In the mid-18th century, British intellectuals began to use philology, the study of texts and languages, as a tool for understanding human societies. The rise of the East India Company, in particular, inspired new research into the origins of South Asian languages. After studying Sanskrit, Persian, Greek, and Latin, the jurist William Jones helped to popularise the idea that these languages shared a common ancestor.
Over time, however, colonial administrators increasingly turned to philology as an instrument of governance. In North India, British officials divided languages into two broad categories. Those that had written literature and material evidence of their antiquity were classified as languages proper. Anything else was classified as a dialect—from the Greek dialektos, meaning what was ‘spoken’.
Colonial administrators also created a hierarchy of standard and non-standard dialects, associating secretive and coded languages with criminality, illegibility, and deviance. This coincided with a wider initiative to categorise and classify colonial subjects. Following the Indian Rebellion of 1857, as the East India Company gave way to the direct rule of the British Crown, governors grew increasingly suspicious of nomadic groups and communities that were considered ‘hereditary criminals.’ These anxieties were eventually codified into the Criminal Tribes Act of 1871.
By the mid to late 19th century, therefore, discourses about ‘elegant speech’ coexisted with efforts to mark certain forms of communication as suspect. Philology served a carceral purpose, helping the state to map, classify, and control populations. Colonial administrators and native informants, as Christopher Bayly has shown, engaged with dialects and trade languages as part of a growing apparatus of surveillance. From secret trade cyphers to fraudulent shawl-weaving, language was treated not merely as culture, but as evidence that could be used to criminalise whole communities.
The most famous example of this linguistic work is William Henry Sleeman’s Ramaseeana (1836), a lexicon of ‘criminal speech’ compiled during his campaign against the so-called Thuggee. Thuggee was a colonial term for semi-organised bands of robbers accused of deception and ritual murder. They were said to communicate in a hidden language unintelligible to outsiders, and Sleeman’s text helped popularise the idea that criminal networks had their own secret modes of communication.
This work would inspire others to create systematic frameworks of ethnolinguistic governance. The Linguistic Fragments (1870, 1872, 1879) compiled by the orientalist G.W. Leitner included detailed notes on the Magadds, a group operating in northern Punjab. Leitner suspected that the Magadds had originated in Khorasan, in present-day Iran, and travelled between Persia, Afghanistan, and India to sell polished stones. This theory was ingrained in colonial policy, and Magadds were often deported to Khorasan by British officers. These ethnographic speculations, however, were only confirmed after Leitner studied their dialect and understood their internal codes.
Leitner observed that the Magadds spoke in a simple code by adding five to all their numerals:
‘I found that, although they knew both the Persian and the Urdū numerals, those in their own vernacular were formed by the addition of 5, just as the extended hand, with the thumb thrown out, forms the Latin V and its double forms the Latin X.’
This linguistic detail was not a neutral observation. Leitner’s research contributed to the emerging sciences of policing and detection – fields that, as Kapil Raj has shown of modern science more broadly, were not developed exclusively in the West.
Beyond his surveys of the dialects of the Bashegālī Kāfirs, Changars, Dards, and Magadds, Leitner also investigated secret communication systems—what he called ‘native cryptography.’ One such method was the Kam Ṣalā cypher, a coded script used by Persian scholars and elite figures. Leitner considered it impenetrable to munshīs (ordinary scribes), making it useful for covert communication.
The cypher worked by letter substitution, using two lines of poetry to inform the reader which letters to swap and which to leave as they were. The gibberish words kam ṣalā instructed readers to replace m with k, and lā with a sibilant ṣ. Readers could then use the second line to understand which letters were to be used as-is. The use of these systems among scholars, criminals, and travelling communities blurred the lines between erudition and deception.
Leitner’s attention to linguistic detail extended to economic life as well. In one section of the Fragments, he describes the illicit appropriation of Kashmiri shawl patterns by Punjabi weavers. This was a time when North Indian artisans were being integrated into colonial capitalism, as Amanda Lanzillo has shown. Kashmiri weavers (shawlbafs) designed the margins of their shawls with unique patterns of colours and symbols. Specific directions for the warp and weft of the cloth ensured that each generation created an authentic Kashmiri design. Leitner’s analysis of the visual language of shawls became part of an effort to distinguish authentic Kashmiri producers from imitators, helping to police trade and protect origin.
One of Leitner’s most important collaborators was Muhammad Abdul Ghaffur. A former schoolmaster, Ghaffur would rise through the ranks to become a thanadar (low-level police official) and later a darogha (prison supervisor) in Lahore. There, he compiled a Dictionary of the Criminal Tribes of the Punjab (1879) for use by police and prison officials. The dictionary documented dialects, community histories, place-names, and the unique speech patterns of criminalised groups.
According to Ghaffur, each category of thief or confidence artist used its own secretive form of communication, or argot. Many of these dialects had never been documented in writing before. In his introduction, Ghaffur stressed the need for linguistic training among colonial officers, arguing that without such tools, criminal intelligence gathering was incomplete. His work was annotated by Leitner and ultimately incorporated into the 11th volume of George Grierson’s Linguistic Survey of India (1903-1928), a monumental study that codified decades of linguistic research.
Much of this survey was supported by the works of jailers, as well as prisoners themselves. One of these prisoners was Maulana Ja‘ffar Thanesari. Famous for having declared jihād against Queen Victoria, Thanesari had been jailed on the Andaman Islands for his participation in the 1864 Ambala Conspiracy Case. There, he served as an amateur anthropologist, lexicographer, and ethnographer of the local Sentinelese tribes. Thanesari compiled prison vocabularies and corresponded on linguistic matters, all while incarcerated. This work points to a larger pattern of extracting philological labour from imprisoned people.
The colonial interest in ‘secret languages’ became more institutionalised through the 1870s and 1880s. R.C. Temple’s An Examination of the Trade Dialect of the Naqqash (1884), for example, focused on the secret vocabulary of papier-mâché artisans in Kashmir and Punjab. Temple argued that while some words were relics of older languages, others were intentionally inverted or layered with nonsense syllables to obscure meaning. Disguised words like gauẍkha (paper) and nath (place) exemplified how language could serve exclusionary and protective functions in artisan communities.
Temple also examined the dialect of Delhi’s dalāls (brokers), suggesting connections to earlier lexicographical works by 18th-century Persian and Urdu scholars. His question throughout was whether these secret dialects were fossilised remnants of an older language or inventions meant to throw off outsiders. Temple’s analysis linked linguistic shifts to changes in trade, mobility, and urban anonymity.
Glossaries, dictionaries, and regional settlement reports all contributed to a colonial archive where language functioned as both classification and evidence. Carceral spaces such as jails and reformatories became unlikely laboratories for linguistic data-gathering. Philology, a discipline ostensibly devoted to the love of language and textual traditions, was deployed to manage populations, regulate trade, and suppress mobility. It moved in parallel with fingerprinting, anthropometry, and other emergent technologies of the modern colonial state.
By the late 19th century, in other words, philology had become a forensic tool. Figures like Leitner and Temple used language to detect, classify, and sometimes deport, while native collaborators like Ghaffur and Thanesari lent both access and credibility to their efforts. This history challenges any clean separation between the human sciences and the machinery of the colonial state. The study of dialects and argots was never innocent. It was shaped by the needs of governance, animated by suspicion, and grounded in institutions of incarceration. Traces of this carceral philology live on in linguistic archives, waiting to be read.
In today’s world, where accents and vocabulary still shape policing outcomes, and where algorithms scan speech for cues of deviance or threat, the colonial entanglement of language and suspicion feels eerily current. Understanding these 19th-century archives is both a historical task and a political one.