Version classiqueVersion mobile

The Politics of Language Contact in the Himalaya

Selma K. Sonntag
Mark Turin

2. What Happened to the Ahom Language?

The Politics of Language Contact in Assam

Selma K. Sonntag

Texte intégral

The modern nation-state, emanating from post-Westphalia Europe, is frequently characterized by linguistic homogeneity, if not always empirically then at least ideologically. A common language is ideologized as the glue of the nation, the emotional tie needed to foment a sense of national identity rising above parochial proclivities. Furthermore, the modern state has an interest in impelling linguistic homogeneity: ‘linguistic rationalization’ (Laitin 1988) allegedly promotes administrative and economic efficiency and, some argue (e.g., Patten 2001), democratic participation. Even multilingual states, such as India, engage in these processes: in the Indian federal system, states, including Assam, are the site of linguistic rationalization (Laitin 1989) and cultivation of the affective ties that fuel linguistic nationalism (Mitchell 2009; Sonntag 2014).

1The association of a common language with the nation-state, both as a marker of identity and a vehicle of communication, has been avowed by such acclaimed scholars as Benedict Anderson (1991), Ernst Gellner (1983) and Eric Hobsbawm (1990). They attribute the association to print capitalism, industrialization, and nationalism, respectively. These attributions impart a sense of inevitability to linguistic homogenization in the modern nation-state, as if linguistic homogenization is the only outcome of modernization and nation-building. They also shift focus away from the state as the purposeful perpetrator of linguistic rationalization to forces beyond the state’s control. But we know that, despite state attempts at linguistic rationalization, most modern states are not monolingual and that, despite Anderson’s claim that historically as well as currently ‘the bulk of mankind is monoglot’ (1991: 38), a majority of the global population remain polyglots (Meyerhoff and Stanford 2015: 3). Moreover, despite its Herderian origins, the modern nation-state fixes its territorial boundaries, which do not always correspond to more fluid linguistic boundaries. Thongchai Winichakul (1996: 67) has chastised Anderson, Gellner and Hobsbawm and their followers for failing to ‘pa [y] attention to the most obvious constitutive element of a nation-state, namely its territory, as if it were merely a noneffective container of […] essential elements [of the nation].’ Refocusing on the agency of the state, i.e., ‘bringing the state back in’ (Evans et al. 1985) to our analyses of language politics, exposes the historical and geographical contingency of the Western European nation-state model of linguistic homogeneity and debunks its accompanying ideology of linguistic nationalism.

2A historical-institutionalist approach to the study of language politics, as developed by Sonntag and Cardinal (2015), lends itself to analyzing language politics in a variety of geographically and temporally dispersed and diverse states, rather than being limited to the modern nation-state. Central to our theoretical framework are the analytical concepts of state tradition and language regime. We posit that state traditions inform language regimes. State traditions are representational sedimentations of how a state governs. They are reflected in historical patterns of institutionalized practices of governance. For example, federalism represents a state tradition of territorial governance that informs the language regime in India, Canada and the erstwhile Soviet Union. In the United States, liberalism, rather than federalism, has been the dominant tradition informing language policy choices (Sonntag 2019). Language policies compatible with state traditions are institutionalized, comprising a language regime. Language regimes are, then, the institutionalized practices of language governance. Because state traditions inform these institutionalized practices, a language regime has a representational or ideological component as well. As such, language regimes tend to be hegemonic and therefore relatively stable. Although specific language policies within the regime may be tinkered with around the edges, the range of policy options is circumscribed by state tradition. In the terminology of historical institutionalism, language policies are said to be path dependent. Only at critical junctures is a wider range of policy choices considered. Critical junctures can, but do not always, disrupt state traditions, leading to new policy regimes. More often, state traditions endure even though the policy path may shift.

3Using this historical institutionalist framework of state traditions and language regimes, my chapter is a case study of Assamese language politics. I cover three different polities or states — the Ahom kingdom, British India, and independent India — and their associated language regimes, all located in the same complex linguistic environment marked by language contact and diversity. My primary focus, however, will be on the Ahom kingdom. As a precolonial state in the Himalayan border region between what we categorize today as South Asia and Southeast Asia, the Ahom kingdom’s state traditions, and by extension its language regime, were decidedly different from either the colonial state or independent India. What makes the study of the Ahom kingdom intriguing is that, at the height of its power during the seventeenth century, prior to British colonization, the Ahom kingdom shifted from using the Tai-Kadai Ahom language to using the Indo-Aryan Assamese language, suggesting a rupture in the language regime. This shift appears to have been subsequently reified under the colonial language regime, setting the stage for postcolonial linguistic nationalism in Assam. Hence the framing of this chapter’s theme as ‘What happened to the Ahom language?’

  • 1 The metaphor of the mandala invokes a cosmic universe depicted as a circle radiating outwa (...)

4I will proceed on the assumption that the Ahom kingdom was similar to other ‘mandala states’1 in Southeast Asia. The key traditions of the mandala states that I discuss below are their non-territoriality and elite hierarchy. I then sketch out a probable linguistic environment of the Ahom kingdom, speculating on its language regime given its mandala-state traditions. I attempt to identify the critical juncture at which the language shift from Tai Ahom to Assamese occurred. I argue that this critical juncture indexed a change in the language regime informed by incipient changes in the Ahom kingdom’s traditions of non-territoriality and elite hierarchy. I will then discuss how this shift was consolidated under the colonial state and its attendant linguistic ideology. Finally, I will address language politics in post-independence Assam, starting with nationalist representations of the Ahom kingdom and its language regime. My research is based on secondary sources, most of them historiographical and using scant or sketchy records, hence my use above of less-than-analytical terms, such as ‘assumption’, ‘sketch’, ‘attempt’ and ‘speculate’.

The Mandala State

Given the dominance of the Western nation-state model, scholars have grappled with identifying and defining ‘Asian forms of the nation’ (Tønnesson and Antlöv 1996). Historians of Southeast Asia have reached a consensus to some extent on the precolonial state, after sifting through the historiography of the region:

historians have found the state-as-mandala productive in their thinking for two reasons. First, it is an ‘indigenous’ model and therefore ‘protects’ early Southeast Asia from Eurocentric concepts. Actually, of course, it is not indigenous; it is Indic. Second, it is a cultural concept, and its very resistance to pat definition as a model of state formation gives it interpretative power (Reynolds 1995: 427).

James Scott (2009) defines the classical mandala or padi state in opposition to Zomia, the anarchic, swidden-agricultural swathe of highlands across Southeast Asia and Northeast India. Largely missing from his analysis is the flip-side of his dichotomy between the mandala state and Zomia: the mandala state in interaction with a transregional Sanskrit cosmopolis, to use Sheldon Pollock’s (1998a) terms. Both Scott’s and Pollock’s analyses are useful in reconstructing the dynamics of the politics of language contact in Assam.

5One of the most significant characteristics of the mandala state is its non-territoriality. In mandala states, ‘ [t] he political sphere could be mapped only by power relationships, not by territorial integrity’ (Thongchai 1994: 79). Yet scholars of language politics (e.g., Laponce 1987) often identify territoriality as essential to the vitality of a language. Pollock labels the transition from Sanskrit’s transregional cosmopolitanism to vernacularization as ‘literary territorialization’ (Pollock, 1998b: 49), when ‘new notions of geocultural frameworks [were developed] for […] literary narrative representations […] in which […] texts would circulate’ (Pollock 1998a: 28). The lack of territoriality as a political concept in the mandala states — what we can call a state tradition of non-territoriality — suggests a different type of language politics and language regime from that of the modern nation-state.

6Instead of sovereignty being defined territorially in precolonial Southeast Asia, power radiated out of the center, hence the mandala trope. Shifting and overlapping alliances, and multiple, shared sovereignty were the norms (Thongchai 1994: chap. 4). There were no fixed territorial boundaries where one mandala state started and another stopped. And they differed in the range or scope of their mandala. The ‘unit of political order’ (Scott 2009: 430) that indexed the state tradition of non-territoriality was muang, a Tai/Thai term that Thongchai Winichakul (1994: 81) glosses as ‘governed area’ and Charles Keyes (2003: 184) translates as ‘principality’. A muang could be within the mandala of one or more powerful states, often called muang themselves (Keyes 2003: 179), and could have subordinate muang within its mandala orbit. A muang could shift territorially from, for example, one side of the Mekong River to the other but remain the same muang (Miles 2014: chap. 6). Muang could divide and separate, and, less frequently, coalesce and combine. Moreover, ‘a subject of a local authority could be at the same time a subject of another authority’ (Thongchai 1994: 73). At least in northern Thailand, spoken languages were referred to as Kammüang, or language of the muang (Keyes 2003: 184). Because of the shifting nature of muang and their populations, people also ‘shift [ed] their language practices’ (Scott 2009: 39). Languages were no more delineated or objectified than territory was. Language contact, diversity and multilingualism were constants. As Oliver Wolters (1999: 159), the historian usually credited for the mandala trope, puts it: ‘There are probably few more influential cultural features in earlier Southeast Asia than multilingualism.’

7At the center of the mandala states was the king or deva-raj. Kings were considered to be reincarnated divinities legitimizing their centrality and rule in what Victor Lieberman (2011) refers to as river-valley civilizations. The ‘Indianization’ of Southeast Asian kingdoms (Coedès 1967) provided legitimation of hierarchical rule centered around the king. As Coedès (1967: 15–16) defines it:

Indianization must be understood essentially as the expansion of an organized culture that was founded upon the Indian conception of royalty, was characterized by Hinduist or Buddhist cults, the mythology of the Purāṇas, and the observance of the Dharmaśāstras, and expressed itself in the Sanskrit language. It is for this reason that we sometimes speak of ‘Sanskritization’ instead of ‘Indianization’.

Sanskritization in this context refers to the adoption of Sanskrit as the aesthetic, political-cultural, and literary language of the court and the legitimation this provided, and not necessarily to emulation by those lower in the sociopolitical hierarchy, as the term implies today throughout South Asia. Sanskrit was never used as an ‘everyday medium of communication […] [n] or even functioned as a chancery language for bureaucratic or administrative purposes’ in the mandala states (Pollock 1998a: 12). However, its use introduced writing to the mandala states; literacy, or in Pollock’s terms ‘literariness’, provided the trappings of hierarchical rule. According to Pollock (1998a), inscriptions from this period would usually begin with accolades to the king in Sanskrit, then revert to details in the local language (s), written in Sanskritic script. Sanskrit functioned as the expressive, aesthetic language, whereas local languages were for enumerative, constative uses (Pollack 1998a: 12). We can identify Sanskritization or Indianization as the source of the mandala-state tradition of socio-political hierarchy. Along with the state tradition of non-territoriality, it informed a cosmopolitan Sanskritic language regime. With the ‘spread of political Sanskrit’, Southeast Asia became part of what Pollock (1998a: 12) refers to as the ‘Sanskrit cosmopolis’.

8This Sanskrit cosmopolis was ecumenical according to Pollock (1998a, 1998b) — religion per se was not the driver of the spread of Sanskrit; instead Pollock attributes its spread to the aesthetic, universalistic traits of Sanskrit’s hyperglossia and accompanying cultural-political practices. And yet it seems that this hyperglossia began to break down with the spread of Theravada Buddhism from Sri Lanka and its attendant Sanskritized Pali liturgy in the early centuries of the second millennium CE. Nevertheless, consistent with Pollock’s analysis, Steven Collins (2003) suggests that the adoption of Pali Buddhism in mandala states was not necessarily or primarily for religious purposes. According to Collins (2003: 681), Pali Buddhism ‘began to be imported by kings as part of their state-building enterprises […] [Its] ideology operated among the elite, primarily as an element in a nexus of power, as local power-holders were organized by a king into a maṇḍala.’ Pali or Theravada Buddhism ‘ [i] n its sociopolitical aspect […] had to do with naturalizing inequality in social hierarchies’ (Collins 2003: 681), i.e., it was instrumental in reinforcing the hierarchical state tradition at a time when mandala kingdoms, particularly newer ones in Siam and Burma, were coming into their own.

9It also appears that the importation of Pali Buddhism and its liturgy spurred the development of local writing systems, albeit derived from Pali’s Sanskritic script (Collins 2003: 683). Hence, with the adoption of Buddhism by the Siamese king in the thirteenth century, royal inscriptions began to appear in what was to become the Thai script (Keyes 2003: 183; Collins 2003: 684). The Burmese script dates about a century earlier, while other orthographic systems in what is now Burma, such as Mon and Arakanese scripts, appeared even earlier, predating Pali Buddhism to some extent (Collins 2003: 683–84). The change in the literary language regime of the Siamese and Burmese mandala states corresponds to Pollock’s vernacularization, that is, a shift from Sanskrit being the ascetic language of royal accolades and poetry to vernaculars playing this role. The language regimes of the mandala states became more parochial and local — in effect, remaining multilingual and multiscriptal but without the Sanskrit hyperglossia.

10Although the development of local scripts appears to signal a language regime change, there was not necessarily a corresponding political change: mandala states tended to endure up until colonial conquest. In particular, state traditions endured. As suggested above, the state tradition of hierarchy was actually reinforced by the introduction of Pali Buddhism. Thongchai Winichakul (1994) argues that what I am calling the mandala-state tradition of non-territoriality was not disrupted in Siam until colonial powers began to vie for power in Southeast Asia, at which time the Siamese kingdom managed to escape colonization by quickly adopting the technology of mapping and territorial-administrative control. Yet Pollock’s vernacularization rests on an assumption of new ‘geocultural frameworks’ enabling ‘literary territorialization’ (Pollock 1998b: 49). So what were the new geocultural conceptions that marked the change in language regime? Culturally, Theravada Buddhism percolated down the social hierarchy (Collins 2003; Keyes 2003), which hadn’t happened with the Brahmanical Hinduism of the Sanskrit cosmopolis (Coedès 1967: 369). Geographically, the spread of Tai peoples south and west from Yunnan had reinforced the muang concept as the primary ‘unit of political order’ (Scott 2009: 430). These cultural and geographical shifts framed the ‘turn […] to the use of local languages for literary expression in preference to the translocal language [s]’ of Sanskrit and Pali (Pollock 1998a: 6). The ‘local’ was not defined territorially but rather according to its place in the mandala of state power. The shift to the vernacular was a conscious choice, according to Pollock (1998a: 7) — it was a policy option that had opened up at a critical juncture.

11The state traditions of hierarchy and non-territorialization endured at this juncture, during the shift in the literary language regime from Sanskrit hyperglossia to vernacularization. Nevertheless, the new regime enabled ‘literary territorialization’ (Pollock 1998b: 49) which was not equivalent to political territorialization but undoubtedly impacted, and gradually transformed, political-cultural concepts of space. Institutionally, this transformation took the form of ‘a more uniform and efficient tax regime and administration, and […] kingdomwide economic integration and militarization’ (Scott 2009: 253) in the increasingly centralized mandala states on the eve of colonialism. In turn, ‘ [t] he process of valley homogenization’ along the Chao Phraya, Salween and Irrawaddy ‘was much advanced by increasing state centralization between 1600 and 1840’, with ‘the valley states […] busy fabricating more uniform Burmese, Siamese […] and Shan’ (Scott 2009: 253, referencing Lieberman). Zomia remained outside of the mandala states’ hierarchy in Scott’s rendition, not least because of the mandala states’ lack of territorial sovereignty and integrity.

  • 2 Sudipta Kaviraj (2010: 141–43) argues that likewise in Bengal prior to colonialism, there (...)

12However, the mandala states were dependent on Zomia for their agricultural and military manpower needs, which they would obtain by periodically raiding Zomia, often for slaves (Scott 2009; Wolters 1999: 164). The mandala states also battled each other, albeit for control of manpower rather than territory (Wolters 1999: 164). Given the lack of fixity of the states and their subjects (including slaves), there was a great deal of fluidity in identity. Although we need not accept Scott’s emphasis on Zomians-versus-the-state as the most relevant distinction, there is a general consensus that neither language nor ethnicity were salient identity markers in the mandala state. As Scott (2009: 253) puts it, ‘ [a] certain plasticity of identity was built into precolonial power relations.’ While the elite literary component of the language regimes of mandala states was initially characterized by Sanskrit hyperglossia, followed by literary vernacularization, the ‘constant movement back and forth between the valleys and the hills’ (Scott 2009: 27) meant that quotidian language practices were undoubtedly multilingual and involved constant language contact, often between unrelated languages. Despite valley homogenization — particularly in terms of governmentality — on the eve of colonialism, most scholars concur that it is indeed difficult to say that there was any concept of ethnolinguistic identity, as we now conceive of it, in the mandala state (Scott 2009: 253).2

The Ahom Kingdom

Hence the difficulty of identifying the Ahom. Yasmin Saikia (2004: 252) insists that the Ahom are not an ethnic group, despite an ethnolinguistic dimension to their current-day efforts at mobilization. More probably, they initially were identifiable as warriors migrating with a Tai king from the east, and constituted the military-bureaucratic class once the Ahom kingdom was established in the upper Brahmaputra Valley in the thirteenth century (Saikia 2004: chap. 3). This ‘class’ of ‘nobles’ (Saikia 2004: 20, 126, 252) shared power with the king, or swargadeo, as ‘Ahoms’ (Guha 1983: 19–20). According to Saikia (2004: 133), these initial Ahoms, i.e., the royalty and the warrior-administrators, were all men and ‘espoused local women’. Amalendu Guha (1983: 39), who more easily uses Ahom as an identitarian category, claims that the initial Ahoms separated from a mung or muang in upper Burma. Guha (1983: 12, 32) glosses mung as ‘political society’; accordingly ‘the Ahom polity started as a loose confederacy of several mungs around a dominant one of the Tai model.’

13It appears, then, that the establishment of the Ahom kingdom in the early thirteenth century was based on mandala-state traditions. The muang geocultural framework and a sociopolitical hierarchy with a noble class were its foundation. Moreover, it was founded in a complex, multilingual environment, which was neither ‘politically void’ (Guha 1983: 12) nor literately void (Gohain 1999: 101), at the critical juncture when mandala states were vernacularizing. As I argue below, this put the Ahom language regime on a different policy path, setting the stage for the replacement of the Tai-Kadai Ahom language with Indo-Aryan Assamese at another critical juncture: when the kingdom’s traditions of non-territoriality and elite hierarchy were changing, although not completely disrupted. I will first reconstruct the Ahom language regime by elucidating how the Ahom kingdom’s mandala-state traditions informed it, then analyze its shift to Assamese.

14Like other mandala states, the Ahom kingdom was not dependent on territorial control, but rather control of manpower for wet-rice-agricultural as well as military purposes (Guha 1983: 34; Saikia 2004: 128–29). Accordingly, there was no land tax imposed in the kingdom, only a labor, and at times military, service requirement (Bhattacharya 2005: 21; see also Gurung 2018: 199, n14; Parwez 2018: 133). Indeed, the institutionalization of manpower in ‘the Ahom peasant-militia’ system of corvée labor was deemed ‘the most important component of the Ahom political system’ by Edward Gait, a colonial authority on the Ahom (quoted in Saikia 2008: 154). The nobility was responsible for acquiring the manpower, usually by absorbing and/or enslaving the diverse population, often through military ventures. These populations could be Zomians: Guha’s (1983) account has been characterized by Nayanjot Lahiri (1984: 60) as ‘the absorption of stateless shifting cultivators into [the padi Ahom] polity’ (see also Saikia 2004: 155). Manpower could also be secured from other muang — usually militarily although without necessarily absorbing the muang themselves, as was typical of mandala states. Some of the earliest secured, if not absorbed, were the Tibeto-Burman-speaking Barahis and Morans (Guha 1983: 15; Saikia 2004: 2), whose women the initial Ahom espoused (Saikia 2004: 32). The acquired manpower was organized at its base into paik, which consisted of three to four individual laborers (or militiamen) who then shared rotation in labor units known as khel (Saikia 2004: 126–27; Guha 1983: 8–9; Terweil 1983: 58). Khel were further aggregated into faid or ‘service groups’ (Saikia 2004: 126). Neither khel nor faid had any territorial base. Nor did they necessarily contain paik from the same locale or speaking the same language: ‘To inhibit khel solidarity, paiks were constantly moved from place to place and between khels’ (Saikia 2004: 127). This system undoubtedly both reflected and reinforced an extremely diverse multilingual environment.

15Khel were under the control of the Ahom nobility. The structure of the nobility was hierarchical, with the rungs expanding and growing more numerous over time. At the apex initially were two great gohain (Terweil 1983: 21), followed by lesser gohain in the hierarchical ranking. The gohain ‘control of the labor force […] made these nobles very powerful, often more powerful than the [royal] princes’ (Saikia 2004: 126). Their relation with the Ahom royalty cohered to mandala-state traditions: according to Barend Terweil (1983: 54), gohain is the Assamese term for Ahom nobility; in Ahom Tai, the titles of these high-ranking gohain contained the word mung, the Ahom cognate of the Thai muang — reflecting the Tai model of political society as a loose confederation, as identified by Guha (1983) quoted above. Also quite powerful, and more numerous, were the phukan and barua, nobles who were responsible for large khel which served in both wet-rice agriculture production and military ventures (Saikia 2004: 126; Guha 1983: 8). Lower ranks, such as the saikia, also constituted the Ahom, or nobility class. Beyond the initial warrior migrants, the Ahom were drawn from earlier populations and autochthonous groups, and possibly from earlier as well as more recent Tai migrations from upper Burma (Guha 1983; Gohain 1999: 102; Saikia 2004: 27). Saikia (2004: 286) cites a source claiming that the third great gohain, created in the late fifteenth century, was initially filled by a Naga (see also Baruah 1999: 32). In Scott’s (2009) terminology, Nagas would be considered Zomians; that they could be enlisted as both paik and the highest-ranking Ahom demonstrates the political-cultural, and by extension linguistic, fluidity of the Ahom kingdom.

  • 3 In a personal conversation (29 October 2018, Dibrughar University, Assam) I had with Jahna (...)

16Just as fluid was the literary environment of the Ahom kingdom. The Ahom mandala state most probably was part of, or at least influenced by, the Sanskrit cosmopolis (see Saikia 2004: 118). At a minimum, the kingdom was sandwiched by it — between Southeast Asia and ancient Kamarupa in the lower Brahmaputra Valley, where the spoken and written linguistic heritage was Indo-Aryan (Saikia 2008: 160–64), dating from the fifth century onward (Gohain 1999: 101). According to Guha (1983: 12), the ‘political heritage of ancient Kamarupa had not left upper Assam totally untouched.’ Other lower Brahmaputra kingdoms, such as the Kachari, Koch and Jaintia, had developed writing systems as well (Saikia 2004: 123). Pollock dates Assamese vernacularization from the mid-fourteenth century when the Ramayana was first composed in Assamese ‘at the request of the Barāhi king’. Although the Barahi king to whom Pollock attributes vernacular Assamese patronage appeared to be further down the valley than the Tibeto-Burman-speaking Barahi with whom the Ahom initially interacted (see above), Pollock’s dating suggests that not long after their arrival in upper Assam from Southeast Asia, the Ahom came into contact with vernacularizing trends originating in the South Asian, as opposed to the Southeast Asian, Sanskrit cosmopolis.3

  • 4 This assessment that the Ahom script was similar to the Mon script was also expressed by G (...)

17It appears that the Ahom brought a writing system with them in the early thirteenth century, possibly pre-dating the emergence of the Thai script in Siam, but post-dating vernacularization in the mandala states in Burma. B. K. Gohain (1999: 101), referencing scholars such as Terweil, claims that the script of the Ahom kingdom was ‘specific only to Ahom’ and that it was ‘derived from old Mon as it was written in the first centuries of the second millennium A.D.’ — the Mon kingdom being one of the first vernacularizing mandala states among the Southeast Asian mandala states, as noted above.4 Other scholars, such as Guha (1983), refer to the Ahom writing system as ‘Tai script’. Ahom literariness took shape in the buranji, which were primarily court chronicles celebrating and delineating the genealogy and exploits of the kingdom’s rulers. The buranji were under the keep of the deodhai, or priests, of the Ahom kingdom. But there is little, if any, historical evidence that the deodhai were Buddhists (or Hindus). The critical juncture in literary vernacularization marked by the introduction of Pali Buddhism in other mandala states, as discussed above, was apparently absent in the Ahom case. Or at least the religious component was absent, but perhaps not the literary marker: Gait, the British colonial ‘expert’ on the Ahom kingdom, claimed that the Ahom language was ‘written in a character derived from Pali’ (quoted in Saikia 2004: 285). If indeed the Ahom script was based on the Mon script, which, as discussed above, pre-dated fullscale adoption of Pali Buddhism, then this dates the establishment of the Ahom kingdom on the cusp of literary vernacularization in Southeast Asia spurred on by Pali Buddhism.

  • 5 Ahmed (2008), unfortunately, does not specify to which inscriptions he is referring.

18At least initially, then, the written language regime of the Ahom kingdom was neither the cosmopolitan Sanskrit language regime nor a fully vernacularized one in terms of literary territorialization, but rather a very fluid, multiscriptal language regime. While Sarharuddin Ahmed (2008: 22) claims that ‘ [t] he inscriptions of the Mediaeval Assam (Āhom period) are written partly in Sanskrit and partly in local Tai language’,5 indicative of the language regime of the Sanskrit cosmopolis, the Ahom buranji, which are more literary and included origin myths as well as royal chronologies and military exploits, ‘were never written in Sanskrit but rather in Assamese, and occasionally in a Tai language that is sometimes referred to as Ahom’ (Saikia 2004: 123). The Ahom language regime was also likely characterized by fluid multilingualism. With literacy limited to the very few, primarily the priestly and royal/nobility classes in mandala states, ‘there was no pressure to impose linguistic uniformity on the peoples’ of these ‘premodern’ mandala polities (Keyes 2003: 186). We can assume that the spoken means of communication in the Ahom kingdom comprised a variety of languages, including pidgins (Baruah 1999: 31). As Terweil (1983: 44) has put it, ‘ [t] he Ahom developed sophisticated communication systems and organizational hierarchies […] Generally they succeeded in keeping a tight hold over a large populace of great diversity.’ Linguistic contact between Tai-Kadai, Indo-Aryan, Tibeto-Burman and Austro-Asiatic ‘language families’ must have been constant in speech interactions among and between the different ‘classes’ and muang in the Ahom kingdom. The language regime of the Ahom kingdom, at least up until the seventeenth century, is therefore probably best characterized as institutionalized multilingual and multiscript/multiliterate contact.

19In the early sixteenth century, the sphere of the Ahom mandala state began expanding through military ventures to include the older Chutia kingdom to the northeast. The recently Hinduized Chutia, who may have been either Tibeto-Burman or Tai-Kadai speakers or both, had supposedly developed ‘a Tai script’ (Saikia 2004: 6) — another example of the very complex environment of language contact throughout upper Assam. Guha (1983: 27) emphasizes the resulting influence of Hinduism on the Ahom kingdom from not only the Chutia but also from further Ahom expansion westward: ‘By 1539, the Ahom territory became at least twice as big as what it was in size around 1407. More important, its Assamese-speaking Hindu subjects were now more numerous than the Ahoms themselves.’ Soon thereafter, in 1562, according to Guha (1983), the Koch invaded from the west, having already absorbed Assamese-speaking Hindu Kamarupa, only to have the Ahom reverse their fortunes by the mid-seventeenth century. Also by the mid-seventeenth century, the Ahom had finally managed to stave off repeated Mughal military incursions. Hence, ‘ [a] t its peak in the seventeenth century the Ahom kingdom stretched from Sadiya [the earlier Chutia capital] in the east to Guwahati in the west’ (Saikia 2004: 8). As Jahnabi Gogoi Nath (2002: 30) phrases it, the ‘small [Ahom] state formed in the extreme south-east corner of the Brahmaputra valley in the early part of the 13th century became the single largest state of Assam covering almost the entire region of the valley by the last part of the 17th century.’

20Evidence of territory becoming a political category in the wake of successful Ahom military expansion can be gleaned by comparing the Ahom kingdom’s early 1500s census with census categories of the mid-seventeenth century. The early sixteenth century census focused on manpower, i.e., counting the adult male population, with no territorial component, according to Guha (1983: 34). In comparison, the mid-seventeenth-century census was much more complex, although still oriented toward addressing manpower needs. Its purpose was ‘to facilitate the swargadeo’s [king’s] mission to become the most powerful king of the region’ by expanding and rationalizing the administrative nobility’s functions through new regulations on land division for wet-rice agriculture (Saikia 2004: 127–28). Guha (1983: 34, 32) also notes that there was a ‘total absence of any land survey and measurement in the Ahom Kingdom until the end of the 16th century’ and that ‘land surveyors’ (as well as scribes) were brought into the kingdom from the Mughal empire in the early seventeenth century. By 1681, the Ahom had ‘started [a] countrywide detailed land survey’ (Parwez 2018: 133).

21The mid-seventeenth century can be taken as the critical juncture for the Ahom kingdom, with its military victories, particularly over the Mughals, and its ensuing expansion up and down the Brahmaputra Valley disrupting the mandala-state tradition of non-territoriality. It was also a period of substantial political and economic reform (Guha 1983). Terweil (1983: 43–44) claims that with the defeat of the Mughals, ‘a new, invigorated Ahom rule was established, ready to try new methods of administration.’ A significant example of these new administrative methods was the adoption of the Mughal system of territorial-administrative control — the pargana system of collecting land revenue — at least in the newly acquired Ahom territories in lower Assam (Goswami 1986: 16–18; Parwez 2018: 132; Gogoi Nath 2002: 46–47). Not only did new administrative policy options and pathways open up at this critical juncture, the language regime also shifted. Terweil (1983: 44) continues: ‘From this time onward the Ahom were firmly set on the path towards full assimilation of Assamese Hindu culture, and the Ahom tongue became obsolete. Assamese script took over from the old Ahom characters.’

22Guha (1983) also dates the culmination of the switch from the Tai Ahom language to Assamese to the mid-seventeenth century. As Guha (1983: 10) puts it: ‘the literate Ahoms retained the Tai language and script well until the end of the 17th century […] In this century, this [Tai Ahom] language first co-existed with and then was progressively replaced by Assamese at and outside the [Ahom] court.’ Saikia (2004: 120) points to the seventeenth century as well, when ‘several Brahmin families were invited to take high positions in the swargadeo’s [Ahom king’s] administration’, testifying to what Guha (1983: 25–30) claims as nearly a century of increasing Brahmanical influence. The literary impact of this evolving state tradition of expanding the ranks of the Ahom nobility can be ascertained from Saikia’s (2004: 121) assertion that ‘ [a] close reading of the [Ahom kingdom’s] chronicles to evaluate their prose style, orthography and language, narrative structure, and stories indicates that buranji writing developed and took off in the late seventeenth century’ — and that these buranji were written in Assamese (Saikia 2004: 123).

  • 6 For comparison’s sake, according to Carmen Brandt (2015), in the early eighteenth century, (...)

23It was not only through the expansion of the elite that Assamese was replacing the Ahom language. Neo-Vaishnavism with its bhakti and anti-Brahmanical practices was spreading rapidly eastward, up the valley, bringing with it poetry and prose in Assamese to both the literate and illiterate (Guha 1983: 30–31, 44). Although Pollock (1998a: 29) argues that bhakti was not a primary ‘dynamic in the history of South Asian vernacularization’, its spread corresponds to what appears to be political and cultural changes in the Ahom kingdom, and to parallel the role of Pali Buddhism in the vernacularization of Southeast Asian mandala states. The Ahom kingdom had converted to Hinduism by the mid-seventeenth century, but it wasn’t until the late seventeenth century that neo-Vaishnavism received official Ahom royal patronage (Guha 1983: 30–32). This patronage — presumably, in linguistic terms, of ‘vernacularizing’ Assamese — may have been more significant than neo-Vaishnavism itself (Pollock 1998a: 31).6

24Equally important as linguistic patronage was the political-economic patronage given to the neo-Vaishnavites. The Ahom kings gave land grants to the neo-Vaishnavite leaders, which dramatically increased from the mid-seventeenth century (Gogoi Nath 2002: 38) — mostly on the huge riverine island of Majuli in the Brahmaputra. These land grants were not subject to the new land taxation system that the Ahom state had adopted from the Mughals. Along with these land grants, paik were allocated to the neo-Vaishnavite satra (monastaries) (Gurung 2018: 101). The satra may have also provided a ‘zone of refuge’, in Scott’s (2009: 22–26) sense of the term, for the paik: according to Tejimala Gurung (2018: 113) a ‘large number of paiks […] escape [d] from periodic paik duty and settle [d] down at satra lands.’ Another option to those seeking to escape paik service was also emerging: ‘in the Ahom territories the relatively affluent paiks wanted, frequently, to commute their service obligation into cash or kind payments’ (Parwez 2018: 133). By the early eighteenth century, ‘commutation of paik services for money was encouraged by the Ahom state which needed money for payment of wages to the soldiers’ of the increasingly mercenary Ahom army (Gurung 2018: 124, n77). Edward Gait (2015: 184, n1) cites figures for the number dead in a battle in which the Ahom defeated the Jaintia kingdom at the beginning of the eighteenth century: of the 2336 men lost on the Ahom side, approximately 60 % were from lower Assam, and hence presumably Assamese speakers, compared to only 40 % from upper Assam. Although Gait does not provide information on this, we can speculate that the foot-soldiers from upper Assam were more likely paik rendering military service than those from lower Assam, where territorial-administrative control through land taxation had been adopted.

25In the end, the Ahom kingdom was brought down by what Guha (1983: 38) calls ‘peasant revolts under a [neo-Vaishnavism] religious garb’, which ‘became endemic’ after 1770 (see also Purkayastha 2008: 193). The revolts were at least partly in response to ‘ [t] he Ahom state […] increasing the duration of service a paik had to render’ to offset the reduction of manpower resources that had resulted from the increasing demand for commutation of paik services (Gurung 2018: 101). Also, the ‘grant of large numbers of paiks [in the mid-eighteenth century to religious institutions such as satras] ultimately drained the state treasury in the form of loss of man-power-revenue’ (Gogoi Nath 2002: 40). This was a period of dramatic riverine changes as well (Cederlöf 2014: 23), undoubtedly contributing to the revolts and the kingdom’s manpower crisis. The implication in analyses of the decline of the Ahom kingdom is that, while a more territorial notion of political control was emerging, control over manpower remained a significant resource for exercising political authority. The Burmese invaded the weakened Ahom kingdom in the early nineteenth century, after the British attempted to help the Ahom king to get the rebellions under control (Cederlöf 2014: 4). In the subsequent Anglo-Burmese war of 1825, the British took control of the Ahom kingdom, first establishing indirect rule, then direct rule in 1838 (Saikia 2004: 97, 101).

  • 7 The linguistic effects of language contact is apparent today, according to Chelliah and Le (...)

26Assuming that the above narrative is more or less accurate, what kind of conclusions can we draw? Using the framework of state traditions and language regimes introduced above, we can postulate that the Ahom kingdom’s mandala-state traditions most likely informed a heavily multilingual and multiscriptal language regime. The critical juncture, crowning language regime change, appears to have been when the Ahom kingdom reached its height in the mid-seventeenth century, after nearly a century of expansion that increasingly took on a territorial dimension and accelerated a ‘liberal policy of offering respectable official positions to new entrants’ (Purkayastha 2008: 180). These newest entrants were in effect ‘a new social class related to state power’ (Purkayastha 2008: 180). They legitimized their new status by having their own buranji written in Assamese (Purkayastha 2008: 180; Saikia 2004: 121, 140). This Assamese buranji writing had an expanded ‘circle of authors and readers’, given the increased extent of the Ahom kingdom (Purkayastha 2008: 180). Adding in the increasing political and linguistic influence of neo-Vaishnavism, it seems, then, at the height of state power, the language of literature, and probably spoken language as well, shifted. The new language regime was informed by state traditions that were morphing into more of a territorial notion of the kingdom and a flatter hierarchy through expansion of the elite. Although these emerging state traditions seem closer to those of the modern state, the language regime they informed remained different. Despite the shift to Assamese, the path dependency of multilingualism through language contact most likely remained.7

The Colonial State

By all accounts, the Ahom kingdom’s shift from a Tai-Kadai language to Indo-Aryan Assamese predates colonialism. While new ‘geocultural’ concepts may well have underlain the ‘literary territorialization’ of Assamese (Pollack 1998b: 49), and the Ahom kingdom’s mandala-state tradition of non-territoriality appeared to be changing, the political concepts of fixed borders and territorial sovereignty were a colonial introduction. ‘Valley homogenization’, including some degree of linguistic homogenization, probably also preceded colonial conquest (Scott 2009: 253). However, the linguistic ideology of a ‘language’ being a discrete, identifiable object ‘belonging’ to a particular people in a fixed territory was a colonial imposition. There was a fundamental congruence between ‘colonial geography’ (Baruah 1999: chap. 2) and the colonial language regime. Colonial geography spawned a ‘new property regime’ informed by the ‘Orientalist view that […] emphasized the discreteness of each village’ (Baruah 1999: 48), just as the colonial language regime was based on the ‘dogma that those who speak a particular language form a unique, definable unit and that this unit had a particular culture and a particular history’ (Scott 2009: 239, quoting Leach 1954: 48). British colonial rule ‘provide [d] a radically new representation of the relation of the speaker to his speech (one language, one name, one identity)’ (Montaut 2005: 81). Under colonial rule, the recognition and status of Assamese as the language of the erstwhile Ahom kingdom was consolidated, and extended to the colonial political-geographical construct of Assam and ethnolinguistic construct of the Assamese people.

27Integral to this consolidation and extension was the colonial patronage of written texts (Mitchell 2009). Colonial patronage differed from the previous royal patronage of what Pollock (1998a) would call aesthetic texts, such as poetry or royal accolades. Because the new colonial rulers assumed that texts were examples of spoken language (Errington 2008: 58), they patronized the production of grammars and dictionaries of Indian vernaculars, often undertaken by missionaries, spurring standardization. Saikia (2004: 6) notes that ‘the written language of Assamese was […] standardized’ by the American Baptist Mission printing a translation of the bible in 1835, ‘merg [ing] into one’ the various scripts used at the time. Written standardization of Assamese in effect reduced, if not eliminated, the multiscriptal characteristic of the previous Ahom language regime. It also undoubtedly accelerated homogenization of spoken Assamese, a process that began, according to Guha (1983: 44), with the ‘neo-Vaishnavite missionaries’. Standardization reflected and reinforced the colonial linguistic ideology that ascribes a discrete, identifiable language as the marker of a discrete, identifiable people or ethnicity. It is the key component of linguistic rationalization (Laitin 1988). The colonial state’s language regime was founded on the enumeration and delineation, and then standardization, of indigenous languages.

28Philology furnished the ‘scientific’ vindication of the colonial linguistic ideology (Errington 2008: chap. 4). Through the comparison of the ‘languages’ in written texts, relations — both spatial and temporal — between languages could be established, according to the philological approach. Since languages mapped onto ethnicities/ peoples in the colonial language regime, philology provided the groundwork for establishing historical relations between peoples and ethnicities (including races). Those relations were perceived to be hierarchical. Although philologically Sanskrit was related to Greek, Latin and Persian, as Sir William Jones ‘discovered’, and had spawned ‘daughter’ or descendent languages such as Assamese (albeit several nodes down the language family tree), the spatial and temporal distance between Sanskrit and its European sisters indicated to colonialists that Sanskrit and by extension India had fallen into decay (Errington 2008: 56 ff). It was a relatively short step from Jones’ ‘discovery’ in 1786 to Macaulay’s declaration in 1835 that ‘a single shelf of a good European library was worth the whole native literature of India’. Although the Orientalists, such as Jones, and the Anglicists, such as Macaulay, disagreed over language policy choices for Britain’s colonial possession (Sonntag 2011), the colonial state’s hierarchical and racist tradition reinforced by colonial linguistic ideology informed the language regime of the colonial state.

29The ‘history’ of the Tai Ahom language and the Ahom kingdom’s shift to Assamese challenged some of the tenets of colonial linguistic ideology. The language shift seemed to indicate that the Ahom ‘conquerors’ did not impose their language on their subjects. This did not quite fit with the biologically-based racist ideology of the inherent superiority of conquerors over the conquered that informed colonial language regimes (Errington 2008: 65, 87). As mentioned above, one of the earliest and best known colonial officials to study Assam, and hence the Ahom kingdom, was Edward Gait in the late nineteenth century. According to Arupjyoti Saikia (2008: 152),

Gait proclaimed that philology could not be a ‘real test of race’ […] [because of] numerous examples where one language had supplanted another, or where conquerors adopted the language of the vanquished. Here he provided the example of the Ahom, who abandoned their ‘tribal dialect’ in favor of Assamese.

Gait’s ‘texts’ for his history of Assam were primarily buranji which, as noted above, were written mainly in Assamese. Buranji were somewhat unusual in that, unlike vernacular texts in many other parts of India, they could be ‘read’ not only philologically but also as historical texts according to the colonial regime of history (Chatterjee 2008: 14). These local circumstances of Gait’s inquiry may have led him to discount philology as providing the ‘scientific’ evidence of racial inferiority in the case of the Ahom and, by extension, the Assamese. But for Gait, Assamese was undoubtedly a more developed language, with its Sanskrit derivation, than the ‘tribal dialect’ of the Ahom. Hence linguistic superiority — a strong ideological tenet of the colonial language regime — was nevertheless reinforced in the Assamese case albeit not linked to race. The colonial language regime rendered the Ahom language as ‘dead’ (Saikia 2004: 80), fitting in with philology’s biological metaphors (Errington 2008). The demoting of the Ahom language vis-à-vis Assamese in the colonial regime facilitated the demotion of Ahoms and, by extension, the Assamese as a race in the colonial mentality, despite Gait’s reservations.

30Further reinforcement of the colonial mapping of racial and ethnic hierarchy was accomplished through a new colonial territorial-administrative ‘hard boundary’ between hill peoples and supposedly more ‘advanced’ valley peoples (Baruah 1999: 29; Saikia 2004: 57–58). This new colonial territorial regime disrupted the constant exchange between river valley padi states and Zomia that had been a defining feature of mandala kingdoms, including the Ahom kingdom (Scott 2009). This disruption facilitated the dismantling of precolonial economic relations and the introduction of a tea plantation economy. For example, slave-hunting was prohibited by the British in the 1830s, for the alleged protection of hill tribes/Zomians, and then slavery itself was abolished in 1860 (Guha 1977: 3, 10). According to Guha (1977: 10–11), ‘ [t] he abolition of slavery almost crippled the old Ahom aristocracy.’ The British took over the sale of opium, which had become an important product for trade and consumption during the decline of the Ahom kingdom. Poppy cultivation was banned at the behest of tea plantation owners in order to coerce the Assamese peasantry into laboring on the plantations (Guha 1977: 6, 9–10, 19). A new land tenure system favored the tea plantations and agriculturally marginalized the peasantry (Baruah 2005: chap. 4). The tea plantations and the colonial administration in Assam tended to employ Bengalis, rather than Assamese, in managerial positions (Saikia 2004: 102). In this new colonial hierarchy, the Ahom, now defined as Assamese — and concomitantly the Assamese language — were ranked below the Bengalis and the Bengali language (although above the tribals in the highlands, i.e., Scott’s Zomia). The colonial language regime, informed by Western European state traditions, set the stage for the demotic politics of the modern state, which in Assam have taken the form of intense linguistic nationalism.

The Modern State

Language politics have featured prominently in independent India. A significant component of India’s postcolonial language regime has been linguistic federalism, institutionalized by the 1956 States Reorganization Act: major Indian languages, including Assamese, form the basis of most states in India’s federal union. The vitality of India’s major regional languages seems assured with this linguistic territorialization, while minority languages within the linguistic states remain for the most part unprotected. Linguistic rationalization, i.e., the institutionalized linguistic homogenization of the nation-state model, happens within India’s linguistic states (Laitin 1989) — in effect, ‘ [l] inguistic federalism has shifted the politicization of language downward, to the state level’ (Sonntag 2014: 96). Major Indian languages are also enhanced through another component of the language regime: the Eighth Schedule of the Indian constitution confers the status of national language on an original fourteen languages (including Assamese), now twenty-two (Sarangi 2015). These two components of modern India’s language regime provide it with the ‘pan-Indian cultural grammar of the nation-province’, to use Baruah’s (1999: 113) terminology. Language politics play out primarily in the linguistic nation-provinces or states within India. Linguistic nationalism is the ideological fuel for language politics within the Indian states; multiple ‘subnationalisms’ (Baruah’s terminology again) make up India writ large. Hence, despite state-level linguistic rationalization and nationalism, India can be described as having a multilingual, and multiscriptal, language regime.

  • 8 The view that Assamese is a dialect of Bengali still finds expression in such prominent pl (...)

31The intensity of language politics in various states of India since independence has been attributed to colonialism: ‘ [l] inguistic consciousness […] seemed to have stemmed from the classificatory passion of the colonial agenda […] providing the grounds for a distinctive language consciousness later on to develop into language claims and conflicts’ (Montaut 2005: 87–88). The raising of a distinctive language consciousness in Assam began soon after the British annexed Assam in 1826. The British had made Bengali the official language in 1837 as they were transitioning to direct rule. According to Uddipan Dutta (2016: 68), ‘ [w] hile this was an administrative decision, it created a situation where Assamese was regarded as a dialect of Bengali.’8 The colonialists also ruled, or administered, Assam from Bengal, not constituting it as a separate colonial province until 1874 (Baruah 1999: 24), a year after they partially restored Assamese for official use (Saikia 2004: 60; Guha 1977: 22). This colonial rendering of Assam as a territorial and linguistic appendage of Bengal sparked Assamese intellectuals, many of them residing in Calcutta and familiar with the Bengal renaissance, to ‘make the case that they [the Assamese] were a distinct people with a distinct language and culture’ (Baruah 1999: 71). Their case against the British colonial language policy in Assam was paradoxically based on the colonial — and modern — linguistic ideology of discrete peoples having their own discrete language, and their demands therefore resonated with the language regime (Sonntag and Cardinal 2015: 8). It was consequently unsurprising when they allied with missionaries complicit with the colonial regime who promoted the Assamese language as the distinctive feature of Assam, for the purpose of disseminating Christian scriptures to the local population in their own language (Baruah 1999: 71; Saikia 2004: 60). Ironically, the tea industrialists helped cultivate Assam’s territorial distinctiveness from Bengal, given the increasing brand recognition of Assamese tea (Baruah 1999: 27). With the association between the Assamese language and the regional territory solidifying, ‘ [l] anguage politics became the channel for new demands’ for the Assamese (Saikia 2004: 60).

32By the early twentieth century, Assamese intellectuals were wedding their anti-colonialist discourse to the nationalist cause. While some nationalists identified ancient Kamarupa in the lower Brahmaputra Valley as the source of pride in Assamese, others, such as the Assamese historian Surya Kumar Bhuyan, attempted to bond the (Indo-Aryan) Assamese with the (Tai-Kadai) Ahom kingdom in order to authenticate a distinctive ethnolinguistic consciousness, especially distinct from Bengali (Saikia 2008: 161; Purkayastha 2008: 196). The singularity of Assamese consciousness was crucial: ‘Bhuyan’s portrayal of a generic Assamese society evolving within the Ahom state trie [d] to assure homogeneity by playing down the legacy of a composite culture’ (Purkayastha 2008: 194). Bhuyan ‘argu [ed] for a language-based nationalism for Assam’, with language being ‘identified as a central feature defining Assamese culture’, for which he ‘used the buranji texts as a powerful weapon […] to assert an Assamese linguistic identity’ (Purkayastha 2008: 196). Furthermore, Bhuyan made the vital territorial linkage: ‘The name (Asam = Ahom) was a symbol of the territorial identity of modern Assam’ for him (Purkayastha 2008: 195).

33The language-territory-people association, reified under the colonial regime, now resonated in the struggle for independence and beyond. Bhuyan had made the territorial linkage, but in Assam linguistic homogenization remained precarious despite Bhuyan’s and others’ attempts. Sylhet, which had been considered ‘as part of the mostly Bengali-speaking Surma Valley as opposed to the mostly Assamese-speaking Brahmaputra Valley’ by colonial authorities was hived off from Assam at Partition, when its population voted by referendum to join (East) Pakistan rather than remain in India (Baruah 1999: 101). However, Cachar, also in the Surma Valley, remained part of Assam. Nagaland was formed in 1963 in a rather unsuccessful attempt to contain an insurgency. Further reorganization of Assam, which had linguistic implications, was postponed until the early 1970s, when Meghalaya, Mizoram and what became Arunachal Pradesh were detached. Even then, linguistic homogenization within Assam remained incomplete: according to the 1991 census, only about 58 % of the population claimed Assamese as their first language (Baruah 1999: 19). By the 2011 census, that figure had dropped to 48 % (‘Assamese Language Under Threat’ 2018). The Assam state government’s inability to carry out linguistic rationalization has been at the base of much of the conflict, often violent, in recent decades.

34Assam’s claims to legitimately constitute a linguistic state in modern India’s language regime has been further compromised by the emergence of a Tai-Ahom identity movement in recent decades. As Saikia (2004: 11) puts it, ‘the assertion of Ahoms as separate from the Assamese is a problematic sign of the internal breakdown of the composite Assamese identity.’ Some self-identifying Ahoms are attempting language revitalization of Tai-Ahom, claiming Thai spoken in Thailand is a related language and seeking assistance from Thai scholars (Saikia 2004: 185–87, 216–21). In this regard, they are downplaying language contact and the multilingualism it has entailed in Assam in their reconstruction of Ahom historical experience, opting for a philological basis to their new, distinct linguistic identity in Assam. The movement also identifies a territorial component to Tai-Ahom identity: the upper Brahmaputra Valley (Saikia 2004: 232). Interestingly, however, one of the main organizations of the movement, the Ban Ok Publik Muang Tai (translated as the Tai-Ahom Sahitya Sabha in Assamese or Eastern Tai Literary Society), whose ‘main agenda [i] s to produce a memory of the Ahom as an ethnic group’ through a reinterpretation of the historical records (Saikia 2004: xv, 180), retains in its name the mandala geocultural concept of muang. The likelihood of success of the movement is minimal, especially compared to the better-known Bodo movement in Assam. In 2003, Bodos attained territorial recognition through the establishment of a territorial council under their governance; the Bodo are now demanding a separate state in India’s federal union. The Bodo language also received recognition in 2003, through its addition to the Eighth Schedule (Sarangi 2015), even though there are, unsurprisingly, many Bodo languages and many Bodo have assimilated linguistically to Assamese (Baruah 1999: 180–83). In contrast, Tai-Ahom activists’ ethnolinguistic demands have not gained the same amount of traction in India’s current language regime. As suggested throughout this chapter, this may be because by invoking the Ahom kingdom as the historical basis of their ethnolinguistic identity, the activists are also unintentionally invoking the legacy of a multilingual, multiscriptal language regime informed by mandala-state traditions.


Colonial and postcolonial language regimes are palimpsests, imposed not on blank slates but in complex political and cultural environments marked by historically dynamic language contact. While language contact has been recognized as a prominent and ongoing feature of the linguistic environment in South Asia (Emeneau 1956), and certainly characterized precolonial language regimes, it was not institutionalized in India’s colonial or postcolonial language regime. In contrast to these latter language regimes, conceptions of language and language use as language contact, without discrete associations with a particular territory or people, informed the language regime of the precolonial Ahom kingdom.

35The Ahom kingdom’s shift from the Tai-Kadai Ahom language to the Indo-Aryan Assamese language signaled the vernacularization of the Sanskrit cosmopolis in Assam, at a critical junction when the kingdom’s mandala-state traditions were being disrupted. Pollock (1998a: 31–32) references the argument that when the kingdoms in the Sanskrit cosmopolis started being spatially defined, then vernaculars took over from Sanskrit, suggesting a close, if not symbiotic, relationship between territorialization and vernacularization. At the height of the Ahom kingdom, the two processes of vernacularization and territorialization were intertwined. That vernacularization happened in Assamese, rather than in the Ahom language, is at least partially explained by the flattening of the elite hierarchy through an influx of Assamese speakers at the critical juncture.

36While the legacy of the Ahom language regime can still be detected in the rich linguistic diversity in Assam today, as well as perceived in India’s ‘multilingual ethos’ (Montaut 2005), language politics in Assam today resonate with the modern nation-state’s language regime of linguistic rationalization — albeit modulated by India’s linguistic federalism, in that linguistic rationalization is the political agenda of states within the Indian union, rather than the nation-state writ large. In Assam’s case, linguistic rationalization grates against its linguistic diversity. That diversity reflects the legacy of Assam’s multilingualism emanating from a historical constant of language contact. The historically and theoretically grounded case study of Assam that I have presented above exposes the contingency of contemporary language politics informed by a language regime that is path-dependent on linguistic homogeneity.



2018 (5 November). ‘Assamese Language Under Threat: AXX’, The Sentinel (Dibrugarh edition), p. 1,

Ahmed, Sarharuddin. 2008. ‘Cultural Ties Between India and Ancient Cambodia’, Bulletin of the Assam State Museum 18: 19–23.

Anderson, Benedict. 1991. Imagined Communities (London: Verso).

Baruah, Sanjib. 2005. Durable Disorder: Understanding the Politics of Northeast India (New Delhi: Oxford University Press).

Baruah, Sanjib. 1999. India Against Itself: Assam and the Politics of Nationality (Philadelphia: University of Pennsylvania Press).

Bhattacharya, Sudhindra Nath. 2005. ‘The North-East Frontier — The Land, the People and their Early History’ (excerpt from author’s A History of Mughal North East Frontier Policy [1929]), in Discovery of North-East India, Vol. 1: North-East India — A Panoramic View, ed. by S. K. Sharma and Usha Sharma (New Delhi: Mittal Publications), pp. 1–39.

Brandt, Carmen. 2015. ‘Writing off Domination: The Revival of the Chakma and Meitei Scripts’, presentation in the Script Systems and Identity Formation in South Asia panel at the 44th Annual Conference on South Asia, October, Madison, WI.

Cederlöf, Gunnel. 2014. Founding an Empire on India’s North-Eastern Frontiers 1790–1840: Climate, Commerce, Polity (New Delhi: Oxford University Press).

Chatterjee, Partha. 2008. ‘Introduction: History in the Vernacular’, in History in the Vernacular, ed. by Raziuddin Aquil and Partha Chatterjee (Ranikhet: Permanent Black), pp. 1–24.

Chelliah, Shobhana and Lester, Nicholas. 2016. ‘Contact and Convergence in the Northeast’, in The Languages and Linguistics of South Asia: A Comprehensive Guide, ed. by Hans Henrich Hock and Elena Bashir (Berlin: Walter de Gruyter), Section 2.5.

Coedès, George. 1967. The Indianized States of Southeast Asia, ed. by Walter F. Vella; translated by Susan Brown Cowing (Honolulu: East-West Center Press).

Collins, Steven. 2003. ‘What is Literature in Pali?’, in Literary Cultures in History: Reconstructions from South Asia, ed. by Sheldon Pollock (Berkeley: University of California Press), pp. 649–88.

Dutta, Uddipan. 2016. ‘Sarbabhoumo Asom: Three Public Discourses’, in Unheeded Hinterland: Identity and Sovereignty in Northeast India, ed. by Dilip Gogoi (New York: Routledge), pp. 49–69.

Emeneau, Murray. 1956. ‘India as a Linguistic Area’, Language, 32: 3–16.

Errington, Joseph. 2008. Linguistics in a Colonial World: A Story of Language, Meaning, and Power (Malden, MA: Blackwell Publishing).

Evans, Peter, Dietrich Rueschemeyer and Theda Skocpol (eds.). 1985. Bringing the State Back in (Cambridge: Cambridge University Press).

Gait, Edward. 2015. A History of Assam (Delhi: Surjeet Publications, 8th Indian Reprint).

Gellner, Ernst. 1983. Nations and Nationalism (Ithaca: Cornell University Press).

Gogoi Nath, Jahnabi. 2002. Agrarian System of Medieval Assam (New Delhi: Concept Publishing Co.).

Gohain, B. K. 1999. Origin of the Tai and Chao Lung Hsukapha: A Historical Perspective (New Delhi: Omsons Publications).

Goswami, Surendra Kumar. 1986. A History of Revenue Administration in Assam, 1228–1826 A.D. (Guwahati: Spectrum Publications).

Guha, Amalendu. 1983. ‘The Ahom Political System: An Enquiry into the State Formation Process in Medieval Assam: 1228–1714’, CSSSC Occasional Paper (Calcutta: Centre for Studies in Social Sciences) 64 (October): 5–56 https://

Gurung, Tejimala. 2018. ‘Crisis of Sustenance and Decline of the Ahom State in Eighteenth Century Assam’, in The Falling Polities: Crisis and Decline of States in North-East India in the Eighteenth Century, ed. by Tejimala Gurung (Guwahati: DVS Publishers), pp. 95–124.

Guha, Amalendu. 1977. Planter-Raj to Swaraj: Freedom Struggle and Electoral Politics in Assam, 1826–1947 (New Delhi: People’s Publishing House).

Hobsbawm, Eric. 1990. Nations and Nationalism Since 1780 (Cambridge, UK: Cambridge University Press).

Kaviraj, Sudipta. 2010. The Imaginary Institution of India: Politics and Ideas (Ranikhet: Permanent Black).

Keyes, Charles F. 2003. ‘The Politics of Language in Thailand and Laos’, in Fighting Words: Language Policy and Ethnic Relations in Asia, ed. by Michael E. Brown and Šumit Ganguly (Cambridge, MA: MIT Press), pp. 177–210.

Laitin, David D. 1988. ‘Language Games’, Comparative Politics 20 (3) (April): 282–302.

Laitin, David D. 1989. ‘Language Policy and Political Strategy in India’, Policy Sciences 22: 415–36.

Lahiri, Nayanjot. 1984. ‘The Pre-Ahom Roots of Medieval Assam’, Social Scientist 12 (6) (June): 60–69.

Laponce, Jean. 1987. Languages and their Territories (Toronto: University of Toronto Press).

Leach, Edmund. 1954. Political Systems of Highland Burma (Cambridge: Harvard University Press).

Lieberman, Victor. 2011. ‘What Strange Parallels Sought to Accomplish’, Journal of Asian Studies 4 (November): 931–38, S0021911811001628

Meyerhoff, Miriam and Stanford, James N. 2015. ‘ “Tings Change, All Tings Change”: The Changing Face of Sociolinguistics with a Global Perspective’, in Globalizing Sociolinguistics: Challenging and Expanding Theory, ed. by Dick Smakman and Patrick Heinrich (London: Routledge), pp. 1–15.

Miles, William F. S. 2014. Scars of Partition: Postcolonial Legacies in French and British Borderlands (Lincoln: University of Nebraska Press).

Mitchell, Lisa. 2009. Language, Emotion, and Politics in South India (Ranikhet: Permanent Black).

Montaut, Annie. 2005. ‘Colonial Language Classification, Post-Colonial Language Movements, and the Grassroot Multilingualism Ethos in India’, in Living Together Separately. Cultural India in History and Politics, ed. by Mushirul Hasan and Asim Roy (New Delhi: Oxford University Press), pp. 75–106.

Parwaz, M. 2018. ‘Crisis in the Ahom State in Eighteenth Century’, in The Falling Polities: Crisis and Decline of States in North-East India in the Eighteenth Century, ed. by Tejimala Gurung (Guwahati: DVS Publishers), pp. 125–48.

Patten, Alan. 2001. ‘Political Theory and Language Policy’, Political Theory 25 (5) (October): 691–715.

Phukan, Girin. 2018. ‘Socio-religious Knowledge Culture of the Ahoms’, Indian Journal of Tai Studies, 18 (October): 193–225.

Pollock, Sheldon. 1998a. ‘The Cosmopolitan Vernacular’, Journal of Asian Studies 57 (1): 6–37.

Pollock, Sheldon. 1998b. ‘India in the Vernacular Millennium: Literary Culture and Polity, 1000–1500’, Daedalus, 127 (3) (Summer): 41–74.

Purkayastha, Sudeshna. 2008. ‘Restructuring the Past in Early-Twentieth-Century Assam: Historiography and Surya Kumar Bhuyan’, in History in the Vernacular, ed. by Raziuddin Aquil and Partha Chatterjee (Ranikhet: Permanent Black), pp. 172–208.

Reynolds, Craig J. 1995. ‘A New Look at Old Southeast Asia’, The Journal of Asian Studies, 54 (2) (May): 419–46.

Saikia, Arupjyoti. 2008. ‘Gait’s Way: Writing History in Early-Twentieth-Century Assam’, in History in the Vernacular, ed. by Raziuddin Aquil and Partha Chatterjee (Ranikhet: Permanent Black), pp. 142–71.

Saikia, Yasmin. 2004. Fragments of Memories: Struggling to be Tai-Ahom in India. (Duke University Press).

Sarangi, Asha. 2015. ‘India’s Language Regime: The Eighth Schedule’, in State Traditions and Language Regimes, ed. by Linda Cardinal and Selma K. Sonntag (Montreal: McGill-Queen’s University Press), pp. 205–18.

Scott, James C. 2009. The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia (New Haven, CT: Yale University Press).

Sonntag, Selma K. 2011. ‘The Changing Global-Local Linguistic Landscape in India’, in English Language Education in South Asia, ed. by Lesley Farrell, Udaya N. Singh and Ram A. Giri (New Delhi: Foundations Books), pp. 24– 35.

Sonntag, Selma K. 2014. ‘Depoliticizing Hindi in India’, in Defining the Indefinable: Delimiting Hindi, ed. by Agnieszka Kuczkiewicz-Fraś (Frankfurt am Main: Peter Lang), pp. 95–105.

Sonntag, Selma K. 2019. ‘The Liberal Tradition in America: A Historical-Institutionalist Approach to U.S. Language Policy’, in Language Politics and Policies: Perspectives from Canada and the United States, ed. by Thomas K. Ricento (Cambridge: Cambridge University Press), pp. 27–44.

Sonntag, Selma K. and Cardinal, Linda. 2015. ‘Introduction: State Traditions and Language Regimes: Conceptualizing Language Policy Choices’, in State Traditions and Language Regimes, ed. by Linda Cardinal and Selma K. Sonntag (Montreal: McGill-Queen’s University Press), pp. 3–26.

Terweil, Barend Jan. 1983. ‘Ahom and the Study of Early Tai Society’, Journal of the Siam Society 81 (2): 42–62.

Thongchai Winichakul. 1994. Siam Mapped: A History of the Geo-Body of a Nation (Honolulu: University of Hawai’i Press).

Thongchai Winichakul. 1996. ‘Maps and the Formation of the Geo-Body of Siam’, in Asian Forms of the Nation, ed. by Stein Tønnesson and Hans Antlöv (Richmond, UK: Curzon) pp. 67–92.

Tønnesson, Stein and Antlöv, Hans (eds.). 1996. Asian Forms of the Nation (Richmond, UK: Curzon).

Wolters, Oliver W. 1999. History, Culture, and Region in Southeast Asian Perspective, rev. ed. (Ithaca: Cornell University Press, in cooperation with the Institute of Southeast Asian Studies, Singapore).


1 The metaphor of the mandala invokes a cosmic universe depicted as a circle radiating outward, the center of which is the most spiritually enlightened or, in the case of states, the most politically powerful.

2 Sudipta Kaviraj (2010: 141–43) argues that likewise in Bengal prior to colonialism, there was ‘no linguistic identity’. He points to the ‘fuzziness’ of space and time in pre-colonial India, distinct from the enumeration and mapping of the colonial world.

3 In a personal conversation (29 October 2018, Dibrughar University, Assam) I had with Jahnabi Gogoi Nath, a historian of the Ahom kingdom (see, e.g., Gogoi Nath 2002), she expressed doubt whether the king whom Pollock references was Barahi. Given my argument that ethnonyms had little or no identitarian significance, or that kingdoms were not territorially defined and marked by a specific language before vernacularization, whether the royal patron of the first Assamese Ramayana was Barahi or not is less significant than that he was Hindu and probably a Tibeto-Burman speaker.

4 This assessment that the Ahom script was similar to the Mon script was also expressed by Girin Phukon, Director of the Institute of Tai Studies and Research in Moranhat, Assam during my visit there on 30 October 2018. However, in an article in the Institute’s journal, Phukon (2018: 212) speculates that the Ahom ‘acquired the script from the Pyu in Upper Burma’. Phukon’s assessments are not necessarily inconsistent, given that the Pyu (Tibeto-Burman speakers) could have themselves adopted the script in which Mon, an Austro-Asiatic language, was written. But they do indicate the high degree of both written and spoken language contact that was prevalent throughout the region historically.

5 Ahmed (2008), unfortunately, does not specify to which inscriptions he is referring.

6 For comparison’s sake, according to Carmen Brandt (2015), in the early eighteenth century, the Meitei king converted to Vaishnavism and, legend has it, destroyed the Meitei (Tibeto-Burman) script in the process of switching to a Devanagari-based script.

7 The linguistic effects of language contact is apparent today, according to Chelliah and Lester (2016: 305): ‘Assamese developed a classifier system unusual to Indo-Aryan on the basis of contact with Tai-Ahom.’

8 The view that Assamese is a dialect of Bengali still finds expression in such prominent places as the Oxford Dictionary. The 2010 edition’s definition of Assamese as ‘ [t] he Indic language which is the official language of Assam, related to Bengali and spoken by around 23 million people, roughly half in Assam and half in Bangladesh’ recently generated an online petition pointing out both its linguistic and territorial errors ( My thanks go to Mark Turin for informing me about this petition.


Professor Emerita of Politics at Humboldt State University in California and Affiliate Professor of Political Science at the University of Colorado, Boulder. Her research area is the politics of language, primarily in South Asia, but also in the United States, Europe and South Africa. Her numerous publications on language politics in South Asia have appeared in Language Policy, The Journal of Commonwealth and Comparative Politics, and Nationalism and Ethnic Politics, among other journals, as well as in over a dozen edited volumes. Her books include The Local Politics of Global English: Case Studies in Linguistic Globalization (2003) and State Traditions and Language Regimes (2015). Dr. Sonntag was a Research Fellow at the Jawaharlal Nehru Institute for Advanced Studies in New Delhi in spring 2012 and the recipient of two Fulbright research awards. She recently completed her tenure as chair of the Research Committee on the Politics of Language of the International Political Science Association. 


Volume papier
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search