Reconstructing Heritage Language: Resolving Dilemmas in Language Maintenance for Sri Lankan Tamil...

DOI 10.1515/ijsl-2013-0035 IJSL 2013; 222: 131 – 155

Suresh CanagarajahReconstructing heritage language: resolving dilemmas in language maintenance for Sri Lankan Tamil migrants*

Abstract: In recent years, Sri Lankan Tamils have fled their homeland as refugees as a result of the ethnic conflict in the country. Despite their heightened linguistic consciousness, community elders claim that Tamil youth are turning their backs on their heritage language. My data from Lancaster (California, US), East London (UK) and Toronto (Canada) shows a more complex attitude towards language maintenance by Tamil youth. Though a majority of the youth declared that En-glish was their dominant language of proficiency, they insisted that it did not affect their positive orientation to ethnic identity and community affiliation. They adopted diverse language practices to enjoy in-group identity: namely, code switching into Tamil; emblematic uses of Tamil; switches into Tamilized versions of English; receptive competence in Tamil which enabled them to respond in English; and ritualized practices of communication where they could participate in communicative events with the aid of multimodal resources. These practices suggest that migrant Tamils are treating languages as fluid resources for identity and community construction. The hybrid and multilingual construction of heri-tage language and identity enables Tamils to shuttle between different languages and communities in migrant settings to resolve the dilemmas of mobility and identity.

Keywords: Sri Lankan Tamils; language maintenance; ethnicity; language ideology.

Suresh Canagarajah: Stellenbosch Institute for Advanced Studies, Penn State University. E-mail: [email protected]

* I thank the Stellenbosch Institute for Advanced Studies for a fellowship in July and August 2011 that enabled me to work on this article.

123456789

10111213141516171819202122232425262728293031323334353637383940

(CS6) WDG (155×230mm) DGMetaScience J-2775 IJSL 222 pp. 131–156 IJSL_222_07-0016 (p. 131)AC1:(KN/)24/6/2013 27 June 2013 11:18 AM

PMU:(YCP)11/6/2013

132 S. Canagarajah

“It sometimes boils down to a choice between saving speakers from their economic predic-ament and saving a language.” (Mufwene 2002: 377)

1 IntroductionIn the context of migration and diaspora life, issues of language maintenance become very challenging. As Mufwene (2002) succinctly puts it, migrants from underprivileged backgrounds and speaking non-privileged languages experience the pressure to move ahead socially by learning the globally privileged languages. This is not an easy choice, as the claims of identity and group solidarity are also important. Migrants are all too aware of the negative consequences of abandon-ing their heritage languages.

Such dilemmas have generated a debate among sociolinguists on the place of heritage languages in contemporary contexts of globalization and migration. Scholars are divided into polarized camps on this question. The discussion has been framed in terms of dichotomies such as the following (to summarize the debate between Brutt-Griffler [2002] and Skuttnab-Kangas [2004], for example):– territoriality/mobility– tradition/modernity– ethnic interests / class interests– community solidarity / individual needs– essence/hybridity

For scholars like Skuttnab-Kangas, language maintenance is connected to ecolog-ical and territorial preservation. Loss of languages also means loss of local knowl-edge, which is tied to ecological resources and cultural diversity. However, for scholars like Brutt-Griffler, maintaining local languages means being left out of conditions leading to cultural change and social progress. It is motivated by a con-servative attitude of valuing tradition at the cost of modernization. Brutt-Griffler also points out that the claim for language maintenance prioritizes ethnic inter-ests above class interests. As we know from history, elite groups have denied members of their community access to privileged languages and resources in the name of preserving traditional identities. Thus the elite safeguard their vested interests tied to proficiency in privileged languages. Similarly, there is a dilemma between community solidarity and individual needs. Should one focus on his or her own economic mobility and learn the privileged languages – or affirm com-munity solidarity at all costs by maintaining heritage languages? Such questions are complicated by postmodern orientations to language and identity. Many


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940

Reconstructing heritage language 133

scholars now hold that identities are splintered and languages are hybrid, as communities are in contact (Hall 1997). Therefore, for scholars like Brutt-Griffler, calling for heritage language maintenance implies essentializing identities and languages (see also Edwards 2001). However, Skuttnab-Kangas considers such discourses as having a debilitating effect on the interests of underprivileged com-munities and their languages.

While scholars are polarized along these lines on the question of language maintenance, migrant groups are discovering their own ways of negotiating these dilemmas. It appears that scholars are not granting enough agency to migrants to creatively devise their own ways of resolving their conflicts. It is possible that migrants may develop attitudes and practices that help maintain heritage lan-guages without hindering their socioeconomic mobility. To study such possibili-ties, we need ethnographies that take a closer look at the ways migrants negotiate these dilemmas. While providing knowledge on the strategies of dealing with heritage language maintenance in migrant contexts, such studies can also help re-theorize language and identity in mobility. In this article, I present findings on the attitudes and practices of the Sri Lankan Tamil (SLT) diaspora members in traditionally English-dominant countries, US, UK and Canada, to theorize their ways of negotiating dilemmas in language maintenance.

2 ContextSri Lankan Tamils are among the newest wave of migrants in North American and European metropolises. They started migrating in large numbers after 1983 when the ethnic conflict in the island took a military turn. Language is at the root of the ethnic conflict in Sri Lanka. When parliamentary negotiations to win official status for Tamil language and a fair share in economic resources, employment and higher education failed to produce satisfactory results after independence from Britain in 1948, the Tamils launched a military campaign for a separate state. In 1990 the militant organization LTTE (Liberation Tigers of Tamil Eelam) set up a de facto regime in much of the area in the North and East (treated as the Tamil homeland) which it claimed to have “liberated” from the Sinhala Buddhist government and provisionally renamed Tamil Eelam. There was hectic linguistic activity of both “purifying” Tamil and raising its status in this emergent state (see Canagarajah 1995). Though Sri Lankan government announced in May 2010 that it has defeated the LTTE, the grievances that led to this conflict remain unre-solved, and Tamils continue to leave the country as refugees.

This study started in response to the concerns of elders in the SLT community that their children were shifting wholesale to English in migrant settings. They

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013

134 S. Canagarajah

considered it shocking that the community should lose its heritage language within just fifteen or so years after migration, especially when this mobility re-sulted from the struggle for autonomy for their ethnic identity and language in Sri Lanka. The elders treat heritage language as directly connected to ethnic identity. They therefore hold that a loss of heritage language also portends the demise of the Tamil community identity in diaspora settings. Statements like the following were commonly heard in my interviews:

(1) Ravi: illai, moLiyai, moLi illaaTTi vaNtu niccayamillai ennenRaal, itukaLai paTRiya, enkaTai values inTa tanmaikalai aRincu kolliratukku moLi nalla visayam. atooTai nalla nunukkamaana visayankaLai terincu koLLalaam. atukku moLi kanTippaai mukkiyam enna? naankal identity-ai maintain paNNonum enRaal moLi kanTippaay mukkiyam enna? allaaTTi enkaTai identity onRum irukkaatu.

‘No, for language, things won’t be certain if we don’t have the language because to learn these things, to learn our values and such aspects lan-guage is an important resource. Also, we can learn more subtle things. For those, language is important, isn’t it? If we want to maintain our identity, language is absolutely essential, isn’t it? If not, our identity and other things won’t survive.’

(Male, 28 years, community organizer, Toronto)1

The informant holds that heritage language is critical to the learning and repre-sentation of the community’s values, and that it is impossible to maintain ethnic identity without the language.

I have been visiting three locations in particular – Toronto, London and Lancaster (California) – since 1996 for fieldwork. Currently, according to the esti-mates of local Tamil community organizations, there are about 150,000 Tamils in Toronto (the largest community of Tamils in a single city outside Sri Lanka) and 50,000 in London. While they are scattered in the United States, Tamils form a cohesive community of about 35 families (around 200 people) in a small town

1 I identify the informants by their gender, age, profession (if available), and location. Code switches are in bold. The following transcription conventions are used:

( ) pause of 0.5 seconds or more _____ emphasis= latched utterances ((text)) explanations by researcher[ ] overlapping utterances @ laughter, sustained intonation italics utterances in Tamil ? rising intonation regular English translations. falling intonation bold codemixed/switched items


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940


populated by 121,000 people in Lancaster (about 60 miles north of Los Angeles). Though this is a smaller community (providing fewer subjects for research), I wanted to compare how a fairly close-knit immigrant community that is upwardly mobile in a suburb reflected the trends of scattered settlements of economically more diverse groups in other urban locations.

During extended periods of stay in each location for ethnographic purposes, I have been conducting sociolinguistic surveys to understand the changing pat-terns of bilingualism and interviewing families to explore their language atti-tudes. Interviews and surveys were mostly held in homes, with all the family members or those available present. I have also been conducting interviews of community leaders and focus groups to obtain community perspectives on lan-guage maintenance. The participants were recruited through snowball sampling. In order to cover a balanced range of families according to the diverse socioeco-nomic categories – i.e., caste, class, religion, educational attainment, place of origin in Sri Lanka, length of settlement and area of settlement in diaspora (e.g., inner city or suburbs) – I had to adopt a more proactive approach to soliciting subjects. Such information is not available from the telephone directory, for ex-ample. Furthermore, personal questions were not permitted in my questionnaire as community members resent too much intrusiveness. Some members even re-sented questions relating to their language of dominant proficiency, as declaring Tamil would immediately place them in the less educated grouping. Therefore, information relating to caste, income level, educational attainment and religion was elicited mainly from informal conversations. Based on such information, I solicited a balanced range of subjects for my interviews and surveys. If I had more families from Hindu upwardly mobile professionals in the suburbs in a particular location, I sought Christian working class families from the inner city – and so on.

In addition to observation notes and collection of published material (i.e., newspapers, books, and magazines by the local community), the data set con-sists of the following: (a) A questionnaire on language attitudes and choice, constituting about 20 close

and open questions. It elicits a linguistic profile of the family. In cases where I visited families personally, I filled them out in collaboration with the family members. To make up for families I couldn’t visit, I also distributed my ques-tionnaires through friends and volunteers. See Table 1 for the spread of ques-tionnaires filled by me or anonymously, by families and by individuals.

(b) Interviews audiotaped with families and focus groups. See Table 2 for the distribution of the groups interviewed. Not all the families I visited agreed to be interviewed on tape. Though I followed an interview module, I didn’t resist participants’ preference for a free-flowing conversation. In addition to topics relating to family life (i.e., socialization of children, lifestyle and cultural

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013

136 S. Canagarajah

practices), the interviewees discussed their sense of identity, core values, community life, and inter-community relationships. About 20 hours of inter-views are available from each of these locations.

Though the questionnaire enabled me to quantify the trends and patterns of bilingualism across three generations of Tamils (i.e., grandparents, parents, and children), it is ancillary to my more important objective of gaining an insider per-spective on how the community explains its language choice and attitudes. In this article, I use primarily the interview data from family members, focus groups, and community leaders to understand attitudes to language maintenance and practices of identification.

3 Language shift The survey data on language of dominant proficiency shows that there was indeed a pronounced shift to English. Table 3 describes the profile that emerges.

It is clear that children in all three locations overwhelmingly claim English as their language of dominant proficiency. There is also a generational shift evident in the data. A majority of the parents of these children declare Tamil as their lan-guage of proficiency. There is a complication in the generational trend as more of the grandparents declare that they are balanced bilinguals or more dominant in English than the parent generation. This is explained by the fact that a majority

Table 1: Surveys

From families I personally visited

Filled out for me by individuals and groups

Total

Lancaster 42 (10 families) 10 52Toronto 100 (24 families) 110 210London 110 (25 families) 119 219

Table 2: Interviews

Families Recorded interviews and questionnaires obtained

Informal interviews and questionnaires obtained

Total visited

Lancaster 18 12 10Toronto 12 12 24London 12 13 25


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940


of them were educated in the colonial English educational system. After indepen-dence, Sri Lanka adopted a vernacular education system, in which the parents have been schooled. The greater proficiency in English by the grandparents can be one reason for the pronounced shift of the grand children to English. Unlike other migrant communities where grandparents help in the transmission of heri-tage languages, SLT grandparents switch easily to English to accommodate the preferences of their grand children.

However, we need to understand the limitations of survey data of this sort. First of all, self report data misses a lot of complexity in practice. The respondents may not be clear about what is meant by balanced proficiency, or may not have the ability to compare different types and levels of proficiency. This limitation may also derive from the way questions of this nature are posed in a survey. Sub-jects don’t have the space to make subtle distinctions on types and levels of profi-ciency. More importantly, the survey treats Tamil and English as monolithic lan-guages. It ignores the fact that there are diverse types of hybridity that characterize language use. We have to wonder what “Tamil” and “English” mean in the con-text of multilingual and hybrid language practices. Note how the informant above (in Transcript 1) switches to English in “maintain” in the middle of making a case for heritage language. These subtleties in attitudes and practice come out better in the interview data. However, it is still useful to survey subjects’ views on their relative language proficiency. It is probably trends of this nature that make the elders fear that there is a pronounced language shift and heritage language loss.

Table 3: Language shift in three locations

Location Language of dominant proficiency (%)

Tamil English Both

Toronto Grandparents (n = 55) 41.8 9.1 49.1 Parents (n = 91) 68.1 1.1 30.8 Children (n = 64) 20.6 41.3 38.0

London Grandparents (n = 67) 47.8 7.4 44.8 Parents (n = 51) 66.7 1.9 31.4 Children (n = 101) 13.9 43.6 42.5

Lancaster, CA Grandparents (n = 3) 00 0.0 100.0 Parents (n = 27) 37.1 22.2 40.7 Children (n = 22) 00 95.5 4.5

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013

138 S. Canagarajah

My interview data reveals that the picture is more complex. The youth made a distinction between language and identity. Many answered that even if they couldn’t claim full proficiency in the heritage language (along the ways expected by their parents), they had a strong commitment to ethnic identity. The views of the two university students below, one a male from Lancaster and the other a female from Toronto, reveal this sentiment:

(2) Bama: I don’t think that knowing how to speak Tamil is what makes me a Sri Lankan Tamil per se –

(Female, 22 years, college student, Toronto)

(3) Nishan: But I think, that’s, being a Tamil, as Americans, it is not, it is often, it is always at a point of a contact. It’s not about just concrete steps or concrete values like language and religion and whatnot. I think I can, I mean I could say I can try to be a Tamil without knowing the language, without knowing the religion, without doing any of the customs or traditions, or mores, traditional mores. But still, at the same time, I am Tamil, even if I don’t have any of those things, and I am happy that I am Tamil.

(Male, 22 years, college student, Lancaster)

While Bama simply asserts that heritage language doesn’t have anything to do with her Tamilness, Nishan attempts to rationalize this disjuncture. He first artic-ulates that Tamil identity has nothing to do with objective (or “concrete”) criteria, such as language or religion. He articulates a vision of ethnicity as relational when he says ethnicity is articulated “at a point of contact”. He goes further to surmise that ethnicity can be performed and consciously acted out when he says that he “can try to be a Tamil without knowing the language”. As we can recog-nize, all these possibilities are being theorized by scholars these days. Socio-linguists are moving away from the position that there is a one-to-one connec- tion between language and identity. They question the popular notions that a language always represents a particular identity; that identities and languages are isomorphic; that identities and languages can be essentialized to always repre-sent a particular community or values (see Blommaert 2010; de Fina 2007). How-ever, such sentiments only make us more curious about the exact relationship between language and identity. In my interviews with youth and families, I ex-plored how they perceived these connections. My findings suggest not only new ways of representing Tamil identity, but new orientations to heritage language.


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940


4 Heritage language and ethnicityMy subjects articulated very subtle and complex ways of representing ethnic identity and enjoying community relations even if they couldn’t claim advanced or complete proficiency in Tamil. Many youth told me that they would adopt code switching in strategic ways for identity purposes. The following is the observation of her own practices by a college student in Lancaster:

(4) 1 ASC: With the siblings, with your brothers and sisters, what ((language)) do you use?

2 Vishi: English. Yeah. Sometimes when I curse or when I am mad I call him paNTi ((‘pig’)) or something @@@ and my brother is funny cause he calls my grandmother caniyan ((i.e., playful insult)).

3 ASC: Right right. . . .4 V: With most of my my Tamil friends, I’ll call them paNTi, but that’s

actually their name, a term of endearment almost, you know. Like ‘paNTi come here and show me this please,’ like that. With them it is always Tamil things here and there, especially for inside jokes and stuff. . . .

5 ASC: How do you relate to your Tamil identity? Do you feel, and how do you feel about it. like, are you under pressure to be American, or=

6 V: = No, you know, I’m not like that. I mean I think I tend to be more Americanized anyway, right? But I do, I love my culture. I, I, you know I am really really keen on keeping it, you know even keep giving it to my children and stuff. It’s even in things I want them to have that American people don’t have. Hm kind of I feel something special that I have that they don’t have. . . .

7 ASC: Suppose, when you raise a family would you feel it’s important to, you know, have a place for Tamil? You know, I am just thinking of the second generation of Tamils.

8 V: Yes, I will do some. I mean I’ll teach them all the bad words, @@@ what little I know I’ll try to teach.

(Female, 18 years, student, Lancaster)

Though Vishi refers to playful insult words in this interview, I have observed switches to other lexical items for family relationships, food and cultural and re-ligious practices in other contexts. As we can guess, the use of Tamil in the family domains referred to in line 1 contributes to bonding purposes. The humor and playfulness only add to the bonding effect. Vishi goes on to say that she would use Tamil also with her friends. The code switch contributes to a shared insider

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013

140 S. Canagarajah

identity, indicated by the fact that she refers to these switches as contributing to “inside jokes and stuff” (as she observes in line 4). It is also clear that when such switches are used in mainstream contexts, they have an effect of representing her ethnic identity and showing to her American friends “something special that I have that they don’t have”. What is important is that these single lexical items are adequate to represent (and show her love for) her Tamil ethnic identity and cul-ture. Vishi goes on further to say that these lexical items and activities of switch-ing are significant enough to be passed on to her children. They would constitute knowledge of heritage language and identity for the next generation (as she indi-cates in line 8). The language ideology that emerges here is important. Bits and pieces of Tamil, in the form of code switches, represent ethnic identity and consti-tute proficiency in heritage language for Vishi.

Another practice that emerged was the use of emblematic uses of Tamil for identity purposes. Unlike the previous case, in emblematic uses the lexical items and phrases may not be understood for their literal meaning. They were used purely for symbolic reasons and performative acts. Even if the Tamil words that were used were not understood by the users, they were significant for the func-tions they performed. For example, I found that children in many families would recite teevaarams ‘chants’, bajans ‘hymns’ and prayers memorized in Tamil. They didn’t have the fluency in Tamil to understand what the words meant. They had been written out for them in English and memorized. It is possible that the chil-dren generally understood the significance of these hymns and prayers, though they didn’t understand the literal meaning of every word they uttered. Similar practices were observed in Tamil speech contests and singing competitions. Such events are widely popular in the diaspora. Prizes are awarded for the most effec-tive singing or speech. However, I discovered that the participants had memo-rized the texts written in English. A mother narrated to me the process behind these practices. The mother first enunciated the whole speech or song in Tamil while her sons wrote it down in English; then the sons memorized what they had written down; finally, the mother perfected their pronunciation and coached them on the appropriate gestures and feelings that should accompany the speech or song. The language ideology that emerges from these practices is that perfor-mative acts of emblematic Tamil can index Tamil identity and count as Tamil proficiency.

Some community elders sometimes criticize such emblematic uses and con-sider this language practice inauthentic. Consider the observations of the Direc-tor of a Community Organization that holds classes for heritage language:


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940


(5) 1 ASC: maTRatu ippiTiyum keeLvi paTTan, teevaaram appiTiyaana itukal ellaam English-ilai eLuti paaTiinam enRu.

‘Also, I have heard that such things as devotional hymns are sung from texts written in English.’

2 Bala: English-ilai taan paaTiinam. cila veeLai cila piLLayal, tamiL-ilai nallaa first prize, first prize eTukkunkal, English-ilai paaTamaakki taan peecuNkaL. appa atellaaam imitation taan, enna? Imitation taan.

‘It is in English that they are singing. Sometimes, some children, they get first prize in Tamil, they get first prize, by delivering their oratory after memorizing it from English. So those things are all im-itation, isn’t it? It is all imitation.’

(Male, 50 years, Director of Community Organization, Toronto)

Though the informant considers this practice “imitation”, as the children are simply imitating their parents who coached them or going through the motions, the fact that such practices are widespread suggests another perspective. For many families, these practices are sufficient to display their children’s proficiency in heritage languages, pride in ethnic identity, and commitment to the commu-nity.

Diaspora youth didn’t count on always using Tamil lexemes for ethnic iden-tity and community relations. Their identification practices were facilitated by codes from other languages too. For example, I found that some used Sri Lankan English to establish in-group identity. Consider the following example in an East London home:

(6) 1 Father: Son, Dr. Canagarajah wants to interview you about something.2 Danny: Hi, Uncle. Do you want to sit here or go to that table over there?(Father: male, 50 years, teacher; Danny: male, 16 years, student; London)

“Uncle” is a Sri Lankanism used for older members of the community, not neces-sarily those who are blood-relatives (Meyler 2007: 274–275). Though Danny rarely used Tamil words to establish identity during the whole conversation, items from Sri Lankan English served to build in-group identity with me. Danny resorts to Sri Lankan English tokens as he is less proficient in Tamil than most other youth I met.

I found that some SLT youth can make subtle distinctions between varieties of English for the purposes of identity representation. In the following example, the interviewees used Sri Lankan English to establish their common SLT dias-poric identity. Sisters Nila and Vatha belong to the “1.5 generation” and are

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013

142 S. Canagarajah

slightly more proficient in Tamil than their peers. They described how their friends take up traditional dance and music for display of identity, though they are not proficient in Tamil language. To mark their own difference and establish identity with me, the sisters switched to Tamil in some statements. Though they were very proficient in British variants of English (note the modal “done” in line 7 below) and had a marked British accent, they also switched to Sri Lankan English at times:

(7) 1 Nila: Some girls, they are like, dance teriyum, viiNai teriyum- viiNaa. ‘Some girls, they are like, they know dance, know violin- sitar.’ But they don’t know tamiL= ‘But they don’t know Tamil=’2 ASC: =language? I see. Yes, that’s very common. When I asked, they said

they want to know something of Tamil culture. So they go for dance classes and all that.

3 Vatha: Yes.4 ASC: But some people said they do it only because their parents force

them to do it. Do you feel that way? Or are they happy doing that?5 Vatha: Some people. They just tell.6 Nila: But Abi- Abi likes to do. ( ) 7 Vatha: So like, me and my friends, they all love dance and they want to do

it. But there are certain people. They have done too many things when they were in a young age. And they don’t like it. Those who follow it through, it’s not bad, they like it. Like my friend, she done violin and viiNaa ((‘sitar’)), goes to the temple, she prays, she has all the Tamil values, but she can’t speak Tamil. There are people like that. Who are quite changed and they can’t speak to anyone.

(Vatha: female, 16 years; Nila: female, 12 years; students, London)

Note the peculiar structures: “they just tell” in line 5, which means “they profess something without meaning it”; “Abi likes to do” in line 6, which doesn’t have an object; and “when they were in young age” in line 7 is a literal translation from idiomatic Tamil. These usages may be motivated partly by the reason that I was using English as my code with them, assuming that they were more comfortable in that language. Even within English, therefore, the sisters are able to use a vari-ant that would establish SLT identity with me

In an interaction I had with brothers Kannan and Nannan in their Toronto home, it became dramatically clear that English items are perceived as markers of Tamil identity for speakers who adopt them for in-group purposes. The brothers had made a distinction with the children in their neighboring house who had re-


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940


cently immigrated to Canada and spoke what they labeled “broken English”. The brothers expressed pride in the fact that their family spoke only Canadian En-glish. While we were talking, we overheard the mother using a Sri Lankanism with an acquaintance on the phone. I queried the boys as to why their mother used a Sri Lankan English item if the family spoke only Canadian English:

(8) 1 ASC: naan unkaTai mother ippa peeceekai kavaniccan avankaL cila words paaviccaankal, Canadians paavikka maaTTaankaL. He can come at nine, no? enRu keeTTaa. No enRatu inka peeca maaTTaankaL.

‘I was observing as your mother was talking just now. She was using certain words which Canadians won’t use. She asked, “He can come at nine, no?” People here won’t say “no” like that.’

2 Nannan: atu atu tamiL tamiL. ‘That, that is Tamil, Tamil.’3 ASC: yaaru? ‘Who?’4 Nannan: atu tamiL. ( ) ‘That is Tamil.’5 ASC: I see, Sri Lankans oTai peecina paTiyaal appiTi peecinaankaL?

I see. ‘I see, she spoke like that because she spoke to Sri Lankans?

I see.’6 Nannan: naankaL annaa colla maaTTam. annaa, tankacci, peTTai

tankacci, cittaa. Canadians-ooTai. aanaaL, I use with my people.

‘We won’t use words like elder brother. Elder brother, younger sister, female sister, cousin. (We won’t use them) with Cana-dians. But I use with my people.’

(Kannan: male, 12 years; Nannan: male, 10 years; students, Toronto)

Initially, I failed to understand Nannan’s explanation when he said “That is Tamil” (line 2). The “no” particle is clearly not a Tamil word. I belatedly realized that the expression was used solely for Tamil interlocutors. He confirms this in line 6 when he illustrates how he won’t use Tamil kinship words when he talks to Anglo-Canadians. He suggests that “no” is, similarly, used with Tamil people (“my people”). In connecting the Sri Lankanism with Tamil words, Nannan sug-gests that he is associating “no” with Tamil values and identities. In effect, it is not an English word for him (as he associates “standard” English with Canadian identity). Like those Tamil relationship words which he uses only when he talks

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013

144 S. Canagarajah

“with my people”, this English expression too is an in-group item. It can be in-ferred that Nannan has processed his mother’s expression as an analogous switch to in-group language. It is clear that Sri Lankan English indexes Tamil values, in-group relations, and identity for these youth.

Other tokens of Sri Lankan English (SLE) that came up were borrowings such as Amma and Appa used interchangeably with ‘Mom’ and ‘Dad’, respectively, when the situation warranted the switch; relationship terms like Anna and Akka for ‘elder brother’ and ‘elder sister’, respectively (in this case, English proficient speakers have to use these terms to mark respect, as English doesn’t have an equivalent for these kinship terms); terms of endearment such as machang ‘buddy’, kunchu ‘darling’, and cultural borrowings such as curry, sarong, sari, pottu (i.e., the mark women wear on their forehead), and maaLu paan (i.e., ‘fish bun’) – which all appear in the Sri Lankan English dictionary – see Meyler [2007]).

Interestingly, some of these words (i.e., machang, maalu paan) came into SLE through the other Sri Lankan local language, Sinhala. In using these words, it is possible that Tamils are still constructing an in-group identity by drawing upon resources from Sinhala. Sinhala is a rival language in the homeland, fiercely re-sisted by Tamils for their language rights. However, outside in the diaspora, espe-cially for youth who don’t have feelings of traditional animosity, it might serve to construct an in-group identity. The youth don’t think of these Sinhala tokens as necessarily alien, but part of the SLT identity. Old distinctions get erased and new associations get constructed in the diaspora. When Sinhala words get reified with Tamil identities and in diaspora social practice, they begin to stand for in-group solidarity and identity. The language identity that emerges through such practice is paradoxical: Tamilness can be indexed through multilingual resources.

More complicated in the ethnic identification practices of the SLT youth are acts of reception in conversational interaction. I found many contexts of Tamil conversation where youth use their passive (receptive) competence to understand Tamil and participate in Tamil conversations while using English for their re-sponse. The sisters Nila and Vatha (whom we met in Transcript 7), also told me that they would use Tamil with their less-proficient friends who respond in En-glish. She considered it “kind of fun”. Even if these friends use non-SLT tokens to respond then, they are able to enjoy in-group membership through their receptive competence of Tamil. It appears as if the production of Tamil language tokens are beside the point in performing ethnic identities and relationships. Communica-tive practices help SLT youth enjoy in-group identity. In the following extended conversation, Rajani uses her passive competence to participate in a conversation with her mother, her uncle, and me, a conversation which the elders conducted largely in Tamil:


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940


(9) 1 ASC: makaL tamiL kataikka maaTTaa enRu worry paNNuriinkaLoo? ‘Do you worry that your daughter can’t speak Tamil?’ 2 Mother: worry paNNi enna, onRum ceiya eelaatu enna? viLankum taanee

avavukku? ‘We can’t achieve anything by worrying, can we? Isn’t the fact

that she understands enough?’3 Uncle: caappaaTu atukaLilai she is more. ‘She is more ((conservative)) in food and things like that.’4 Mother: oom. maTRatu paavaaTai caTTai atukaL pooTa konca naaL

pooTaamal iruntava, piraku orumaatiri friends aakkalooTai ceerntu skirt pooTa atukaL itukaL ellaam viruppam.

‘Yes. She was earlier disinterested in wearing skirt and things like that. After joining some friends, she now likes to wear skirt and stuff.’

5 ASC: Temple atukaL ellaam? ‘What about temple and things like that?’6 Rajani: Not regularly, but yeah.7 ASC: Hmm would you maintain Tamil culture and language in the

future?8 Rajani: Culture, culture yes. I can’t see myself speaking in Tamil now,

because I think I have left it too late. Xxx it’s hardly useful, to choose between English, so.

9 Mother: pirayoosanam illai enRu colluvaa xxx @@@ ‘She says it is useless.’10 Rajani: No I am just saying. It’s the truth though. You don’t speak so

much in Tamil, so.11 ASC (to the mother): niinkaL muyaRci eTukkira niinkaLoo? ‘Do you take any effort?’12 Mother: avavai tamiL paLLikuuTattukku anuppina naankaL. vaacikka

eLuta teriyum. peecuraTu taan karaiccal. viLankum, ellaam viLankum.

‘We did send her to Tamil schools. She knows how to read and write. It is speaking that is difficult. She understands, she under-stands everything.’

13 ASC (to Rajani): So is it simply because you are shy or –14 Rajani: It’s because I have lost touch. I used to speak at a very young age.

It became less later, and I switched more to English. But you know they are trying to get me to speak Tamil again, and I used to try, and with my accent it sounded funny. So my relatives

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013

146 S. Canagarajah

tended to look down at me, so that’s why I couldn’t be bothered after that, you know.

(Mother: female, 45 years; Uncle: male, 54 years, teacher; Rajani: female, 16 years, student; London)

In choosing to interject meaningfully and strategically in contexts where the mother, uncle, and I talk in Tamil, Rajani is choosing to display her Tamil identity and establish community with us. As the mother apologizes for Rajani’s lack of Tamil proficiency by saying that she adopts other cultural practices in food and dress, I ask the mother if her daughter goes to the temple (in line 5). Rajani chooses to answer herself, and gives a more qualified response, as her mother and uncle seem to have exaggerated her adoption of Tamil practices in the previ-ous turns. Rajani shows that she has processed the Tamil conversation to inter-vene critically in English. Note that though I had myself used a borrowing (“temple”) which Rajani would have understood, she wouldn’t have understood the force of the question or the context of the utterance by that word alone. She should have understood the previous sequence of utterances in Tamil in order to answer meaningfully in English.

Rajani responds again in line 10 when the mother says that the daughter doesn’t find Tamil language useful. Rajani clarifies her reasons for saying so, in-sisting on her position. As we (the elders) had laughed at the mother’s character-ization of her daughter’s explanation, Rajani sees a need to clarify. Note, how-ever, that in providing an explanation in Tamil for her daughter’s answer in English (line 9), the mother assumes that her daughter will understand the Tamil statements. In other words, the mother and the uncle are assuming that they can carry out this conversation multilingually.

Later, in line 12, the mother states that her daughter understands Tamil and that is satisfactory to her. I ask Rajani why she doesn’t speak in Tamil if she under-stands the language (line 13). Rajani’s explanation (line 14) might also suggest the reason she doesn’t use Tamil tokens as other youth do. She fears that she will be stigmatized for her pronunciation. This perhaps motivates her to represent her ethnic identity through other communicative practices. Note that at this stage of the conversation, I too have started assuming that Rajani understands Tamil and will respond in English to sustain this polyglot dialogue. After the mother’s contribution in Tamil, I pose a follow-up question to Rajani in English (line 13), assuming that she would have understood what her mother had said earlier in Tamil (line 12). I thus piggyback on the preceding Tamil utterance, and ask a question from Rajani in English. Though I am new to the realization, the mother and uncle seem to have assumed throughout that Rajani is part of the conversa-tion taking place in Tamil. Their assumptions and practices suggest that such


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940


conversational strategies are everyday reality for such families. This is how family conversation takes place in many diasporic homes. The language ideology that emerges through this practice is noteworthy: receptive proficiency counts as com-petence in heritage language, and indexes ethnic identity.

I also found in many contexts that the youth didn’t require any language tokens to “perform” their ethnic identities or maintain in-group relations. They could participate in communicative events through ritualized practices that also depend on multimodal communicative resources. For example, the youth could follow religious, social, and cultural rituals through a knowledge of the discourse conventions. When they go to the temple for pooja, they could participate in the rituals, going up to the priest at the appropriate time for example, through famil-iarity with the components of the ritual and by watching others in the event. Christian youth would participate in a Tamil worship service similarly – they stand up with others for the reading of the scripture, go up for communion, and kneel for prayers without having to understand every Tamil word in the service. At wedding ceremonies, they would participate in the rituals through familiarity with the different stages in the ceremony. A young man told me that he could adopt multimodal resources also to participate in conversations:

(10) Niranjan: =sometimes I do [understand]. I can understand Appa and Amma. But I don’t understand anybody else. Like basics. If it’s something familiar. But sometimes I use the body language as well, than the language.

(Male, 22 years, medical student, Lancaster)

Though he had earlier said that he is not proficient in Tamil, Niranjan goes on to say in this interaction how he does participate in conversations with his parents by reading their body language. Many parents were happy to accept these prac-tices for purposes of family and community bonding, and treated them as their children’s proficiency in heritage language. The language ideology that emerges through these practices is even more surprising: multimodal communication and participatory practices in community events count as communication in heritage language and representation of ethnic identity.

5 Implications for heritage language maintenanceThe findings in this study convey the importance of shifting to “a sociolinguistics of mobile resources” from “a sociolinguistics of immobile languages”, as Blom-maert (2010: 43) has recently argued. This paradigm shift requires two major

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013

148 S. Canagarajah

changes in our orientation to language. Firstly, we have to think of “language” as made up of diverse symbolic resources. That is, languages are not separate mono-lithic wholes. Communication involves resources from diverse languages and multisensory symbol systems. This activity of borrowing codes from diverse sys-tems for one’s communicative needs has also been called translanguaging by other scholars (see Canagarajah 2011; Garcia 2009). It is from this perspective that we can understand how switches, emblematic uses, and borrowings from diverse multilingual codes can serve to construct identities for SLT youth. In a related move, Pennycook (2010: 142) has argued that we should go beyond language re-sources to treating communication as involving practices: “discourse, genre, and style, when viewed in terms of practices, direct our attention to different ways in which we achieve social life through language”. It is this perspective that will help us address how receptive competence, multimodal resources, and ritualized practices can be communicative and contribute to significant identification prac-tices even when some of my subjects don’t have the competence to produce Tamil language tokens.

Heeding such recent calls for sociolinguistic reorientation will also enable us to address how language resources function in mobility. Semiotic resources take new values and indexicalities when they are uncoupled from their rootedness in monolithic languages, communities, and territories. This theoretical orientation would help demystify the traditional equation of “1 language = 1 community”. This isomorphic relationship is what motivates community elders to think that a loss of Tamil language would mean the loss of ethnic identity. As Hymes has ob-served, “The relations between the units classified by linguists as languages, and the communicative units of ethnology, are problematic” (1968: 30). If we are pre-pared to adopt this orientation, we can see how resources from diverse languages are able to index new values and represent new identities in mobility. We can then address how resources from Tamil can be used in new ways for communica-tive objectives and multilingual resources may come to represent Tamil identity. More radically, we can understand how resources from diverse languages and semiotic modes can still be treated as constituting “Tamil language” in diaspora contexts.

We will now move to answer the questions posed by SLT diaspora community elders: is the heritage language dying? And is ethnic identity lost with the heri-tage language? What we find from the data is that while ethnic identity is alive and well, perhaps defined and practiced in novel ways, the heritage language is also going through redefinition. Tamil identity and ethnic relations in the dias-pora can be constructed through multilingual resources and multimodal prac-tices. An obvious question that then arises is: What is Tamil – and what consti-tutes proficiency in Tamil? In diaspora life, Tamil includes considerable mixing of


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940


English and other languages. Emblematic and ritual uses constitute knowing Tamil. Also receptive proficiency constitutes Tamil knowledge. It appears that heritage language is not lost; it has simply been redefined with new values for a different time and place. Mobility of the community has resulted in the recon-struction of heritage language for diaspora contexts with new indexicality.

It is clear that the way Tamil is defined and practiced in diaspora is not simi-lar to that in the traditional homeland of Sri Lanka. In Sri Lanka, especially in the context of heightened linguistic nationalism, Tamil proficiency is treated as es-chewing any mixing from other languages. It involves full proficiency in receptive and productive skills, in both oral and literate modalities. Language attitudes and policies in the Tamil homeland are currently influenced by the dominant ideolo-gies of “Tamil Only” and “Pure Tamil” (see Canagarajah 1995). It is clear that the migrant community has detached itself from these ideologies in diaspora con-texts to define Tamil language and proficiency in new ways.

A reconstructed heritage language may sound like an oxymoron. It may appear according to traditional assumptions that such a primordial construct as heritage language should remain unchanged and pure. People assume that heri-tage language should stand stable across time and space to represent community identity and preserve its grammar (as the subject in Transcript 1 expressed). However, many scholars in applied linguistics are becoming comfortable with the possibility that heritage language cannot be essentialized. Agnes He (2010: 77) argues in a recent state-of-the-art essay, “HL [heritage language] is not static but dynamic; it is constantly undergoing transformation by its learners and users, so that at the same time it serves as a resource for the transformation of learner iden-tities, it is also transformed itself as result of learners’ and users’ language ideol-ogies and practices”. Part of this reconstruction process is that, while acquiring new values and indexicalities, the heritage language also colonizes resources from other languages and communities and makes them part of its corpus. In this sense, resources from English, Sinhala, and other languages in contact situations can get redefined as Tamil and get accommodated into the heritage language by its users, as we see in the case of the SLT diaspora.

To further demystify the notion that heritage languages are static and mono-lithic, it is possible to show from the history of the SLT community that heritage language and identity have always been changing. To recount recent history, when SLT community members encountered British colonial forces from the mid 18th to the 20th century, they developed the attitude that learning English didn’t affect community solidarity. Remaining Hindu was what preserved ethnic identity for them at that time. In other words, it was religion not language that constituted ethnic heritage. Therefore, some of the Hindu reformists established Hindu schools for English education. These schools adopted the English medium,

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013

150 S. Canagarajah

and there was a healthy multilingualism in the construction of Hindu identity and practices. Language mixing, therefore, didn’t create a problem for heritage language, ethnicity, or religious identity at that time. However, after decoloniza-tion in 1948, when the SLT community encountered what it perceived as Sinhala colonization – which found most direct expression in terms of language, as Sin-hala was declared the official language of the country – the Tamil community developed a different language ideology. It favored the policies of Tamil Only and Pure Tamil. At this point, the community reacted against English as well. Mixing of not only Sinhala, but English and even Sanskrit, was censored.

As times and places have changed further, migrant Tamils have constructed a different language ideology for the 21st century in diaspora contexts. Now it is not religion or language that constitutes the core values. Practices are now de-fined as more important. What makes one Tamil is the ability and willingness to participate in community life through social, multimodal, and multilingual prac-tices. One’s cultural values, religion, or language proficiency don’t violate one’s lingua-cultural identity, as Niran explained earlier (in Transcript 2). In this defini-tion, language too is defined as a form of practice. Tamil language also accommo-dates diverse linguistic and symbolic resources, and gets defined in heteroge-neous terms. It is not full and total grammatical proficiency in all skills, or even an understanding of literal meanings of every word used, that define knowledge of Tamil. One’s ability to “perform” in Tamil by strategically employing its sym-bolic resources for one’s purposes constitutes maintenance of heritage language.

It is evident from my observations that families and community organiza-tions are developing heritage language pedagogies based on these assumptions. Wherever I went for my interviews, parents would call up their children to demon-strate that their children “knew” Tamil and maintained their heritage language. What transpired was a display of their children’s codeswitching abilities or em-blematic uses. Many parents consciously adopted a practice-based orientation to intergenerational language transmission. Such contexts are often framed as learning-and-display situations where no attempt is made to develop full compe-tence in the language. Displaying appropriate tokens to represent Tamil identity is the main objective. Parents call up their children in front of guests to display their Tamil knowledge, in effect providing an opportunity to perform their iden-tity. Both purposes, learning and performance, are often fused as in the following case. While I was interviewing the parents in East London, the father called his 7 year old son, Sinju, who was playing outside with some friends, to display his Tamil knowledge. Since there were three boys speaking in English as they played together, I used English initially to ask who Sinju is:


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940


(11) 1 ASC: Who is Sinju? 2 Sinju: naan taan sinju. ‘I am Sinju.’3 Siva: paattiinkaLaa? avan tamiLiLayee colliiTTaan. ( ) ‘Did you see that? He said even that in Tamil.’ appaavaa niinkaLaa keTTikaaran? ‘Is your dad or you cleverer?’4 Sinju: I don’t know.5 Siva: ( ) ah?6 Sinju: I don’t know.7 ASC: ettinai vayasu unkaLukku? ‘How old are you?’8 Sinju: Seven.9 Siva: Seven-ukku tamiL illayaa? ‘Is there no Tamil word for seven?’10 Sinju: eeLu. ‘Seven.’(Siva: Male, 45 years, engineer; Sinju: Male, 7 years, student; London)

The conversation moves between a display of the son’s Tamil proficiency, mean-ingful information for the guest, and teaching/learning. The father begins by gloating how his son had answered my English question in Tamil (line 3). Though this was a genuine information-seeking question, the father’s following display question (i.e., whether he or his son was more clever) receives an ambiguous answer. The son’s reply in English (line 4) can mean he doesn’t know how to answer that in Tamil or he can’t say who is cleverer. The father asks for clarifica-tion (line 5), but later drops the question, and doesn’t insist on Tamil usage. In this case, it appears as if the father recognizes that passive competence is good enough. He probably realizes that the Tamil phrase (enakku teriyaatu) is a bit difficult for the son, given his level of language proficiency. My question about age (in line 7), however, requires an unambiguous answer. So the father insists on a Tamil reply. In all this, the elders ask questions that are within Sinju’s profi-ciency level for both comprehension and production. Occasions like this are quite common in diaspora homes. No one in these speech events is claiming or aiming at advanced proficiency in Tamil. Parents and children are satisfied with code switching and emblematic uses for ethnic identity display and in-group bonding. More importantly, this performance qualifies as proficiency in heritage language.

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013

152 S. Canagarajah

6 Agency in language maintenanceWe can now return to the way in which the approach to heritage language dis-played by my subjects resolves their dilemmas and answers the debates posed by sociolinguists. The orientation to language as mobile resources and the construc-tion of a pragmatic language ideology enables SLT youth and families to move on with their new life in the diaspora setting while holding on to ethnic identity and community life. In a sense, they are able to have it both ways. They can both de-velop new identities and symbolic resources for life in the West, and also main-tain heritage language for ethnic solidarity. They probably realize that holding on to the language ideologies, definitions of proficiency, and language norms in the SLT homeland will be counter-productive to their needs and interests in the changed time and location of diaspora.

The attitudes and practices of the community provide another way to ap-proach the scholarly debates on language maintenance. While scholars fero-ciously debate the seemingly irreconcilable claims of tradition/modernity, com-munity/individual, ethnicity/class, essentializing/hybridity, and territoriality/mobility, the migrants adopt creative strategies to resolve these dichotomies in their favor. They are able to address their personal interests of socioeconomic mobility by constructing hybrid identities, without abandoning affiliation with their heritage language and ethnic community. They are able to be mobile with-out abandoning inherited identities and community solidarity. They accomplish this feat by constructing ideologies of language and ethnicity that are flexible enough to let them shuttle across spatiotemporal contexts and communities easily. We can understand this possibility only if we understand identities and community membership as situational rather than static. It is not impossible for people to adopt one set of identities, values, and language practices in one space/time, while adopting others (often conflicting ones) in other contexts.

This perspective on language maintenance also helps us adopt a critical ori-entation to other sociolinguistic constructs relating to globalization and mobility. While the orientation to language as mobile resources enables us to address how heritage languages acquire new indexicalities in new spatiotemporal contexts, the orientation to these practices as “fragmented and ‘incomplete’ – ‘truncated’ – language repertoires” (Blommaert 2010: 9) leads to some simplification. Though it is true that a mixture of codes from different systems and different symbolic practices constitute the identification practices and competence of the SLT youth, it acquires a wholeness in the context of their language ideology. The communi-ty’s language ideology accommodates this linguistic hybridity and gives it coher-ence. To consider this proficiency as “fragmented” is to fall into the trap of mono-lingualist ideologies.


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940


This perspective on language maintenance also compels us to adopt a more positive orientation to language ideologies. In current sociolinguistic scholar-ship, language ideologies have been critiqued for constructing language regimes for mostly limiting and suppressive purposes. In order to develop a notion of lan-guage as mobile resources, sociolinguists have had to deconstruct language ide-ologies that totalize language systems. Therefore, Pennycook has pointed to the limiting ideologies of linguistic science: “Languages, as described by linguistics and applied linguistics, are inventions of the disciplines that make them” (Penny-cook 2010: 129). Similarly, there are other language ideologies that have been cri-tiqued in recent literature. Pennycook has also criticized missionary ideologies that invented languages in colonial contexts to suit their purpose. Blommaert (2010) has critiqued nation-states and commercial agencies for constructing lan-guage ideologies that serve their political and marketing interests, respectively. Such criticism may give the impression that language ideologies only serve pur-poses of false consciousness and manipulation. That is of course the classic Marxist perspective on ideology. However, poststructuralist perspectives have opened up the possibility that ideologies are also illuminating, enabling, and em-powering (see Canagarajah 2010). The more important question is: which group and what purposes are we addressing when we consider the functions of lan-guage ideologies? For the SLT youth and families, the construction of a pragmatic language ideology helps resolve their dilemmas in diaspora life. From this per-spective, there is nothing practical like an ideology. This is not to say that there is no contestation of this ideology. As we saw in the comments of some community elders (in Transcripts 1 and 5), some still hold on to the language ideologies of the homeland. They consider the practices of the youth inauthentic and insincere “imitation”. However, there is an emergent language ideology in the diaspora context that helps the younger generation deal with their social challenges stra-tegically. Even among the elders, there is a tension between their professed and implicit ideology. Though they are not always willing to accept the emergent pragmatic language ideology as satisfying, in practice they are reconciled to the new identification and communicative practices of the youth and community members in everyday life (as we in the mother and uncle of Rajani in Transcript 9, and the father in 11).

7 Pedagogical implicationsFrom this perspective, we have to ask if there is any value in teaching heritage languages in diaspora contexts. What are the pedagogical implications of this study? I do see value in heritage language education. However, it has to be con-

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013

154 S. Canagarajah

ducted with a new orientation. We have to teach Tamil in order to provide more semiotic resources for youth to fashion ethnicity and community solidarity in novel ways. In this kind of pedagogy, there is no need to aim for “full” or ad-vanced proficiency in heritage language, such as the ability to both speak and write or form long stretches of utterances. What diaspora youth need is the capa-bility to draw from Tamil symbolic resources for their multilingual communica-tive practices. They need the resources for performative practices such as code switching, crossing, and styling. From this perspective, the current pedagogy of teaching Tamil through English medium in many diaspora contexts is not un-acceptable. Though teachers are often unhappy with this pedagogy, as it goes against the long established “direct method” approach, such a pedagogy pro-vides practice to students on the hybrid language practices of diaspora life. In effect, this pedagogy facilitates the multilingual construction of ethnicity and communicative life.

Beyond formal considerations, teachers have to also work on ideology and attitudes. Teachers have to facilitate language awareness among student to help them understand the effects of changing spatiotemporal contexts on indexicality. Diaspora youth have to learn that heritage language doesn’t remain unchanged as it travels across time and space. Apart from responding to the changing index-icalities of the language, students can be proactive and reconstruct the heritage language by accommodating newer semiotic systems and practices. If students hold non-essentialist and constructivist orientations toward language, they will appropriate resources from other languages to reconstruct heritage language and ethnic identity. They also have to develop the pragmatics and discourse strategies to shuttle across contexts, and acquire new language resources and identities, without jettisoning their linguistic and cultural heritage.

ReferencesBlommaert, J. 2010. The sociolinguistics of globalization. Cambridge: CUP.Brutt-Griffler, Janina. 2002. Class, ethnicity, and language rights: an analysis of British colonial

policy in Lesotho and Sri Lanka and some implications for language policy. Journal of Language, Identity, and Education 1(4). 207–234.

Canagarajah, A. Suresh. 1995. The political-economy of code choice in a revolutionary society: Tamil/English bilingualism in Jaffna. Language in Society 24. 187–212.

Canagarajah, A. Suresh. 2010. Ideology and theory in second language writing: a dialogical treatment. In T. Silva & P. K. Matsuda (eds.), Practicing theory in second language writing , 176–190. West Lafayette, IN: Parlor Press

Canagarajah, A. Suresh. 2011. Translanguaging in the classroom: emerging issues for research and pedagogy. Applied Linguistics Review 2. 1–28.


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940


De Fina, Anna. 2007. Code-switching and the construction of ethnic identity in a community of practice. Language in Society 36. 371–392.

Edwards, John. 2001. The ecology of language revival. Current Issues in Language Planning 2. 231–241.

Garcia, Ofelia. 2009. Bilingual education in the 21st century: a global perspective . Oxford: Wiley-Blackwell.

Hall, Stuart. 1997. The local and the global: globalization and ethnicity. In A. D. King (ed.), Culture, globalization, and the world system , 19–40. Minneapolis, MN: University of Minnesota Press.

He, Agnes. 2010. The heart of heritage: sociocultural dimensions of heritage language learning. Annual Review of Applied Linguistics 30. 66–82.

Hymes, Dell. 1968. Linguistic problems in defining the concept of “tribe”. In J. Helm (ed.), Essays on the problem of tribe , 23–48. Seattle: American Ethnological Society & University of Washington Press.

Meyler, Michael. 2007. A dictionary of Sri Lankan English. Colombo: Samayawardhana Printers.Mufwene, Salikoko. 2002. Colonization, globalization, and the plight of “weak” languages.

Journal of Linguistics 38. 375–395.Pennycook, Alastair. 2010. Language as a local practice. London: Routledge.Skuttnab-Kangas, Tove. 2004. Do not cut my tongue, let me live and die with my language: a

comment on English and Other Languages in relation to linguistic human rights. Journal of Language, Identity, and Education 3(2). 127–134.

123456789

10111213141516171819202122232425262728293031323334353637383940


PMU:(YCP)11/6/2013


PMU:(YCP)11/6/2013

12345678910111213141516171819202122232425262728293031323334353637383940

Reconstructing Heritage Language: Resolving Dilemmas in Language Maintenance for Sri Lankan Tamil...

Documents

Transcript of Reconstructing Heritage Language: Resolving Dilemmas in Language Maintenance for Sri Lankan Tamil...