Orthography and Semiotic Ideologies of Pronunciation

This case study builds on recent scholarship on orthography and scripts (Sebba Reference Sebba2007; Ahmad Reference Ahmad2008, Reference Ahmad2011; Jaffe et al. Reference Jaffe2012) as the metapragmatics of cross-modal correspondences that serve as resources for sociolinguistic and political action. These approaches define orthography is a system of “correspondences between elements of the script and elements of the language” (Sebba Reference Sebba2012, 4). This definition makes it possible to investigate the production of these correspondences as mechanisms for creating differentiated alignments between communities, linguistic forms, and evaluative orders. For example, LaDousa’s (Reference LaDousa2002) account of school advertisements in North India demonstrates that mixing scripts and codes produces new opportunities for identity construction amidst what seem to be rigid social stereotypes indexed by linguistic competency—thus creating powerful marketing resources for language schools. In Choksi’s (Reference Choksi2015) study of Santali writing on public signs in eastern India, the affordances of the graphemes and the signs themselves enable multiply scalar relationships between several linguistic varieties. These studies of the graphical-linguistic interfaces on public signs offer useful approaches to official naming politics, especially in places like South Asia, where intense multilingualism and multiscriptalism mean that any public name will be written in multiple scripts on signs and in official documents directed at a variety of linguistic communities. These approaches allow us to ask how changes in toponyms and linguanyms, as in Odisha, assert correspondences between graphical forms and linguistic code that in turn project social imaginaries authorized by the state’s official recognition of the names.

While these studies emphasize how productive script-code innovations can be for creating new political relations or recognizable social identities, in South Asia, the “sound shape of language” (Jakobson and Waugh Reference Jakobson2002) is also a central component of orthography’s correspondences. Many writing systems that developed on the South Asian subcontinent have robustly institutionalized ideologies about script-sound correspondences that make any discussion of writing potentially a discussion of sound. Thanks to this context, in cases of public signs or proper names in South Asia, writing choices become a potential source for indexing particular pronunciations and, from those pronunciations, identifiable sociolinguistic identities and stereotypical characters.

In addition to sociolinguistic identities, pronunciation may index speakers’ sociolinguistically entrained bodies. The basic fact that linguistic practice is constantly indexical of sociologically distinct bodies has been a key insight of sociocultural language studies of race, gender and sexuality (see Hall and Bucholtz Reference Hall1995; Zimman and Hall Reference Zimman2009; Alim, Rickford, and Ball Reference Alim2016). To think about how orthographies presuppose distinct bodies, I find inspiration in Bahri’s (Reference Bahri2017) account of the biology of postcolonial hybrids in the example of the name change of her native city. Historically, some residents produced Calcutta with its final syllable an unvoiced retroflex stop, and others said Kolkata with a dental stop in its final syllable. This single sound contained a world of difference: the dental stop spoke of laboring classes, localness, and the city’s small and crowded lanes, whereas the retroflex stop mapped the unvoiced alveolar consonant characteristic of British English into Bengali mouths. Upper-class residents’ pronunciation of Calcutta thus mimicked the British pronunciation, wrought out of difficult linguistic labor, both emblematizing and helping to entrain the class of English-oriented Bengalis imagined in nineteenth-century educational programs. More recently, as reparative or identity assertions over urban spaces have become a key mode of political enunciation, these colonial diglossic inheritances have become a political resource of a new kind. In a political reorganization of indexical orders, now the pronunciation previously marked as lower class and uneducated has become a sign of one’s alignment with anticolonial identities and loyalty to a primordial Bengali culture.

While such ideological transformations are well familiar to students of language and semiotics (e.g., Irvine and Gal Reference Irvine2000), Bahri’s account helps us to narrate these transformations as embodied activities. In her account, official names work as body activities through which “tongue, teeth, breath, diaphragm, palate, mouth, and elocutionary training collaborate in the pulse production of language lined with flesh” (Bahri Reference Bahri2017, vii). This is similar to Bourdieu’s concept of “habitus” or “bodily hexis” (Bourdieu Reference Bourdieu1990) in its attention to how bodily dispositions are entrained through class-based value production and socialization. However, rather than seeing the body’s incorporated history as a strong determining force (Throop and Murphy Reference Throop2002), Bahri’s materialist account of the biology of postcolonial dispositions is more open to the body’s repurposing for new ethical, aesthetic, and political ends. This approach to embodiment shares the spirit of the sociosemiotic study of qualia (Chumley Reference Chumley2013; Harkness Reference Harkness2015), as well as the study of sign communication gesture (Friedner and Helmreich Reference Friedner2012; Green Reference Green2017; Hoffmann-Dilloway Reference Hoffmann-Dilloway2018), especially insofar as it refuses to separate the body as a prior (or natural) phenomenon to which semiotic meaning is a mere cultural add-on. These approaches help us account for how the human body plays multiple semiotic roles—as the historically enskilled material producer/interpreter of sign activities, as a sign vehicle itself, and as an object of signification. Embodiment, and stances toward it, may be a significant feature of semiotic activities in general, even those that apparently have little to do with meaning-making about bodies.

Here, I seek to illuminate only a narrow component of the body’s enrollment in semiotic activities, namely, that the embodiment of linguistic signs is itself a focus of contrasting views or orientations. I show that arguments about orthography differently imagine linguistic signs—here, pronounced sounds—as either being produced by real bodies or as existing in an abstract, unembodied way. Actual speaking bodies figure differently into the debaters’ accounts of how signs work, and these differences are part of what give shape to their distinct political visions. I analyze these different accounts as semiotic ideologies, by which I mean the explicit and implicit assumptions about what signs are and how they function (Keane Reference Keane2003, Reference Keane2006, Reference Keane2018). Each of the two main viewpoints on the naming of Odisha and its dominant language, the view in support of the name changes and the view against it, assert visions of the region in relationship to the national public. Their different orientations to speaking bodies are integral to these different visions.

While clashes over semiotic ideologies have characterized major political and religious conflicts, sociologically the debates over orthography that I describe here closely resemble the language ideological debates described by Blommaert (Reference Blommaert1999). Like those Blommaert’s volume addresses, these debates are structured by state governance procedures, and political infrastructures provide their discursive framework and telos. Additionally, these elite arguments enroll varieties of linguistic expertise in discourses that are “loaded with all kinds of tropic, metaphorical or symbolic associations” (429), thereby turning problems of language into problems of national identity—in this case, regional identity nested within the nation.

The rest of this article is organized around the arguments for and against Odisha’s name changes. I draw entirely on published sources, mostly from the official publications of India’s houses of parliament, the Lok Sabha and the Rajya Sabha, as well as published media commentary.Footnote ¹ Like other studies of language ideological debates, this case study is limited to published statements by elites, here all elite men. Though my evidence may not be sociologically generalizable, it can help us understand how differences in semiotic ideologies can be enrolled in and managed by routine liberal-democratic discursive procedures at the highest level.

In the next section, I describe the sociolinguistic and political context of the debates. After these foundations, I turn to a detailed discussion of the arguments themselves, focusing first on the overarching name change justifications and then on three specific script-sound correspondences. The arguments involve, implicitly, a shared model of the correctly spelled name as a semiotic process involving the name’s graphic representation, its pronunciation, and its resemblance to the authentic name; it is the sign relations between these parts that the proponents and opponents of the name change conceive differently. Investigating the semiotic ideologies that mediate different proposed spellings opens up the problem of the differently speaking bodies that the regimentations do or do not admit, which I explore by returning to Bahri’s account of the body politics of naming. I conclude with a discussion of semiotic ideology as a framework for studying orthographic politics.

One of the challenges of writing about official name change politics is that it requires taking a position in the debate. Here I follow official usage unless I am describing features of the debate.

The Odia Language and Its Script

According to the most recent Indian census, about 37.5 million people speak Odia as a mother tongue, inclusive of a wide range of regional varieties (Office of the Registrar General & Census Commissioner, India 2011). Most of these speakers are concentrated within or near Odisha state on India’s east-central coast. Odia is an Indo-Aryan language that shares many features with Bengali, Assamese, Nepali, Hindi, and other North Indian languages. Like other Indo-Aryan scripts developed from Brahmi, Odia’s script is an alphasyllabary in which the graphic unit is an akshara, usually a consonant-vowel cluster, that can also be decomposed into phonemically distinct visual units (Salomon Reference Salomon2007). This script has been the primary regional script for writing Sanskrit manuscripts since the medieval period, and there is a great deal of regional curiosity about and pride in its history.

The close identification between Odia and its associated script developed through a linguistic nationalist movement in the late nineteenth and early twentieth centuries. This linguistic movement developed, like many of India’s linguistic nationalist movements of the colonial period (such as Mitchell Reference Mitchell2009), as a modernization and identity project, but also as a mode to address political concerns about the marginality of the region in the British awareness. The region that is now Odisha state was at the time split across three different British administrative units, and residents believed this marginality helped lead to a devastating famine in 1866. The linguistic movement first articulated itself as a response to a series of debates in Calcutta in the 1860s about whether Odia merited status as a separate language, or whether Bengali should be made the literary standard and language of education in Odisha, on the rationale that Odia would trap its speakers in poverty, backwardness, and superstition (Dash Reference Dash1983; Mohanty Reference Mohanty2002; Mohanty Reference Mohanty2005). The existence of a script so visually distinct from Bengali—Odia has no top line, and its letters are almost entirely round—became central to arguments for Odia’s distinct status as a full language (Beames Reference Beames1870). The initial success of Odia’s self-assertion against Bengali was later replicated in the consolidation of many Odia-speaking areas in a separate province of “Orissa” in 1936 and the consolidation of that province with most of the region’s princely states in 1947, both based on the logic of shared language and script use (N. Mohanty Reference Mohanty2005). Thanks to this political history, despite a historically low rate of literacy in the region, the script is an important source of identity and solidarity for Odia speakers.

The name changes under discussion here could be seen as an extension of these language politics, but they have also served a canny electoral strategy by Odisha’s ruling party. When first proposed at a state cabinet meeting in 2008, Odisha’s name changes were an apparent outgrowth of the legislative coalition that Odisha’s dominant political party, the Biju Janata Dal (BJD), had formed with this same Hindu nationalist party. The BJD initially formed in the wake of the death of an extremely popular local politician, Biju Patnaik, when his son Naveen Patnaik entered politics. The early electoral successes of the younger Patanik and his new party relied on the strong coalition that they formed in 2000 with the well-established machine of the Bharat Janata Party (BJP). When first proposed, rumors that the name change had been an agenda item for the coalition since its establishment in 2000 saw it as a continuation of the BJP’s own nativist renaming strategy. However, when Naveen Patnaik strongly condemned the BJP for its response to anti-Christian violence in southern Orissa and broke the coalition in 2009, the year after the first name change vote in the state assembly, the political strategy of the name change took on a new dimension. Instead of continuing the explicitly nativist-Hindu line of the BJP, Odisha’s name change became a way for the BJD to take over the Hindu cultural territory claimed by the BJP. Since that time, the BJD has worked hard to take over the economic space occupied nationally since the 1990s by the Congress Party, providing popular state poverty-relief programs alongside FDI- and market-friendly economic reforms, all while generally avoiding the cultural and religious issues that mark the divide between the BJP and the Congress Party.

In the recent 2019 national election, the BJD was one of the ruling BJP’s fiercest foes in its effort to dominant state electoral politics around the country. Precisely because of moves like this renaming, the semisecularist state party can compete with the BJP’s claim to protecting traditions and identities. Indeed, the renaming project is perhaps the clearest articulation of a BJD cultural project, if there could be said to be such a thing, as something like “regional pride in all things regional.” The so-called correction of the “wrong spelling” of the state’s name has helped support this self-referential, controversy-free regional pride. Involving a mere shift in Hindi and English orthography, this name change is a regional assertion that has not invoked any of the sectarian issues frequented by the Hindutva political groups. This is not to say that it is politics-free, but as a social and political project it has been largely uncontroversial—the debates have focused on the linguistic methods of determining the accurate names, not the fact of the change itself.

What Official Names Can Do: The Justifying Discourses

I now turn to what formed the majority of the content of the parliamentary debates: speeches explaining the value of the name changes and what they will achieve. The name change bill was debated for one day in each house, first in the Lok Sabha in November 2010 and then in the Rajya Sabha in March 2011. In both of India’s national houses of parliament, the lower Lok Sabha, or House of the People, and the upper Rajya Sabha, or Council of States, the parliamentary arguments were consistent, even uniform, in their assertion that the name change would reclaim pride for Odisha and Odias. As I describe below, even opponents of the name changes did not contest the rhetoric of reclaimed pride; they assented to the premise that a public name produces self-respect and merely disagreed about how to achieve it. I describe these justifying statements in detail because they illustrate the overarching interpretive frameworks for the proposed orthographic shifts—they narrate how official names work.

The parliamentary speeches widely asserted a view that the pride of the state, the language, and the people would be achieved by use of the right or authentic name. Here, the authentic pronunciation is clearly “ours”—it is how Odia speakers pronounce the names of their own language and state. Speaking in the upper house, MP Rama Chandra Khuntia, a vocal supporter of the name change since 2008, declared, “This is a historic moment as the long-standing wishes of the people have been fulfilled and their dignity upheld.” Other speakers in both houses spoke similarly:

Explicit phrases like “restoration of Oriya pride,” “sense of achievement,” and “Odia self-respect” justify the name change by making it an issue of how Odias see themselves in relationship to “the rest of India.” While linking linguistic forms to a complex of assumptions that merge identity with political rights and linguistic preservation (Duchêne and Heller Reference Duchêne2013). These tropes of pride also suggest expansive and elevated qualities: the right names help the status and reputation (and, implicitly, the economy) of Odisha grow upward and outward.

Similarly, parliamentary speeches’ narrative structures mimed a temporality of self-respect. These narratives began with the inspiring names and accomplishments of an earlier time, delved into the dark times of the mispronounced name, and then turned finally to hopeful statements about the present or future. They declare that Odisha’s past glories were lost to colonial subjugation and forgotten, but the name changes will allow us to reclaim of this past glory:

Odisha-the-place parallels the Odisha-the-name; the same process of colonization that corrupted and obscured Odisha’s civilizational achievements also corrupted its name. These chronotopic narratives of historical pride, lost and regained, help both to define the nature of the name changes themselves and to establish their political necessity. They anchor the new orthographies in the glorious precolonial pasts and reconfigure Odisha’s place within the nation. In this vision, the name changes become agents of the ultimate goal in Odishan politics: to transform a marginal language and an impoverished or “backward” state into a widely respected language and a profitable, highly developed state.

This matters for an account of orthography because the name change’s multilingualism is essential to achieving such reclaimed pride. At the center of this imagined transformation in Odisha’s scalar position vis-à-vis the nation is the focus on how outsiders, non-Odias, speak and write Odisha’s names. This is evident, first, in the participation format of the official name change, moving from the state legislature to the national parliamentary bodies, which explicitly frames the entire naming process as recognition by the nation of the state’s self-identification. Second, the format of the parliamentary speeches restages this national recognition—it is a forum for explicitly addressing the nation. The parliamentary recognition of the new name changes the relationship of Odias to themselves by transforming the relationship between Odias and non-Odias. The discourse of “pride” glosses the essential role of the audience in the name change: when they call us by our right name, then we will be fully ourselves again.

Finally, Odisha’s name changes entirely focus on the outsider audience. While many other recent South Asian name changes, like Madras to Chennai, or the more recently discussed shifts from Ahmedabad to Karnavati and Allahabad to Prayag, involve changes to the names for everyone who speaks them, Odisha’s name changes affect only non-Odia versions of the name. The goal is that Odia speakers will no longer suffer the “the ignominy of [their name] being pronounced erroneously,” as Pyarimohan Mohapatra put it (Rajya Sabha Debates 2011, 20). This is the basic assumption of the supportive parliamentary speeches, as voiced by MP Bhatruhari Mahtab, who stated that “after 75 years, the name of Orissa is being changed to the actual pronunciation which we do in Odisha” (Lok Sabha Debates 2010, 89). The promise of the orthography shift is not only that Hindi and English speakers will speak the names the way that Odias say their own names but also that these national- and international-language speakers’ entire orientation toward the state and its language will change as a result. It suggests that, through the right names, English and Hindi speakers—an audience of outsiders—will begin to see the until-now hidden glory of Odia, and that this will raise the status of the language, its state, and its speakers to a position more akin to the national and international languages English and Hindi.

These narratives justifying the name change contain within them a semiotic model of how this naming should work. We can describe this model in Peircian semiotic terms of the sign, or representamen; the object that the sign signifies; and the interpretant that construes the relationship between the sign and the object (Parmentier Reference Parmentier1994). What emerges from these statements is a model of official names that begins with the semiotic object and works backward to the sign vehicle. Here, the agreed upon political goal is for the name to signify the quality “like we say it,” which, here, is a semiotic object that itself bundles an additional semiotic construal of key sensuous characteristics of local pronunciations as authenticity, correctness, and local belonging. Having established the object or immediate goal for the names’ spelling, the project of official renaming is to construct the sign vehicles, which here are the names’ spellings, that will best allow some interpretants to construe them as signifying the desired object.

This raises the question, what is the interpretant of the name? To reiterate so far, the graphic appearance of the name is the sign or representamen, which must be designed so that it can determine a particular interpretant, which will, in turn, construe the sign as signifying a particular semiotic object. In this model, the interpretant is the name’s sound sequence or pronunciation. More specifically, it is the pronunciation of the Odia names by Hindi and English speakers. This is evident in the discourse about the expansiveness of pride, as well as in the whole debate’s focus on the names “in Hindi” and “in English.” The entire imagination of how the naming process works—its ability to unleash the sought-after political transformations of development, national respect, and self-esteem—are focused on how to spell the names so that how these outsiders say them will connect the written names to “like we say it,” to correctness and authenticity. However, as I consider below, even as all of the debate’s participants share a model constructing the desired interpretant in this way, the proponents and opponents of the name change disagree about the character of the interpretant and especially about whether it serves as a sign of embodiment. I now turn to how the proponents of the name change solve for the sign vehicle and come to propose the new names’ orthography.

Supporting the Name Change: Script-Sound Isomorphism and Sanskritic Authority

Amidst their elaborate speeches on the benefits and reasons for the name changes, the name changes’ proponents in parliament said remarkably little on behalf of the linguistic features of the name changes themselves. The most explicit talk described the name changes as “corrections.” For example, early news reports on the proposed name change in 2008 quoted an anonymous state officer saying, “We want the Union government to make the changes so that state’s name is spelt and pronounced correctly” (Times of India 2008); the state’s home secretary was quoted saying that the pronunciation disparities were due to the “wrong spelling of state’s name” (Rediff India Abroad 2008). As talk of “correction” suggests, their implied criterion for evaluating the rightness of a name is its degree of correspondence to a legitimate standard or source. The proponents’ view of the name is not that they must design the right spelling but that they must discover the right spelling. The sign vehicle that signifies the desired object is out there and knowable; the right spelling preexists the search for it.

To make sense of how the orthographies proposed in the constitutional amendments serve as the sign vehicle that will signify the desired semiotic object, members of parliament drew on a relatively explicit South Asian ideology about orthography as consisting of naturally occurring grapheme-phoneme correspondences. This widespread view assumes an isomorphism between writing and pronunciation that is prevalent among Odia speakers and many other South Asian language users. South Asian scripts are notable for their high degree of grapheme-phoneme correspondence, as reflected in their linguistic categorization as alphasyllabary scripts (Mahapatra Reference Mahapatra1996; Salomon Reference Salomon2007). However, in fact, in speech strings the phonological mapping of such syllabalic, or akshara, graphic clusters are contextually variable (Nag Reference Nag2014). The persistent strength of the belief in sound-script isomorphism has been shaped by a variety of traditions and institutions that narrate linguistic correspondences across North Indian languages as an outcome of their genesis in Sanskrit—a view that has had an enormous impact on generations of primary students learning to recite the basic graphemes, or aksharamala (barnamala in Odia), according to the Sanskrit grammarians’ organization of grapheme-phoneme units by place of articulation (see Deshpande Reference Deshpande1995). This Sanskritic history infuses the presumed sound-script isomorphism with authority and links everyday orthography to an imagination of Sanskrit as the “eternal language” that exists as a “linguistic flatland beyond time and space,” as the true ideal language that lies behind all other language (Deshpande Reference Deshpande1996, 401).

Accompanying this expected isomorphism between writing and pronunciation in one language, such as Odia, is a parallel expectation that script-sound isomorphism makes it possible to establish accurate and straightforward correspondences between the script and sound pairings across different languages. We can see it at work in one of the elements of the orthographic shift that has not been the focus of debate. The previous Devanagari writing of the state’s name, उड़ीसा (uḍīsā) begins with a vowel [u] that is mutually recognizable across Hindi and Odia speakers, and written in Odia’s script as ଉ. Similarly, the vowel [o] that begins the Odia script writing of the same name, ଓଡ଼ିଶା (oḍīśā) has a direct correspondent in Devanagari’s ओ and is mutually recognizable across speakers of Hindi and Odia. In short, both scripts have graphemes that correspond to both phonemes. In this situation of relatively functional isomorphism, to make readers of both languages say the name approximately the same requires simply deciding which grapheme-phoneme pair to use—the pair that Odia script-writers use (ଓ or ओ, [o]) or the pair that Devanagari-writers use (ଉ or उ, [u])—and then doing so across both languages. Focusing on this example makes the name change look like a straightforward change from “their” pronunciation and writing to “ours.” This view of graphic-sound correspondences mirrors how denotational referential ideologies imagine translation as finding transparent correspondences between languages (Haviland Reference Haviland2003), similarly imagining an implied abstract, authoritative reference that anchors cross-linguistic sound-grapheme clusters.

Underlying this ideology of script-sound isomorphism, then, is a strong belief in traditional authority generally, especially in and through Sanskrit texts. Script-sound isomorphism authorizes and anchors sounds in received linguistic traditions. The important role of authority in the name change becomes especially evident in a Lok Sabha speech about a sound that interrupts the ideology of script-sound isomorphism. Here, even while disagreeing about the proposed orthography, the speaker appeals to textual authority:

At issue in this statement is the sibilant in the third syllable of the state’s name, equivalent to the “sh” in Odisha. Both Odia’s script and Devanagari have three different graphemes for sibilants that Sanskrit grammarians distinguished by their ideal place of articulation, either as palatal, retroflex, or dental (Deshpande Reference Deshpande1995). Most Odia speakers pronounce all three of these graphemes identically when they speak Odia, often as [s] (see Mahapatra Reference Mahapatra1996). For speakers of some dominant Hindi varieties, the graphemes associated with the palatal/retroflex sibilants and the dental sibilants are phonemically contrastive (Masica Reference Masica1993)—the contrast in Hindi maps more easily on the dominant English phonemic contrast between the sounds associated with “sh” and “s” than it does onto the Odia sibilants.

In his statement, Mahtab points to a disruption of the ideology of cross-linguistic orthographic transparency by this mismatch between Hindi, English, and Odia sibilants. Mahtab proposes that since there is a historical variability in the grapheme used in Roman script and Devanagari for what is written in Odia as ଶ (ś), it is totally reasonable to continue using the Roman s in the name. What is notable is that he refers to the Odia character in question by using a name, talvesa, that reflects popular understandings of Sanskrit grammarians’ accounts of Sanskrit’s three different sibilants. His appeal to the Sanskritic category “talvesa” establishes that his preference is in fact the most textually sound and authoritative. Even as Mahtab makes an argument against the name change and for the older spelling with s, his argument’s appeal to traditional, textual authority is in line with that of the name change proponents. Even disagreeing with the specifics, Mahtab’s argument is friendly to the proponents’ imagination of the orthographic change as guided by Sanskritic textual authority.

Contesting the Name Change: Sociological Variation

Opponents of the name change share the proponents’ general goal of having an official name that is “like we say it.” Even as they share this goal, however, they disagree with proponents about how to reach this goal—they disagree about the sign vehicle that will signify the widely recognized desired object. The point of difference here is how the opponents conceive of the sign vehicles working. While the name change proponents drew on script-sound isomorphism as a means of establishing textual authority through the new name, their critics drew on cosmopolitan, sociologically sensitive approaches to pronunciation and local history. This section describes the clearest articulation of the opponents’ views, which were focused on the second syllable in both the state’s and the language’s names, written as either an r or d in the Roman script. The consonant in question is a voiced retroflex flap [ɽ]. Linguists routinely note that, in Odia, this voiced retroflex flap is an allophone of the stop [ɖ], one which appears in both intervocalic and final positions (Masica Reference Masica1993; Ray Reference Ray2003, 490). The Odia script grapheme associated with the voiced retroflex stop, ଡ, is routinely altered with a diacritic underneath, ଡ଼, to mark that it is a flap, not a stop. That the diacritical mark provides a graphic contrast for this variation suggests that it may also function as a phonemic contrast. This sound and its graphic representations pose a major challenge to an expectation of cross-linguistic isomorphism and a reliance on textual authority because it is a sound characteristic of Odia, and found in other South Asian languages, but entirely missing from non-Indian English.

Debates about how to represent this voiced retroflex flap in Odisha’s names when writing “in English” follow from of over two hundred years of politically motivated Indic language Romanization projects. The Government of India’s current official Romanization standard relies on the Hunterian system, adopted by the colonial government in 1872 following a campaign by William Wilson Hunter (Reference Hunter1873). Coincidentally, Hunter advocated for his system by circulating a pamphlet of place names transliterated in his system in the very same month that his historical account, Orissa (Reference Hunter1872), was published, which his biographer described as transforming him into a “literary lion” (Skrine Reference Skrine1901, 207). As Hunter’s own book title suggests, there is no special place for the voiced retroflex flap in the Hunterian system, which sometimes writes it as a d and sometimes, as in Orissa, as an r; indeed, the system does not distinguish between dental and retroflex stops at all, though they are phonemic contrasts present across Indic languages. Hunter explains this choice as intentional, arguing that “no typographical device will enable the ordinary English tongue to understand or to utter the nice modulations of oriental speech” (Reference Hunter1873, 24). This line displays Hunter’s motives: he sought a system to make Indic languages plain for English-only speaking mouths, not a system that would tell Indian mouths how to say words reliant on Indic languages’ phonemic contrasts. The scalar associations of the English language with the “global” and Indian languages with the “local” (LaDousa Reference LaDousa2005; Choksi Reference Choksi2015) are thus historically built into Roman script orthographies in India.

The voiced retroflex flap was the focus for two committed opponents of the name changes. Senior Bhubaneswari journalist Subhas Chandra Pattanayak wrote numerous energetic, detailed accounts of how the change is wrong on his personal blog—a prolific and popular blog among local literary activists that is predominately in English. Pattanayak’s overarching argument has two steps. First, he argues that the proposed name change—from Orissa to Odisha—dishonors the Odia language by ignoring the spoken and sometimes written distinction between the voiced retroflex stop and the voiced retroflex flap. The new spelling’s d erases this distinction entirely. He uses a side-by-side image of two Odia characters to draw attention to the frequent use of the diacritic mark, or nukta, in the Odia grapheme, writing, “The uniqueness of Oriya language lies in [the contrast in the two sounds and graphemes]. The single alphabet D appears in two forms: D in the first position of a word and D with dot underneath in subsequent positions. Therefore, the alphabet in second position has a dot underneath as in the name of our motherland and in the name of our mother tongue” (Pattanayak Reference Pattanayak2012). Pattanayak writes that the “archaic magnificence” of the Odia language is the precision with which the use of the diacritic maps the stop/flap variation in speech. Though Pattanayak emphasizes the literary and textual history of the graphemic contrast, this differs significantly from the textual authority focus of the proponents. First, Pattanayak proposes to celebrate the voiced retroflex flap as a meaningful feature of Odia. Emphasizing this sound not found in Sanskrit distances the names from an association with Sanskritic textual authority. Second, though Pattanayak is precisely interested in the existence of one-to-one correspondences between sound and script in Odia, he constructs this correspondence as the language’s own sociolinguistic attention of how Odia people speak. This is not a question of people correctly reading the script to make the right sounds, in which case the text can act as an authority controlling speech, but of the language noticing and reflecting how people speak. Here, the “archaic magnificence” of Odia is that, as a language, it became a repository for social knowledge about the Odia people.

As a result of Odia’s linguistic sensitivity to its social world, Pattanayak argues, the names should maintain the distinction between the flap and the stop. But how to do this? He writes, “We Oriyas understand the difference and use the two different forms of pronunciation marked by the dot under the second avatar of 13th consonant in tune with the phonetical necessity. But how the non-Oriyas, particularly the English-speaking foreigners are to utter the said alphabet used in the second or third or any later position in a word without violating our peculiar pronunciation? This question had engaged our founding fathers in deep cogitation for quite a long period and eventually the approximate correct phonetic presentation was evolved” (Reference Pattanayak2015). Pattanayak here makes explicit that the problem with the name is the non-Odia-speaking mouths. In his view, which contrasts with the position implicitly held by the name change’s proponents, the Roman script version of a name that Odia speakers can correctly pronounce will not necessarily produce the correct sounds when read by non-Odia speakers. Instead, again evoking the linguistic history of Odia, he proposes that the original Roman script orthography for the name—Orissa—was developed precisely as a solution to this problem of English mouths.

A second vocal opponent of the name change has been Tathagatha Satpathy, a prominent MP and the owner and editor-in-chief of one of Odisha’s most popular Odia-language daily newspapers. Satpathy has refused to change the name in his English-language publications, including the title of the English daily Orissa Post. For Satpathy, the argument against the name change parallels Pattanayak’s: shifting the spelling from r to d will make the pronunciation of the names by English speakers more incorrect. Making his argument on a national news program in 2009, he said, “My point is that if a word like Bombay was not pronounced by Maharastrians as Bombay but as Mumbai for ages, changing the name from Bombay to Mumbai makes sense to me. But Odias, per se, never said Odisha or Odia, we never did that. And, it doesn’t actually change in the Odia language.”Footnote ² Here, Satpathy makes his point with an elaborate pronunciation of the consonant in the second syllable of the emphasized Odia and Odisha. He does not pronounce the word with a voiced retroflex flap but rather stabs his tongue into his palate to emphasize the stop, clearly contrasting it with his relaxed flap in the rest of the statement. With this emphasized stop, Satpathy offers a performance of the name as it would be spoken by nonlocal English speakers based only on the Roman script orthography. In the November 2010 Lok Sabha debates on the bill, Satpathy similarly argues for the inaccuracy of the proposed name change. In these debates, Satpathy’s pronunciation distinctions likely matched those from the prior year: “We all know that in Orissa, we always say, we are Oriya; we do not say, we are Odiya; we do not say that my State is Odisha. It is not the pronunciation of the local people” (Lok Sabha Debates 2010, 100). In these statements, Satpathy presents the orthography in its starkest light. For this politician-editor, orthography is not connected to Sanskritic traditions or the magnificent histories of Odia’s own sociological awareness; it is merely a utilitarian tool to achieve the most accurate, authentic pronunciation from sociolinguistically diverse mouths untrained in Odia.

Of course, there do exist multiple Romanization standards for Indic languages that can represent the contrast between the voiced retroflex stop and flap, often as ḍ and ṛ following Monier Williams’s Sanskrit-English Dictionary (Williams Reference Monier-Williams1872); this is the Romanization standard I have used in this article. It is the persistence of the Hunterian system across public uses of Roman script, such as official maps and road signs, that enables an Indian political debate to focus on the interlingual representation of a single grapheme-phoneme pair. While Romanization systems have been the site of a great deal of political (and language ideological) activity elsewhere in Asia (see Kim-Renaud Reference Kim-Renaud1997) as well as an intense interest of computer scientists in India, the Indian government has shown little concern for them (United Nations Group of Experts on Geographical Names 2007).

The Embodied Publics of Official Names

In many political clashes that are also semiotic ideological conflicts, the key point of differentiation is focused on the character either of the sign vehicle or of its object. Keane (Reference Keane2018, 67) provides two characteristic examples: During twentieth-century secularization movements in both Russia and China, atheists attacked religious shrines to assert that the shrines’ purported semiotic object (God) did not exist. During the Protestant Reformation, doctrinal differences did not center the existence on a transcendental God but on the nature of religious sign vehicles, like the liturgical ritual. By contrast, in the case study of an official name, there is little disagreement around the semiotic character of either the sign vehicle or the object it signifies. Here the disagreement focuses on the character of the interpretant of the names’ orthography: the pronunciation. Because of that focus, the disagreement also addresses the characteristics of the sign processes that condition that interpretant—that is, embodiment.

To see this more clearly it is helpful to return to the semiotic model promoted by the debate’s justifying discourses. According to these justifications, Odisha’s name changes assert that the change will improve the accuracy of the names and thus reclaim the pride of Odias. If constructed successfully, the graphic forms of the names do this by making English and Hindi speakers say the names in a way that establishes the names as authentic and pride-worthy. For this process to effectively expand esteem for the language, its state, and its people, the interpretant must be successfully uttered by outsiders. In this model, the graphic forms are the sign vehicle, which determine the sounds as the interpretants, which in turn construct the relationship between the graphic forms and the correct and authentic name. The specific interpretant is essential for the entire name change to function successfully—without it, the sign vehicle fails to signify that the name is “like we say it,” and the rest of the name’s political aspirations fail.

Despite implicitly sharing this model of the official name, the proponents and opponents solve for the written form—the spelling—in opposing ways. The proponents of the names seek authoritative anchors in Sanskritic textual traditions, imagining script-code isomorphisms across codes that have been made possible through Sanskrit’s existence as an ideal outside of history. They see themselves as not creating a new name so much as discovering the correct one. Opponents, by contrast, seek the best graphic form as an empirical project, considering how individual letters are likely to shape the sounds that non-Odia speakers produce. For them, the Roman script spelling is merely a practical problem of managing outsiders’ unfamiliarity with Odia sounds.

At the heart of these different methods for composing the correct names are fundamentally different approaches to the interpretant, that is, to the sounds of the name. The actual spoken sounds are the epistemological center of the name change for the opponents, who focus precisely on the limits of outsiders to produce Odia’s distinct sounds. Meanwhile, the sounds actually produced are almost entirely invisible to the proponents, who are not concerned about the potential for pronunciation difficulties. For them, sounds exist abstractly, not as pronunciations made by speaking bodies, but as sound ideals. The different approaches to the materiality of the interpretant, specifically whether it is an abstract sound ideal or an actual set of sounds pronounced by some real people, reflect conceptions of the semiotic processes that produce those sounds. At issue is whether the sounds, to be accurate, must reflect an authoritative tradition. The divergent answers to these questions reflect different orientations to speakers’ mouths and, through them, to speakers’ bodies.

What is at stake in these different orientations to an official spelling’s sonic interpretants is the character of the name’s audiences or publics—audiences who not only hear names but also speak them. Bahri’s account (Reference Bahri2017) of the embodiment of official names suggests why such different orientations might matter. Real readers and speakers of names are fleshy non-abstract people whose linguistic capacities have been shaped by their lifetimes of linguistic habits. She recalls Fanon’s description of pronunciation practice as a bodily discipline, in which his narrator takes “great pains with [his] speech because [he] shall be more or less judged by it” (Fanon [1952] Reference Fanon2008, 20; cited in Bahri Reference Bahri2017, v). Material-semiotic capacities such as sound production reflect concerted embodied labors and socializations. To make new sounds, the bodies of the speakers have to change, they have to let the language in—or force it to take hold. In pronunciation, even with the production of new habits, it is hard (or takes exceptional personal skill) to lose all traces of early oral/aural habits. We see this with the pronunciation of Calcutta in Bahri’s account, where the retroflex t stands for the British alveolar t but does not actually reproduce it. Though, to Bengalis, the retroflex in Calcutta has been a sign of elite belonging and British orientation, to British ears, the retroflex t has been a sign of even educated Bengalis’ perpetual exclusion from authentic English. For the British, the transposition of the alveolar to the retroflex consonant marked the unbridgeable gap between native bodies and English civilization. Such resistances of tongues, lips, and jaws to new linguistic habits in adulthood is part of what makes speech such a potent focus for social differentiation and the control of human bodies. Whether or not an official name incorporates the recognition of such embodied limits, then, constructs a different political existence for these publics, and especially the political relevance of their socially differentiated bodies.

For the name change opponents, some awareness of the limits of speaking bodies makes it possible to emphasize sociological variabilities. In contrast, the proponents’ views on orthography do not value sociolinguistic variability; instead, they seek to manage variations in sounds across languages by nailing them to Sanskrit—they seek to constrain linguistic diversity through reference to an authoritative ideal. This approach to the interpretant suggests that Hindi and English speakers evoked in the framing justification discourses are entirely abstract for these proponents. These outsiders serve merely as a point of structural distinction, as a quality of outsideness, but are not imagined as substantively real humans. Instead, the reliance on Sanskrit mirrors an effort to purify the interpretant of its materiality, leaving in its place a disembodied interpretant of pure, ahistorical sound.

Semiotic ideologies help explain how the treatment of language is also implicated in the treatment of things and people (Keane Reference Keane2006). In this case, the treatment of orthography is also implicated in these politicians’ treatments of political authority. For the name change proponents, authority is not simply a question of state sovereignty or democratic legitimacy. The authority to name, here, is a dance between the nation-state and Hindu tradition. Traditional authority buttresses or legitimates the nation-state, even as the nation-state must recognize or baptize that very same traditional authority in order for it be effective. Adopting this set of assumptions is a canny political move for the state’s regional Biju Janata Dal party. It is in line with the competing Hindutva Bharata Janata Party’s political agendas, but it also reflects a longer standing political role for Sanskritic traditions in coastal Odisha borne of a close association between regional identity, the local Jagannatha cult, and the state’s important pilgrimage sites (see Pati Reference Pati2012). Thanks to these Brahmanical Hindu sources for local political belonging, an appeal to Sanskritic authority can be a unifying, low-controversy move for the regional political party, even as it takes over the Hindu-tradition territory that the BJP has sought to stake out for itself.

In complete contrast, the opponents’ views embrace sociolinguistic variability. Pattanayak celebrates Odia precisely because the language, in his view, remembers its own folk traditions and sociolinguistic histories. Substantive authority, then, is created by the practices of the populous; the duty of the nation-state is to recognize and support its people. This is a vision of nationalism not unlike that reflected in Gandhian activism or in Rabindranath Tagore’s poetry, a nationalist yet liberal vision based on a reification of “the people” as the site of political authority and legitimacy.

The implication is that the opponents’ views of embodied sounds do not constitute a radical embrace of biological variation. Their approach to official orthography, like their politics, has been structured around normative elite conceptions of sociolinguistic difference—it reflects their own habituated positions as cosmopolitan, regional elites who can write in strong literary English. We can see this most clearly when we consider what the proposed name change would look like and sound like from their critical perspective. When the supporters of the name change propose the Roman script spelling Odisha to produce the closest sounds to “like we say it,” it suggests that the speakers are people who already know how to say those sounds, who already know that the Roman script d stands for the voiced retroflex flap. In other words, it indicates that the proponents’ sonic imaginations of how “English speakers” sound are fully intertwined with their knowledge of English as it is spoken in Odia mouths. It reads as a view that Odia English is unmarked, legitimate English, which in turn indexes a lack of exposure to non-Odia English and a fully provincial sonic imagination. The name change critics, Pattanayak and Satpathy, make a point of this perspective when they repeatedly emphasize that the name poses challenges for “foreign” speakers of English. It is indeed possible that the proponents’ views are a signpost of shifting attitudes toward local varieties of English (see Schneider Reference Schneider2007), or even a political gambit to foist regional speech onto national audiences. But the concerning issue from the perspective of the name change critics is how this new official name indexes provincial ignorance and lack of sophistication. From the critics’ perspective, the new official name enshrines a limited provincial view, especially vis-à-vis “foreign” or global English, not only because it promotes an erroneous pronunciation of the name among outsiders—think Satpathy’s emphasis on the hard stop—but because it turns the state’s official name into a document of the state’s own insularity and backwardness.

This critical imagination of the proposed name change, and the social world that it presumes, provides a different perspective on the embodied world to which the opponents’ orthographic proposals are sensitive. For these English-educated and regional yet cosmopolitan media elites, not unlike the Bengali elites whom Bahri describes, the world is striated by embodied rhematized indexicals that must be constantly managed so that one can be both English and regional, cosmopolitan and loyal. For regional elites, the national audience plays an essential role in this dynamic because, just as in the name change’s justifying discourses, the national audience’s recognition confirms the success of these performances. The opponents’ view of orthography is an instantiation of the very embodied postcolonial semiotic capacities forged through the history of entraining South Asia’s modern elite, from Macaulay’s educational programs, through nationalist movements, to global market competitiveness. It is not only pronunciations that reflect embodied political histories—semiotic ideologies do too.