thieves and burglars endangering the right to property and security of life of the propertied classes, the pillars of the raj. They were interned in the reformatory industrial and agricultural settlements where “they should be provided work and compelled to work for their living” [Punjab Government: 122].


The special features of these settlements were: (i) There was one reformatory settlement at Amritsar where hardened criminals of these tribes were detained for a period of five to 10 years. It was practically a jail. (ii) Industrial settlements were meant for criminals of lesser degree. The members of criminal tribes were provided work either in forests, factories or railway workshops.

(iii) In agricultural settlements, at various places, 10 acres of government land was allotted free of cost to each member of the criminal tribe who had not been convicted of a non-bailable

o ffence or absence during the last 10 consecutive years. (iv) The members of criminal tribes were also given land on ‘batai’ system and the government share was collected in the form of produce.

In the reformatory school at Amritsar, children of these tribes between the ages of nine and 18 years were removed for imparting education and industrial training (ibid: 122-23).

The Criminality Tag

As mentioned above, vimukt jatis had been struggling to remove the tag of criminality and subsequently to acquire the status of STs. Thus they formed numerous state and India level organisations to spearhead their struggle such as All India Denotified Tribes (Vimukt Jatis) Sewak Sangh, All India Tapriwas and Vimukt Jatis Federation, Uttari Bharat Vimukt Jati Sangh, All I ndia Nomadic Tribes Sangh, All India Tribal Communities Dal and Punjab Scheduled Tribes Dal.

Despite persistent efforts, they have succeeded in only obtaining a mention of the correct identity in their “caste certificates”. A letter to this effect issued by the department of welfare on September 6, 1996 reads: “State government has considered this matter and has decided that the competent caste certificate issuing authorities will specifically indicate that these communities (list enclosed) belong to Vimukt Jatis while issuing the caste certificate to them” (No 10/1/96/2 SCW 1/7143). It was later realised that outside Punjab these “certificates” had no validity. During the field work I interviewed some young men of these communities who had been rejected by the Indian army on the grounds that their “caste certificates” did not tally with the list of SCs with them.

Besides political action a civil writ petition was also filed by Buta Ram Azad and 17 others against the union of India and the state of Punjab in 1975 in the high courts of Punjab and Haryana challenging the Constitution (Scheduled Castes) Order 1950, for their caste status. Justice Ajit Singh Bains having examined the criteria of the GOI for defining a SC (extreme social, educational and economic backwardness arising out of traditional practice of untouchability) and a ST (indications of primitive traits, distinctive culture, geographical isolation, shyness of contact with the community at large and backwardness) concluded:

The argument of the learned counsel for the respondents (Government) that the petitioners’ tribes do not fulfil the criteria for inclusion in the list of Scheduled Tribes is misconceived. The petitioners’ Vimukt Jatis are not untouchables… I am of the view that the Vimukt Jatis to which the petitioners belong have been wrongly included in the list of the Scheduled Castes. In fact they should have been included in the list of Scheduled Tribes and the Government of India may consider their deletion from the list of the Scheduled Castes and they may be included in the list of the Scheduled Tribes [Bains 1982].

december 20, 2008

Despite political mobilisation and rallies over the last many decades and the judgment of the high court in their favour, vimukt jatis could not achieve their goal. They attribute this f ailure to the illiteracy and poverty of their people. Moreover, they are scattered all over the state and hence unable to influence the electoral results. “We have no political leader either. Of the 29 r eserved constituencies there is none reserved for us”, the leaders lament. “Thus we could neither influence the government nor get the decision of the high court implemented. The politics of Punjab bureaucracy is more responsible for our plight than the apathy of the political leaders.”

The problem of “correct” identification and classification of these tribes is rightly attributed to the lack of seriousness on the part of administration. For instance, in the Report of the Evaluation Committee on Welfare: Regarding the Welfare of Scheduled Castes, Backward Classes and Denotified Tribes in Punjab State for the period commencing from 15th August, 1947 (December 1965 – August 1966) it is seen that but for a chapter No XVI on the vimukt jatis, the whole Report is devoted to the issues, concerns and information about two dominant SCs – balmik/bhangi and chamar/ ramdasia/ravidasia only. It is not understandable how could those 11 communities – (i) bangali, (ii) barar, (iii) bauria, (iv) nat,

(v) gandhila, (vi) tagus of karnal, (vii) dhinwara and (viii) meena of Gurgaon, (ix) bhora brehmans of Kangra, (x) mahatams of Mamdot and Fazilka, and (xi) sansis with all their “sub-castes” – listed under vimukt jatis were also classified under SCs. It is mentioned: “Of these, those at serial nos (i) to (v) and (xi) are scheduled castes and others fall under the category of backward c lasses” (Punjab government: 124).

Caste and Tribe: The Confusion

Interestingly in the annexure to the chapter there is a statement about the vimukt jatis’ position as consolidated on February 25, 1965 following the criteria: “Whether the tribe possesses tribal characteristics, viz, love of isolation (geographical, social, cultural), community way of life such as their own folklore, dances, etc, primitive way of life).” Once again there is further addition to the already existing confusion. Sansi, bauria, nat, bangali(a), barar(d) and gandhila too (no mention of bazigar banjara) i ncluded in the present study are stated to be tribes positively. It is also mentioned that “these tribes do not mix with other c ommunities.” It is qualified:

All denotified tribes except Mahatams have a tendency to live separately amongst themselves. Their customs of marriages, etc, are as heretofore. Even amongst themselves, they do not mix up and no inter se marriage takes place. They are fond of hunting and 50 per cent of their population has not left the years old evil habit of begging. Sansis are prominent beggars. Although commission of burglary is rare yet it has not vanished altogether (ibid: 131).

Just two pages later, some of these (sansi, barar(d) and bangali(a)) are referred to as castes: “Economically, socially and educationally all these castes are at the lowest rung” (ibid: 135). The Report also mentions: “As for the economic uplift schemes of the vimukt jatis, which are similar to the schemes of scheduled castes, the Committee would make analogous recommendations as done in favour of scheduled castes with following variations” (ibid: 127) (emphasis added). If tribe and caste are to be used

Economic & Political Weekly

december 20, 2008

i nterchangeably, then why are two concepts used in the same R eport? Why is there a separate chapter on vimukt jatis?

The Institute for Development and Communication (IDC) r eport has also used the government’s classification and hence a llowed the confusion to perpetuate. It does refer to the vimukt jatis, and also recommends their inclusion in the list of STs but for its own research calls them “depressed castes” [IDC 1996]. Stephen Fuchs (1992: 117) also refers to this confusion between caste and tribe. He mentions: “There are numerous castes of jugglers, tumblers, snake-charmers and the like, each with a different name, but all connected, at least in upper India, under the general term of nat or bazigar. It is difficult to state how far the term nat is the designation of a caste or a function. In Punjab, for instance, nat is usually held to be a caste, and bazigar is a subdivision of the nat.” Ibbetson and Rose (1970) subscribe to the idea that both bazigar and nat are acrobats, the former are M ohammedan and the latter Hindus.

Is this confusion between caste and tribe, a legacy of some streak in anthropology or is it a result of non-seriousness on the part of administrators and/or researchers? In the present case the latter seems more valid as it has happened so in a single R eport. Or still, is it a legacy of the political usage in vernacular of the term ‘jana-jati’ adopted as such in academic and administrative reports. I would like to suggest that when political leaders used this duo – jana-jati – in their post-independence harangues they were referring to two distinct populations ‘jana aur jati’, that is, tribe and caste. The clubbing of terms thus coalescing into one concept was done only while penning down the administrative reports and policies. As has already been mentioned above that jana and jati are two different concepts with definitely different connotations, and as suggested by Niharranjan Ray (1972) the difference between the two helps sort out the confusion in definition. The tribes are the jana and the caste is jati.

This distinction also helps us sort the matter out in the present study even if there are critics of this formulation:

Today we are not sure of this position because such categories

appear to be rather amorphous in a fluid social situation, with

frontiers continually shifting… Therefore any attempt to identify

the jana with the present day tribe is not free from difficulties” [Singh

1993: 5].

Thus we may say that besides other characteristics one dominant feature of a tribe is that it neither has caste like hierarchy within the tribe nor operates within the caste system. As a matter of fact it operates outside it. The patron-client relation characteristic of caste is absent within and outside the tribe as well. A tribal old man very aptly puts the relation of his community with the larger community(ies) as that of ‘mangan-khan di sanjh’, literally a relation or bond of begging to eat.

In the context of the denotified tribes or vimukt jatis two observations in the Report are important. One refers to the people’s perception about them. It says:

They are looked down upon as criminal tribes. It is stated those who have to watch or keep control over the commission of crimes actually exploit these people for the commission of crimes. Some of these persons have built their thatched houses and indulge in criminal acts forced by their social and other environments as and when there is an opportunity [Punjab Government: 125].

The other statement is: There had been, of late, a good deal of criticism both in the Press and in the Legislature regarding the implementation of these schemes and, by and large, it has been felt that corresponding benefit has not a ccrued to the individuals or the community as a whole, due to erroneous execution and inefficient implementation in the field. It has been accepted at all hands that the pace of progress has been rather slow and the correct implementation is further obstructed by inefficiency

and lack of consciousness among the field officers in the department (ibid: 1) (emphasis added). The situation over the past 40 years has not only remained the

same but deteriorated significantly. It would be no exaggeration to state that their “backwardness” at all the three levels – economic, social and political – has worsened beyond des cription. It is not a metaphorical but a literal statement. It only shows weaknesses and improprieties in the system that despite all paperwork and allocation of funds the designated individuals, households or communities could not get their due. The “ erroneous execution and inefficient implementation” have r emained constant over the last 40 years.

Lack of Awareness

The Status of Depressed Scheduled Castes in Punjab completed in 1996 was an attempt to study the impact of government’s welfare schemes on the depressed castes. Of the different vimukt jatis they included only bangali(a), bauria, bazigar and sansi in the study. It has projected their utterly dismal picture. Of the 13 d epressed castes 10 per cent only owned a house out of which only 2.4 per cent were ‘pucca’. There were 97 per cent households that were landless and 84 per cent were living below the poverty line. A community-wise break-up shows that 80 per cent bangali(a), 96.7 per cent bauria, 73.8 per cent bazigar and 92 per cent sansi live below the poverty line [see IDC 1996, Table 2.3 (a): 12]. The overall literacy rate of these castes is 20.4 per cent only while that of the vimukt jatis among them is abysmally low. Bangali(a) and bauria have 10 per cent each, bazigar 12.31 per cent and sansi 14 per cent only [IDC 1996, Table 2.5: 14].

It is dismaying to note that despite a threefold increase in the annual Special Component Plan for the SCs from Rs 71.15 crore in 1990-91 to Rs 205 crore in 1996-97, the benefits reaching the target groups are negligible. The survey has revealed: “Lack of awareness about government schemes for the welfare of the Scheduled Castes was one of the important reasons for very few respondents availing themselves of these schemes. It was found that only 8 per cent of the respondents had availed themselves of the schemes. Out of these 8 per cent, only 5.2 per cent had benefited from the schemes” [IDC 1996: 43]. The study points out that f ormulation of welfare schemes for the uplift of SCs is not enough. What is equally important is their dissemination to the beneficiaries as more than 60 per cent respondents were not aware of the schemes.

There is no employment potential in Punjab except in agriculture or else as daily wage labour in both rural and urban areas.

Only some of these communities and (a few amongst them), such as bangalas, earn their living through snake charming and by selling indigenous medicines. Nats perform their shows and bazigars too follow their traditional occupations to some extent. All others are given to daily wage unskilled labour in villages and towns, rag-picking or selling petty goods as ‘pheriwalas’. In all the communities we studied an insignificantly small section is working in the service sector, more than 98 per cent are given to petty jobs and errands. What is the future of these communities under existing conditions? What is their strategy for survival? The only way as their leaders have seemingly understood is that without getting the status of STs there is no way they can b enefit from the reservation policy.

A perusal of the ethnographic details collected for each of the seven denotified tribes or vimukt jatis of Punjab shows that they definitely meet all the criteria for d efining them as tribes. The standard definition of the GoI and those of the social anthropologists too have little to differ on this count. The Report of the Evaluation Committee of the Punjab government has also certified these communities as tribes. Besides, the judgment of justice Bains too has not only upheld the Rajput status of these communities but also upheld their inclusion in the list of SCs wrong. He further recommended to the government to include them in the list of STs.

After examining various definitions given by the social anthropologists and critically scrutinising various ethnographic reports prepared by us and allied information in the light of the defining parameters suggested by the government of India, I tend to conclude that all these seven communities have been historically differentiated and isolated from the mainstream village or urban society in Punjab. One very strong element that strengthens their characterisation as tribal is their nomadic and semi-nomadic n ature. By virtue of this very feature alone they do not become a part of the city or village community and hence of the caste system. Their bonding with the main society is only need based. That is why a tribal old man’s characterisation of his community’s relation with the main society is very apt – ‘mangan-khan di sanjh’

– a relation of begging to eat, or supplying such items of domestic use that were not available there and in which these communities specialised. For instance, bangalas still supply various types of medicines for eyes, skin or joint ailments, etc, and get some thing in return. Earlier they got some grain, now they take money. Similarly barads used to supply ‘kanghi’ (comb), their speciality for spinning purposes. They excelled in its manufacture. They also supplied brooms and ‘enu’. Labana or banjaras were salt traders at the all-India level. At the local level they sell needles, ‘suian’ (small) and ‘gadhuian’ (large). They also sold meat in a ‘tokra’ (basket). People preferred ‘tokrewala’ meat to any other. Besides, the bazigars and nats performed ‘bazi’ and acrobatic shows for people’s entertainment.

It may be true as well or it may be a reconstruction of their past especially after they had been forcibly settled by the colonial power and made a part of the village community where caste and varna systems are strongly entrenched, that they claimed

Economic & Political Weekly

december 20, 2008

‘kshatriya’ past. Once made to confront high and low, pure and polluting they started looking into their own social status vis-a-vis the dominant society. Hence they made an attempt to situate themselves in the overall hierarchy. K Suresh Singh (1994 (2001): 7) mentions:

The tribes have generally remained outside the varna system. Therefore, only 11.8 per cent of them recognise their place in it. Another 31.6 per cent are only aware of the varna system. Among those who recognise their place in it nearly 8.3 per cent claim to be Kshatriya, 7.5 per cent Shudras and 0.9 per cent Brahman…When it comes to the selfperception of a tribal community in the regional hierarchy we find that 171 tribes, that is, 26.9 per cent see themselves as being of a high status, while 298 tribes (46.9 per cent) perceive themselves as being in the middle position. About 25.3 per cent, that is, 161 tribes see themselves as being of low status.

Therefore this feature of the tribal communities to glorify their kshatriya past is reasonable and plausible. The problem of criminality is an issue related with this glorification. What was the n ature of crime they committed? What were the objects of their theft or stealing? And most important element was the construction of criminality by the powers that be.

Looting for Survival

Their oral history would reveal that having been pushed into the forests by the Mughal rulers, they were coerced to become nomadic and survive through hunting for almost “three centuries”. In such conditions it is difficult to ascertain if their rebellion in the form of loot and arson, individually or collectively against the rulers, earlier Mughal and later British and the propertied classes, can be interpreted as an act of nationalism and patriotism or as problem of law and order. In the words of Hobsbawm (1959: 102), the primitive rebels: “are peasant outlaws whom the lord and state regard as criminals, but who remain within peasant society, and are considered by their people as heroes, as c hampions, avengers, fighters for justice, perhaps even leaders of liberation, and in any case as men to be admired, helped and supported”. Referring to the marauding pastoralists in the 17th century Punjab, Chetan Singh (1991: 269) also mentions: “This last method (marauding) of procuring non-pastoral goods often placed the tribes in direct confrontation with the powerful Mughal state which increasingly encroached upon their traditional lifestyle as cultivated area expanded” (emphasis added).

Thus it may be interpreted that for the powers that be it was convenient and useful to classify such acts of loot and arson as problems of law and order and label the individuals or groups, and subsequently the whole community as criminal.

Bhargava’s (1949: 20) observations about the bawaria (bauria) criminality give credence to the above formulation. Referring to them as mostly illiterate and scattered all over India, having a distinct language, customs and code words

they are able to be in close communication with the entire tribe throughout India and Burma…They are particularly notorious for committing thefts from camps specially of Government officials and are probably the most clever of camp-robbers and tent-cutters to be found in I ndia (emphasis added).

He also mentions their adeptness in ‘nakabzani’, that is removing goods from a building through a hole made for the purpose and a custom of throwing three small stones one following a nother in the house they want to burgle.

According to them if a man gets up on this warning, it shows that he has earned money by hard and honest labour; if he continues to sleep, they conclude that he has acquired money by dishonest means. Hence if one does not wake up, they consider themselves justified in robbing him and distributing his wealth among those who have a better claim to it” (ibid: 20-21) (emphasis added).

Mayaram’s (2006: 141) observations about the meena criminality are relevant here too:

the stereotype of meena criminality is equally problematic and needs investigation… the ‘evidence’ of crime was often assumed rather than established… Attributing an intrinsic criminality to various groups masked the inroads made by colonialism into the livelihoods of various groups. So-called tribal peoples belonging to forest communities were inducted into systems of settled agriculture as landless labourers and in unequal trade relations that led to debt bondage. Administrative and ethnographic reporting merge crime and resistance.

She also mentions the case of aggarwal, major social anthropologist of the meos who “describes their ‘anti-social activities’ and their ‘vocation’ of ‘stealing cattle and looting pilgrims’. In a later work he confesses his naïve acceptance of the colonial ‘bias’ without indicating what made him abandon his earlier views” (ibid: 137).


Looking into the history of the itinerant community of Madras presidency, koravas, the traditional salt traders, Radhakrishna (2001: 12) writes about the prejudices of the administration: “since the koravas were an itinerant community, the administrators found it difficult to shake off some of the prejudices they c arried with them regarding European gypsies, and they seem to have simply superimposed some of these on the Indian counterparts. Moreover, the bulk of their own prejudices were shared by the high caste landlord sections.” And these prejudices were that koravas as itinerants had an “insatiable lust for wandering aimlessly”, they were idle and lazy, their “lack of any social norms, especially regarding their women” and finally about their “ostensible criminality” (ibid: 12-13) (emphasis added).

The problem of banjara criminality is no different either. W riting about the north India banjaras, Robert Varady (1979: 11-12) mentions:

Banjaras have been on the Indian scene for more than seven hundred years. While others have crumbled, these nomadic communities have demonstrated their versatility by surviving several onslaughts against their way of life…In the nineteenth century, Banjara tandas, confronted with extinction, evolved modes of adaptation which featured permanent settling, criminality, and the development of the draft oxen trade. Through emphasis of strengths and alterations of lifestyle, they entered the twentieth century with renewed vigour.

These citations and examples from numerous researchers make it amply clear that labelling the tribal communities as criminal was not based on serious consideration but mere articulation and imposition of the British administrators’ preconceived notions about them, their work and lifestyles, etc. It was not pecu liar to the colonial authority to label the local people especially those who posed resistance but equally true of the Mughals too. Mayaram (2006: 94) shows how by a sleight of language the invading group becomes “defensive”, and the subordinated peoples become “oppressors” and “plunderers”, violators of the contract whereby revenue is to be paid to the rulers.

As a matter of fact these were the “primitive rebels” of Hobsbawm who were posing problems of law and order to the authority and to the propertied classes – landlords and traders. The terminology used by the colonial administrators to refer to them in their reports speaks of their attitude towards these c ommunities. It is not one of conveying difference and distance but smacks of contempt and hatred. Sleeman (1999: 664) calls the bauria a “predatory tribe”, “infesting the lower Doab…” Sleeman further quotes from A Handbook of the C riminal Tribes of the Punjab that sansi men were “generally dark in complexion with bright sparkling eyes”, their faces were “cast in the aboriginal mould” and are said to be “very foxy” in expression; being often eaters of vermin, they could “always be detected by their smell, which is said to be a combination of “musk-rat and rancid grease”; and their religion, mostly a form of Hinduism, was “of very primitive, mixed and debased nature” (ibid: 671).

It is quite likely that the element of persecution in the oral historical description of the tribal community “of being hunted out” is actually an infiltration into their residential and economic space by the Mughal and British authorities for whom the forest was a source of great revenue, more so for the latter than to the former especially in the wake of increasing trade and commerce. In the 17th century rural Punjab

it was not only agriculture which was commercialised. The extent to which trade had become part of the ethos of Punjab is indicated by the Guru Granth. It reveals to us the idiom in which a large section of s ociety…expressed its economic values and social aspirations. From its passages it appears that both the rural and urban value systems were strongly influenced by a consciousness intimately connected with trade and commerce” [Chetan Singh 1991: 259-60].

A Cosmetic Treatment

Let me suggest at the end that under given conditions of socioeconomic inequality and existing political structure of the Indian state (including Punjab), even if the vimukt and/or other tribes are asking for tribal status, it is not going to make much difference to their social status or economic conditions. It would only be a cosmetic treatment of their problems without actually benefiting the communities as a whole. But there is no alternative either before them, except to struggle for such demands that can accrue some benefits for them, howsoever meagre and partial in the prevailing circumstances, and the given reservation policy. The government should look into their issues with foresight. I fear that the silence has the potential for a v olcanic eruption if the grievances of these communities are not properly redressed.

It is clearly reflected in the rising incidences and intensity of conflict to obtain the constitutional benefits due to them. The escalated violence in Rajasthan in June 2007 is a pointer to the likely coming events. The gujjars were not only pitted against the

december 20, 2008

government but also got entangled with the fellow community, during these years no elected Punjabi politician officially questioned the need to retain the Act, and while several assemblymen did query

meena. I must share my apprehensions that if such situations are

the registration of specific criminal tribes in particular districts, an

not handled dexterously by the government the socio-political

equal number rose to put questions concerning the communal distri

milieu will become ever more volatile. The recent decision of the

bution of posts within the Criminal Tribes Department. For the Punjabi

Supreme Court upholding 27 per cent reservation to the OBCs

elite, there were clearly more important issues at hand than the social will add fuel to the fire. salvation of the criminal tribes. For the criminal tribes, on the other hand, there must assuredly have been something faintly suspicious

It would be apt to close the discussion by the concluding

about the freedom’s ever-louder clarion call [Major 1999: 687-88]

r emarks of Major in his paper:

( emphasis added).

Interestingly, Act VI of 1924 provided for individual provinces to re-

It is dismaying to note that the role of the elected represen

peal the Criminal Tribes Act within their territories if they wished, but the Punjab Government never took up this option, even after 1937

tatives in independent Punjab is no different. They have not when a Unionist ministry came to power… A reading of the Punjab yet given the vimukt jatis and allied communities their due in a

Legislative Assembly Debates for the period 1937-46 is instructive...: free country.


1 Bauria men tell numerous anecdotes of how police used to harass them. They were made to sit at ‘thana’ (police station) or ‘chauki’ (police post) for hours together and asked to do ‘begaar’ that included odd jobs like doing domestic or agriculture work for the police officers. For details see Bauria in Appendix.

2 The list of communities given to us included banjara only and not bazigar. When we went to the field we could not find banjara anywhere but bazigars in large numbers. We were informed that the two are synonyms as the former is an all India name and bazigar is specific to Punjab. Thus they are bazigar banjara.

3 Gandhilas’ presence in Punjab is not clear. The Census of Punjab 2001 informs that they are c oncentrated in Ludhiana and Jalandhar districts only. Our field investigators found them concentrated only on the periphery of Patiala district touching the Haryana border.


Arnold, David (1985): ‘Crime and Crime Control in Madras 1858-1947’ in Yang (ed), Crime and C riminality in British India, University of Arizona Press, Tucson.

Bains, Ajit Singh (1982): ‘Judgement’, Civil Writ Petition No 132 of 1975, Punjab and Haryana High Court, Chandigarh.

Beteille, Andre (1999): Society and Politics in India: Essays in a Comparative Perspective, Oxford U niversity Press, Delhi.

Bhargava, B S (1949): Criminal Tribes, Bhargava Printing Works, Lucknow.

Bokil, Milind (2002): ‘De-notified and Nomadic Tribes: A Perspective’, Economic & Political W eekly, Vol 37, No 2, January 12, pp 1-8.

Census of India (Punjab) 2001. Government of Punjab, Chandigarh.

Freitag, Sandra B (1991): ‘Crime in the Social Order of Colonial North India’, Modern Asian Studies, Vol 25, No 2, pp 227-61.

Fuchs, Stephen (1992): The Aboriginal Tribes of India, Intrer-India Publications, Delhi.

Government of India, Annual Report (2003-2004): Ministry of Tribal Affairs, Delhi.

Hobsbawm, Eric J (1959): Primitive Rebels: Studies in Archaic Forms of Social Movements, Manchester.

Ibbetson, D and H A Rose (1883; 1970): A Glossary of the Tribes and Castes of the Punjab and NWF-P,

Languages Department Punjab, Patiala. Institute for Development and Communication (1996):

Status of Depressed Scheduled Castes in Punjab,

IDC, Chandigarh.

Major, Andrew J (1999): ‘State and Criminal Tribes in Colonial Punjab: Surveillance, Control and Reclamation of the ‘Dangerous Classes’ ’, Modern Asian Studies, Vol 33, No 3, pp 657-88.

Mayaram, Shail (2006): Against History, Against State: Counterperspectives from the Margins,

P ermanent Black, Delhi.

Miri, Mrinal (ed) (1993): Continuity and Change in Tribal Society, Indian Institute of Advanced Study, Shimla.

Punjab Government: Report of the Evaluation C ommittee on Welfare: Regarding the Welfare of Scheduled Castes, Backward Classes and D enotified Tribes in Punjab State for the Period C ommencing from 15th August 1947 (December 1965-August 1966), Department of Welfare, C handigarh.

Punjab Government (2004): Statistical Abstract of Punjab 2004, Punjab Government, Chandigarh.

Radhakrishna, Meena (2001): Dishonoured by History: ‘Criminal Tribes’ and British Colonial Policy,

O rient Longman, Chandigarh.

Ray, Niharranjan (1972): ‘Introductory Address’ in K Suresh Singh (ed), The Tribal Situation in India, Indian Institute of Advanced Studies, Shimla.

Roy, Burman, B K (1993): ‘Tribal Population: Interface of Historical Ecology and Political Economy’ in Miri (ed), Continuity and Change in Tribal Society, Indian Institute of Advanced Studies, Shimla.

Singh, Chetan (1991): Region and Empire: Panjab in the Seventeenth Century, OUP, Delhi.

Singh, K Suresh (1993): ‘Tribal Perspectives – 19691990’ in Miri (ed).

Singh, K Suresh (ed) (1994, 2001): People of India, N ational Series Vol III: The Scheduled Tribes, ASI and OUP.

Sleeman, W H (1971): Sleeman in Qudh: An Abridgement of W H Sleeman’s A Journey Through the Kingdom of Oudh 1849-1850, R D Reeves (ed), Cambridge University Press.

Varady, Robert G (1979): ‘North Indian Banjaras: Their Evolution as Transporters’, South Asia: Journal of South Asia Studies (ns), Vol II, Nos 1 and 2, March and September, pp 1-18.

Xaxa, Virginius (2003): ‘Tribes in India’ in Veena Das (ed), The Oxford India Companion to Sociology and Social Anthropology, OUP, Delhi.

Yang, Anand A (ed) (1985): Crime and Criminality in British India, University of Arizona Press, Tucson.

– (1985): ‘Dangerous Castes and Tribes: The Criminal Tribes Act and the Magahiya Doms of Northeast India’ in Yang (ed), pp 108-27.

Review of Labour
May 31, 2008
Class in Industrial Disputes: Case Studies from Bangalore –Supriya RoyChowdhury
Employee Voice and Collective Formation in Indian ITES-BPO Industry –Philip Taylor, Ernesto Noronha, Dora Scholarios, Premilla D’Cruz
The Growth Miracle, Institutional Reforms and Employment in China –Ajit K Ghose
Soccer Ball Production for Nike in Pakistan –Karin Astrid Siegmann
Labour Regulation and Employment Protection in Europe: Some Reflections for Developing Countries –A V Jose
Labour, Class and Economy: Rethinking Trade Union Struggle –Anjan Chakrabarti, Anup Kumar Dhar

For copies write to: Circulation Manager

Economic and Political Weekly,

320-321, A to Z Industrial Estate, Ganpatrao Kadam Marg, Lower Parel, Mumbai 400 013.


Economic & Political Weekly

december 20, 2008

Dear reader,

To continue reading, become a subscriber.

Explore our attractive subscription offers.

Click here


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top