My French isn’t very good, but I have an interest in this topic as part of the pre-history for British colonial ethnography in South Asia. If you know anything about this journal, or its founder, Alexandre Lacassagne, please share in the comments.

I have not been able to figure out the reason for the silence on this topic. One possibility is that it is simply too complicated for newspapers to explain the category of DNTs – a category which is not well known by most Indians. Another is that the Gujjars are themselves resistant to being thought of as DNTs. The “Gurjar’s Community Online” website refers to the Gujjars as upper caste Kshatriyas, which they may have been in Rajasthan, although many Gujjars are Muslims and Sikhs as well. In fact, it seems they specifically rejected a move by the Rajasthan government to have them listed as DNTs.

The question of categorization lies at the heart of the current conflict. The Gujjars are agitating to have their official status changed from “Other Backward Classes” (OBC) to “Scheduled Tribe” (ST). These are two broad categories in India’s complex system of “reservations.” As the BBC explains:

The communities listed as the Scheduled Castes (SCs) are essentially the lowest in the Hindu caste hierarchy locally referred to as Dalits.

The Scheduled Tribes (STs) are the people living in the forests or on the hills, physically isolated from modern life, but are not necessarily socially backward.

The Other Backward Classes (OBCs) comprise the castes – in the middle of the Hindu caste hierarchy – who do not face so much exclusion or isolation in society but are educationally and economically backward.

The identification of communities in the three categories is based on a data prepared in 1935 by the British when they ruled India.

The current situation was provoked by the electoral politics of the right-wing BJP party, who won the support of the prosperous Jat community in the 1999 elections by promising to have them listed as OBCs. The Jats make up nearly 15% of Rajasthan’s population.

“Once Jats were identified as OBCs , the Gujjars who were already placed in the OBC category felt threatened. They felt the better-off Jats would corner the benefits of reservation,” said Professor Sheth.

Their move has also brought the Gujjars into conflict with the Meenas, another DNT community who are listed as a Scheduled Tribe in Rajasthan. Last year fighting broke out between the two communities as the Meenas fought to prevent any expansion of the ST category.

This situation is typical of a problem faced by the DNT activists we’ve been working with in India. In each state DNTs are listed under different categories: OBC, ST, SC, DNT, or nothing at all. And, as we see with the Meenas and the Gujjars, sometimes different DNT communities are listed differently within the same state. The result is that it is very difficult for DNT communities to come together over their commonalities in order to forge a nation-wide DNT movement.

Parama Roy does just that in the chapter on thuggees in her book Indian Traffic:

The lack of independent witnesses, the unavailability in many cases of both bodies and booty—the sheer paucity of positivist evidence, in other words—could only be resolved in one way. The most important criminal conspiracy of the century (of all time, some of the authors claimed) could be adequately engaged only by a new conception of law. … Since the law as currently defined made the complicity of individuals in particular crimes almost impossible to establish, specific criminal acts were no longer punishable as such. Instead, it was … enough to be a thug, without actually being convicted of a specific act of thuggee, to be liable to the exorbitant measures of the Thuggee and Dacoity Department. … It permitted the arrest of entire families, including women and children, as legitimate means of entrapping active (male) thugs; since thuggee was supposed to be a family affair anyway, transmitted in the genes and passed on from father to son, wives and children were also fit targets for the colonial state’s punitive and corrective measures. The act admitted the testimony of approvers [convicts who confessed in exchange for a pardon] in lieu of the testimony of independent witnesses (which had been disallowed under Islamic law), a move which created a remarkable mechanics of truth production and conviction.

… All those identified as thugs by approvers’ testimony were automatically guilty, even if no specific crimes could be proved against them and even if there was no (other) evidence of their ever having associated with other thugs.

Of course, the British where themselves a little worried about the quality of such evidence:

The fact that approvers’ testimony was “tainted” and that they might either wittingly or unwittingly implicate the innocent was undeniably an issue, though anxiety on the score was aired only to be promptly shown up as unfounded. … These testimonies were not required, under Act XXX, to be matched against the reports of independent witnesses or against the weight of circumstantial evidence; and none of the accused had the benefit of counsel, so the approvers were never cross-examined by anyone other than the officers of the Thuggee and Dacoity Department.

As reported by Kim Wagner, at one point “the government went as far as removing a judge from his post because he claimed thuggee did not exist and refused to cooperate in the operations against them.”

The anti-thuggee campaign undertaken by Sleeman and his successors was quite extensive, eventually stretching across the continent, and marked a significant change in the relationship between colonizer and colonized. Mike Dash’s book states that between 1826 and 1848 4,500 men were tried for being a thuggee. (Dash actually says for “thug crimes,” but as we’ve learned from Roy it was not necessary to prove involvement in specific criminal acts.) Of these 4,500, 504 (one in nine) were hanged, and “three thousand more were sentenced to life in prison,” with “most of the rest” either serving between seven and fourteen years’ hard labour, or dying in prison awaiting trial. That’s a lot of executions and prison sentences on very questionable legal practices, although, to be fair, our current system still executes a lot of people on pretty flimsy evidence.

So, did the British invent thuggees?

I suppose that really depends on what you mean by “invent.” Certainly there were highway bandits in India before the British. Many of these used the Thuggee trademark method of strangulation. It is even very likely that these murderers practiced some rituals in connection with their activities. All these are confirmed by Kim Wagner’s carefully researched paper on the subject, which stands out in its careful exploration of pre-Sleeman sources. He argues that while after 1830 the British very likely compelled or encouraged prisoners to adopt their own narrative to a pre-defined script, the earlier sources did not.

But what did Wagner find? For one thing, he found absolutely no evidence that thuggees were part of a wide-spread cult engaging in Kali-worship. As he says, even ordinary criminals,

who were never assumed to be motivated by religious fervor, would also hold a ceremony or puja after a successful robbery and make votive offerings to a deity. Yet nobody would suggest that they were religious fanatics who robbed and plundered as a means of worship to the Goddess.

He points out that the confessions never mentioned Kali (although Sleeman did in his notes). Wagner attributes the focus on Kali to later Orientalists who had a very limited grasp of Hindu goddess-worship. He even suggests that some informants emphasized the religious aspects of their crimes in the face of the “extreme interest in the subject exhibited by the British” as well as the desire to be absolved of responsibility for their purported actions.

Even the supposed signature methods of the thuggees turns out, upon further examination, to be quite varied, often involving swords or poison rather than just strangulation. Moreover, when we explore the political economic context, rather than an ancient ritual cult stretching back centuries, we find thuggees emerging in the context of regional power struggles, often being supported by local landlords.

Wagner wants to reclaim thuggees from the dustbin of history, arguing that “travellers were strangled and plundered by bands of robbers in early 19th century India if not earlier.” And while he makes some important correctives to the revisionist accounts, I think he misses the point being made by Roy and others. They are not claiming that the British actions were completely divorced from local realities. They are arguing that the British conception of these local practices tell us more about the fears and interests of the colonial rulers than they do about the local reality. Wagner has told us something valuable about that local reality, but not about how and why that reality came to be what it was under Sleeman.

I see an analogous situation over the use of the term “‘al-Qaida’ fighters” to refer to the enemy in Iraq. Sure, there is a group in Iraq which calls itself al-Qaida, but if we want to understand how that term is being used we have to understand the dominant narratives surrounding the War on Terror. And, indeed, part of that narrative is framed by movies like The Deceivers and Indiana Jones and the Temple of Doom, both of which draw on Sleeman in their depiction of thuggees. Indeed, it may not be a coincidence that Mike Dash’s 2005 book became a best seller.

Related: Colonial Ethnography

UPDATE: Seems that Kim Wagner has a new book out on the topic: Thuggee: Banditry and the British in Early Nineteenth-Century India

UPDATE: I wrote this post primarily as a preparation for challenging the Wikipedia page on the subject, and was overly harsh on Mike Dash because I hold his book responsible for that page, even though he specifically distances himself from the myth of the Thuggee in several places. Mike Dash left some comments on my Wikipedia talk page, and I feel it is worth reprinting them here and addressing them:

Hi Kerim

Thanks for your note on the Thuggee talk page, which I try to monitor even though I’ve sworn off actually contributing to the article.

I’ve had a look at your blog post and have to wonder if you’ve actually had a chance to read my book? If you have, I’m rather puzzled as to why you present my point of view as being that of a believer in the old colonialist view of Thugs as members of a religious cult. In fact the book features a whole chapter which discusses the issue and concludes there’s absolutely no evidence that Thugs were anything other than especially unpleasant and ruthless robbers, whose worship of Kali was entirely typical of Indian criminals of that period.

The meaning of my subtitle is a subtle one: that the “true story” is that there was no cult. Sadly, the fact that the chapter discussing religious beliefs falls starts on page 219 of the book has fooled more than one lazy reviewer who’s not bothered to read that far into assuming my views are of the old-fashioned sort.

In fact I spent three years doing primary research in the archives in the UK and India perfectly aware of the revisionist perspective and on the lookout for evidence for and against the reality of Thuggee. Again, if you’ve read my book you’ll know there are lengthy discussions of the reliability of the evidence presented at the various trials.

In case you haven’t, my position is this:

[i] The alleged modus operandi of the Thug gangs – ”invariably” seeking to murder their victims before robbing them – is highly distinctive and apparently unique. As such it should be possible to distnguish alleged Thugs from other sorts of criminals, and Thug crimes from other robberies

[ii]  Close reading of thousands and thousands of pages of the MS material in London and Delhi shows that the British used “approvers” to exhume a minimum of 1,100 corpses from spots identified by the informants, which has to imply they had knowledge of at least that number of murders

[iii]   While Sleeman’s legal processes were far from displaying modern concern for the rights of the accused, he and his associates did go to considerable lengths to separate informants at the time of their arrest and cross-check their stories. No one was executed on the word of a single informant. I don’t say no alleged Thugs were innocent – almost certainly some innocent men were executed – and I do feel standards of evidence clearly became considerably more lax when new laws were passed in the mid 1830s to make it easier to convict alleged Thugs who were only peripheral members of their gangs. However, it would be wrong to suggest that the East India Company was uniquely biased or racist in the way it organised its trials. In fact many accused murderers in the US, UK and independent Indian states experienced trials that were at least as weighted in favour of the prosecution in the 1830s. This too is clearly laid out in my book

[iv]   In some cases, though certainly not all, there was a good deal of corroborative evidence in the shape of recovered loot, and even the testimony of survivors, which suggests at least some approver testimony was pretty reliable

[v]   Roy and other revisionists have, so far as I can tell from their writings, not bothered to consult primary sources to check or verify any of this; their writings are based on secondary material, which is much less satisfactory.

In short, I agree almost entirely with Wagner, whose views I note you cite with approval, and who believes in the existence of Thuggee as a distinct form of crime, but not as a religious cult of any sort.

If you’ve read my book I’m rather surprised that you misrepresent my views so badly. If you haven’t then I do think it might be an idea to pick it up!

All of this said, I do think it would be an idea for the article to be rewritten to include a section setting out the arguments in the dispute between Roy and Wagner, say. (Wagner is actually pretty critical of Roy, certainly much more so than he has been of me.) I think the debate breaks down more as one between historians and anthropologists, which means it’s certainly an interesting one.

Best

Mike

First of all, I want to thank Mike Dash for taking the time to respond. Secondly, I thank him for helping to improve the Wikipedia article, which was a major motivation behind my writing this post. I’m sure it will be better as a result. (Dash seems to be very active on Wikipedia.) Third, I have read Mike’s book – very carefully. Fourth, I want to remove the unstated impression that Mike Dash supports the notion of a Kali cult. His chapter on this regard is very clear that this was largely Sleeman’s invention. As he says: “The emphasis placed by Sleeman … on the role of religion in Thug life was thus enormously exaggerated.”

However, I do stand by my comments about Dash, even if I regret the tone. Namely, I believe he is wrong to present the “Thuggee as a distinct form of crime,” and I think this view comes from his placing too much reliance on the Sleeman archive and the testimony of convicts in an enormously unfair system. Here is what Dash says in his “Notes on Sources”:

Ramaseeana is far from an ideal source; the ‘Conversations’ have been translated and perhaps edited, losing nuance in the process, and the Thug prisoners answer only the questions Sleeman saw fit to pose, which are not always those we might wish to ask today. Nonetheless, the material – containing as it does numerous repetitions, contradictions and even statements that fly directly in the face of opinions that Sleeman himself put in print – does seem to have been published in a more or less raw state. The ‘Conversations’ offer the most fascinating and compelling insight into the thoughts and motives of the Thugs themselves.”

This reads to me like the story about the person who looks for their keys under the lamppost because the light is better there. Although I have not spent the time with the primary material that Dash has, I have looked through the archives from that period and I wouldn’t want to have to write a book which relied so heavily on such sources (see my previous posts on this topic). While the Wagner article (I have not yet read the book) does give lip-service to the view that Thuggee is a “distinct form of crime,” his own account seems to share more in common with the revisionists, highlighting as it does the importance of local politics, the varied methods of killings, etc. In the end, the only thing that Wagner proves in this regard is that it is likely the word “Thuggee” was used to describe crimes before the British became obsessed with the topic. See my comments above about al-Qaida in Iraq.

Reading Dash’s book, and even Wagner, one can not help but feel a strange tension. They attack the revisionists, and yet are themselves revising the history. They point out the unreliability of the archive, but then fault the revisionists for not placing more credence in it. They distance themselves from the myth of the Thuggee even as they seemingly trade in this myth. In the end it seems to be a matter of emphasis. Does one emphasize the unreliability of the archive and the political economic context, or does one dig through the archive to find the molehill of truth upon which the mountain was built?

When doing research last summer in the British colonial archives I read numerous colonial ethnographies of the so-called “Criminal Tribes” (as DNTs were then known). Many were written by policemen, and the information in them was written for the express purpose of identifying such criminals. Gunthorpe’s 1882. Notes on Criminal Tribes Residing in, or Frequenting the Bombay Presidency, Berar and the Central Provinces, Lemarchand’s 1915, A Guide to Criminal Tribes, and, also from 1915, Naidu’s The History of Railway Thieves : With Illustrations & Hints on Detection are all in many ways the same book with slight variations. They freely stole from each other and the style was essentially the same. Numerous other such guides were circulated among the various colonial agencies.

They are like bird watching guides, identifying common habits and markings which will help you spot a criminal among the crowds. From Lemarchand:

Bhampta: Working in lots of three. Often disguise themselves as Marwadi or Hindu traiders, Lingayats, Jangam, Brahmans or shepherds. They are sometimes seen as minstrels, Sanadikorwas or Dakkhani Bhats. They are most commonly met with as Marathas. When posing as Gosains they add the suffix “das” to their names.

Barwar: Accompanied by women who pose as Brahmains and keep their faces veiled.

Sanoria: Gang consists of 2 to 15 or 20. Never accompanied by women.

Chandravedi: Gang comprises 10 or 20 half men, half boys. They alwasys work with a boy between 8 and 12 years of age called the “Chawa”, the man being styled Upaidar. They work by signs and secret vocabulary.

Some of the information was gathered from the confessions of convicts, but much of it seems to have been the result of embellishments and variations of previous works (“remixing” might be a polite way of describing it). A fair amount has been written about such colonial practices, but it wasn’t until I immersed myself in descriptions of which tribe ate jackal meat and which did not and which community’s women were faithful to their men (with each book contradicting the previous one) that I became aware of the true absurdity of this literature.

What is really shocking is just how little has changed a hundred years later. I was motivated to write this post when I stumbled upon this 1999 article from the Indian Express News Service:

The Modus Operandi branch of the police force, which works under the DCB, based on the evidence and eyewitness accounts can thus exactly point out the gang involved in the crime, sometimes even making it possible to identify gang members based on information provided and previous records.

Among the main gangs active in South Gujarat are the Chaddi Banian Dhari, Dafer, Kevat, Waghris, Bawaris, Nats, Sansis, Shikliyar, Jhaver Thutho, Chharras and other gangs. Police records made available to Express Newsline list distinguishing features of various gangs that help the police identify and track them down.

For example, the Bawari gang is known to camp at railway stations before striking. They use the `rumali’ method, where they bend grills of houses to force their way inside. Other gangs like the Dafers and Chaddi Banian Dharis survey possible targets by posing as beggars, vendors and the like. Dafers are known to possess firearms but use these only when challenged. The Chaddi Banian Dhari gang, as the name suggests, are dressed in shorts and banians and have their faces masked. They strike only on highways and of late, have been known to raid houses on the outskirts of cities and towns. The Shikliyars are known to manufacture country made firearms and sell these to gangs with whom they are connected.

Of course, the continuation of these practices requires explanation. There is no reason the past must necessarily burden the present. A proper critique cannot be content at simply pointing out the crimes of the past, but must also ask why colonial practices are still so prevalent in modern India. (It would also be interesting to compare this to other forms of “racial profiling.”) Still, pointing to these continuities is at least a start.

Previously:

• Fingerprinting, Thievery, and Bob Marley
• Anthropometry: Alive and Kicking
• Hidden World: Visiting The British Colonial Archives

It was in the course of searching for some more information about the topic that I came across Vinay Lal’s review of Rai Bahadur M. Pauparao Naidu’s 1915 book: The History of Railway Thieves, with Illustrations and Hints on Detection. Lal’s article discusses the role of colonial anthropology in creating the category of “criminal tribes”, but since I am already well aware of this story, my attention was caught by his tangential account of the origins of fingerprinting in colonial India:

Naidu’s matter-of-fact references to fingerprinting scarcely reveal the manner in which fingerprinting came to be developed and the extraordinary role of the Indian police in enabling its use as the most reliable method for the detection of criminals the world over. It is just shortly after the Rebellion of 1857-58 that William Herschel, Magistrate at Jungipoor on the upper reaches of the Hooghly, realized its uses as a method of identification. … Herschel then left for England, but in India fingerprinting had another proponent, Edward Henry, who in 1891 was appointed Inspector-General of Police for the Lower Provinces, Bengal. Henry first experimented with the anthropometric system, but was not satisfied with the accuracy of the measurements. In a report submitted to the Government of Bengal in 1896, Henry detailed the experiments he had conducted with fingerprints, which he observed were not only inexpensive to obtain, but also a surer means of detecting and confirming the identity of any given person. Henry is then said, with the aid of a team of Indian assistants, to have developed a system of classification under which 1,024 primary positions were identified, which when considered along with secondary and tertiary subdivisions, made fingerprinting a fool-proof form of fixing identity.

Lal goes on to explicate the vital role played by of two of Henry’s subordinates: Azizul Haq and Hem Chandra Bose, as well as to discuss how the Indian origins of fingerprinting came to be suppressed.

If the English police in England had been unable to devise such a system, clearly fingerprinting could be of no great use. This was put bluntly in a letter appearing in an English newspaper, signed by one “disgusted Magistrate”: “Scotland Yard, once known as the world’s finest police organization, will be the laughing stock of Europe if it insisted on trying to trace criminals by odd ridges on their own skins. I, for one, am firmly convinced that no British jury will ever convict a man on ‘evidence’ produced by the half-baked theories some officials happened to pick up in India.”

What redeems this foray into forensic anthropology from being completely tangential is the a comment Lal makes about the lack of any empirical evidence in the prosecution of members of criminal tribes:

Much like the thugs, the criminal tribes were said to be endowed with an innate criminality, and as another official stated apropos the Bawarias, one could easily gain an estimate of the “conditions under which their natural aptitude for thieving has been fostered until the practice of it has become ingrained into their daily life as to assume the features of a hereditary and criminal profession.” Colonial officials had little more to do than to assert this genealogy for the criminal tribes, and as Sanjay Nigam has aptly noted, once the incidence of crime associated with the criminal tribes had been understood “as a species of a well-known, dangerous genus, empirical detail counted for little.” So much for the much-vaunted regime of fact, the dedication to empiricism, on which the English prided themselves.

This is still very much the case. Just today we interviewed a man who was arrested, horribly beaten, and sentenced to ten years in jail just because he was a Chhara. He eventually won his case on appeal and got out of prison after two years. True, he makes his living as a thief, but he was not caught thieving. Moreover, he was charged with “armed robbery,” a much more serious crime, far outside the modus operandi of the Chhara. As he recounted his story, I couldn’t help but think of this famous Bob Marley song.

Beyond the historic injustices Denotified Tribes (DNTs) faced during the British Colonial period, Chharas (and other DNTs) continue to suffer from ethnic discrimination. Stigmatized as thieves, it is difficult for them to get legitimate jobs in mainstream society. As a last resort, they turn to criminal activity. It is a vicious circle from which only a few are able to escape.

But how do people know they are Chhara? They don’t look noticeably different from the rest of the population, and even if they did, they could easily be from a neighboring state. They speak their own language (Bhantu), but they can speak Gujarati as well as anyone else.

The answer, it turns out, couldn’t be simpler: they ask.

As Roxy put it, the third question when you apply for a job, after “What is your name?” and “Where are you from?”, is “What is your caste?” Technically Chharas are outside of the caste system, but that doesn’t help. They have to answer the question. Even if there are laws against discrimination on the basis of caste, there don’t seem to be any laws against asking such questions. Roxy says it would be even worse for a Chhara if they lied and were found out.

Recently, a courier service opened up near Chharanagar. With a recommendation from a high level government official, several Chhara went and applied for the job. They were turned down. The owner of the local franchise said he would rather close shop than hire a Chhara. Not that he’s racist, you understand, but his couriers handle lots of money and precious items. His clients wouldn’t trust him if he employed Chharas…

So I’m proud to announce the release of Acting Like a Thief! A short documentary film I shot and co-produced with Shashwati, who did an amazing job editing it.