Ancestry and Ethnic Identity in the Australian Census… and thus to Sri Lanka

The 9th of August was census night in Australia. The census form has three boxes relating to “Country of Birth,” one’s “language other than English at home” and “Ancestry”— all interesting formulations that bear on one’s ethnic subjectivity and one’s explicit identity.

Ethnicity is a complex phenomenon that is nourished over the years by the influence of many factors. Ethnic self-perception is always inter-relational and thus inter-subjective.[i] It can rest lightly on some and weigh heavily on others. Ethnic terminologies deployed in official domains and brought into everyday speech are among the factors that mould these self-perceptions. As the subaltern historians of India have revealed, census making and bureaucratic categorization in the everyday world had a considerable bearing on the shaping of ethnic identities from the colonial period onwards.

Placed within this introductory note let me refer to the decision taken by the Australian Bureau of Statistics to seek information not only on the “Country of Birth” in Question 12, but also to pose Question 16 as “Does the person speak a language other than English at home?” and press forward with Question 18 on “Ancestry.” It is the last that is the most significant and can generate both amusing and confusing results.

Question18 is posed thus: “what is the person’s ancestry”… followed by several boxes below catering to specific ethnic identities:

Other – please specify…

with four lines left for this last miscellaneous section.

Given the individualistic epistemology that prevails in Australia and the West, clearly, most people will adhere to their self-perceptions in answering this question. It is likely that quite a few Australians whose arrival dates back to the nineteenth century would mark their Irish, English, Scottish or Welsh ancestry, even though they are ardent Australian patriots whenever, say, the Aussies face England or Ireland at soccer or cricket. Even those of Irish lineage who do not have any interest in hurling and are ardent fans of footie and root for Collingwood or Geelong may well indicate their Irish ancestry/heritage. Again, speculatively, I surmise that quite a number of second- and third-generation Italian migrants would tick the box “Italian,” but a few may well disclaim their Italian-ness and insist on being Australian and Australian only.

Let me add here that when the Socceroos played Italy in the World Cup a few years back a significant number of Italian-Australians were cheering for their homeland, though others were ambivalent; and even among those shouting for the Italian team there was a degree of affinity for the Australian side.

This has not been so for those Sri Lankan Tamil migrants in Australia who have been partial to the LTTE or expressed their fervent attachment to Tamil-ness in recent times during the emotional moment attending the collapse of the LTTE military enterprise. When the Sri Lankan cricket team faced Australia in late 2010 (and now when they confront each other in Sri Lanka) most of these migrants are said to be partial to Australia; while some express this Aussie patriotism and/or hostility to Sri Lanka with considerable fervour.[ii] These acts of vociferous expression can be interpreted as acts of retribution inspired by their sense of humiliation and bitterness after the defeat of their LTTE talaivar, Velupillai Pirapāharan, and the demise of the Eelam project.[iii] The expression of such sentiments, of course, is an exercise of their democratic rights and one cannot object to their actions (unless they peddle lies).

From within the same consequences of war and its setbacks, the Australian Tamil Congress has stepped forth to advice Sri Lankan Tamil residents in Australia how they should fill their census forms.[iv] Tamils are requested to fill Question 12 as “Ceylon” and refrain from any reference to “Sri Lanka.” They are advised to say “Tamil” to Question 16 (it is not clear to me if this answer was to hold even where English is the predominant mode of exchange at home).

The more significant advice is with reference to Question 18 on Ancestry. They offer the following advice: “at the last census (in 2006) all the Tamils who responded to this question as Sri Lankan have been counted as Singhalese. This question is not about country of birth. Ancestry does not depend on the place of birth. it relates to heritage. Australian politicians consider the number of voters impacted when they evaluate any policy matters. Tamils have a long and colourful heritage and we should be proud to be part of it.”

No one can deny the richness of the Tamil heritage. It is the associational activity seeking to expand the clout of the Tamil lobby in the Australian political circuit that I highlight as a significant fact (again an exercise of democratic rights). But I am also interested in what sort of statistical results will emanate from this process. How does the Bureau of Statistics compute the results of data conveyed in Question 18? And is the Tamil Congress’s allegation that those who said they were “Sri Lankan” in the previous census of 2006 were thereafter transformed into “Sinhalese” in the census enumeration a valid claim?

The further problem here is that Australia contains a number of people of Tamil heritage from India, Malaysia and Fiji among other countries. Where they too indicate their Tamil ancestry within this box, then one has results that amalgamate all of them as one category. To the extent that these figures boost the number of “Tamil Australians” in particular electoral constituencies, this request is an intelligent tactic on the part of the Tamil lobbyists.

Personal Idiosyncrasy
As a personal statement of some import, let me indicate here how I filled the boxes within Question 18. I inserted four categories within the frame “Other”.


The simple point is that my father is Barbadian. But being coffee-coloured rather than deep black, he probably had quite a smattering of white genes (untraced) in his distant parental past. My mother’s father was a “Perera” according to her birth certificate; but her mother was a Bastiansz from Matara. So she herself was a mix of Sinhalese and Burgher in the Sri Lankan scheme of things. Thus I am a quintessential thuppahi in Sri Lankan terms.

Thuppahi is a disparaging term, part of the weaponry of denigration. It derives from the history of Portuguese Empire in the Indian Ocean and the emergence of a category of mixed bloods and/or indigenous Christians who were called “Topaz,” “Tupass” and other such related derivatives, that is, of mixed ethnic-stock. Given the prevalence of caste thinking derisive of mixtures, and thereafter the deepening during the colonial era of racial thinking hostile to racial mixtures or people of colour, one had the making of a pejorative amalgam.

Those interested in this terminology should dip into Hobson-Jobson, being A Glossary of Anglo-Indian Colloquial Words and Phrases, prepared by Col. Henry Yule and AC Burnell in 1886 (London, John Murray) which is available in modern editions. They should also consult my clarification of this process in the world of Sri Lanka and more specifically within Sinhala nationalist thinking and everyday speech during British times which can be found in the chapter “Pejorative Phrases: the Anti-Colonial Response and Sinhala Perceptions of the Self through Images of the Burghers,” in the book People Inbetween. The Burghers and the Middle Class in the Transformations within Sri Lanka, 1790s-1960s, (Ratmalana, Sarvodaya Publishing Services, 1989).

Thus, in summary, thuppahi (more correctly transliterated as tuppahi) means low, inferior, intermixed and pariah in the Asian categorical order. The best translation of such Asian terms as “pariah” and thuppahi for those who are not familiar with this vocabulary is “mongrel.” I mark it here because it is also a sign deployed as a masthead for my web site – as, indeed, clarified in “Why Thuppahi.”[v] By stressing this dimension of my bloodlines I am deliberately seeking to confront, challenge and undermine the thinking of those who are attached to notions of caste distinction and/or “racial” superiority.

An incidental question arises from this intervention: how will the Australian Bureau of Statistics treat my answer? One presumes that I will be the only one to use the term “Mongrel.” It is probable that this answer will be slotted into the miscellaneous box “Others” created by the Bureau to encompass categories of Ancestry that are too small to be enumerated separately. That is fair dinkum of course.

However, there is an intriguing possibility. One of my tennis mates, Mike Morley, is a Kiwi long resident in Adelaide. He told me that he had indicated his Irish ancestry in Box 18. When I remarked that I had also inserted “Mongrel” in that listing, he immediately laughed and said he would have added that too if he had considered matters carefully. It is now possible that the two of us may initiate an association called the Mongrel Peoples of Australia to convey a point to both politicians and the Australian universe. Such a move will not create waves: most Aussies are likely to see the funny side and deem it a case of taking the piss out of the enumeration process.


From Australia to Sri Lanka

For Sri Lanka, however, such a step would not be light-hearted affair. For over ten years[vi] in deadly seriousness I have advocated a census enumeration which explicitly recognises “Mixed Lankan” as an ethnic identity within a scheme that also denotes other specific categories. This was spelt out in print form in “Some Pillars for Lanka’s Future” in Frontline, Vol. 26/12, 19 June 2009, pp. 24-27.[vii]

In this argument the government should ask each citizen/resident of Sri Lanka to indicate his or her ethnic identity in pre-determined categorical terms by ticking boxes which propose several alternatives, while leaving space – as in Australia – for individuals to specify other terms that are of their own choice.

Thus, the official listing should have the following categories: Sri Lankan, Muslim Lankan, Malay Lankan, Sinhala Lankan, Tamil Lankan, Malaiyāha-Tamil Lankan, Burgher Lankan, Eurasian Lankan, Colombo-Chetty Lankan, Borah Lankan, Memon Lankan, Malayālam Lankan, Parsee Lankan, Väddā Lankan, Sindhi Lankan and, last but not least, Mixed Lankan (or mishra sinhala in the Sinhala language).

This is radical move. It seeks to undermine prejudices in ways that will contribute to the development of unity in diversity over a period of time. For one, it attempts to initiate the development of hyphenated identities of the type embodied in such self-perceptions or descriptions as “Italian Australian,” “Jewish American” or “Polish American.” To install such a transformative set of concepts in Sri Lanka will not be a simple matter. For the process to have any impact the terminology has to overcome the resistance within Sinhala speech to the use of hyphens and thus to hyphenated labels such sinhala lānkika or burgher lānkika, etc etc. Thus, to repeat, this is a suggestion that can only take root over generational time – in the long haul.

Secondly, let me stress that my intention is to use this device as a means of undermining the insidious, but powerful, tendency among some Sinhala people to treat “Sri Lankan” and Sinhalese” as synonyms.[viii] This type of slippage enables the Part (Sinhalese) to equate itself with the Whole. It is a hegemonic act, even when working without aforethought below the surface.

When President Mahinda Rajapaksa’s victory speech on 19 May 2009 contended that ethnic identity was of no relevance within Sri Lanka and beat the drum of “Sri Lankan-ness” by indicating that the only relevant differentiation was between those patriotic (rata ādhara karana aya) and those unpatriotic (rata ādhara nokarana aya), the potential danger was missed by those who applauded this pitch. The emphasis concealed the possibility that the Sinhala part, a majoritarian part in the island’s demographic configuration, was encompassing the mosaic whole under the parasol of Sri Lankan-ness.[ix] As I argued in “Some Pillars for Lanka’s Future,” this potential was marked immediately afterwards at the celebration honouring the war heroes on Friday 22nd May. Here, the President

spoke of the jātika kodiya, sinha kodiya (national flag, Sinha flag) in the same breadth. In this critical conceptualization a part of Sri Lanka, the Sinhalese people, is equated with the whole of Lanka. This ideological act of merger is presented in a taken-for-granted manner, thus, insidiously and powerfully.[x]

This weakness also attends the remarkable and stirring expressions voiced by Kumar Sangakkara at the very end of his Colin Cowdrey Lecture in July this year.[xi] Speaking from a background and position that differed from that of President Rajapaksa, Sangakkara expressed a form of cosmopolitanism and idealism that fused a number of central identities within Sri Lanka into one being, Albeit well-intentioned and liberal in inspiration, the problem with this stirring formulation has been identified independently by both Kalana Senaratne[xii] and Aachcharya (alias Guruparan) in a different vocabulary from that used in my article, but with the same critical thrust.

Let me reiterate some of Aachcharya’s criticisms[xiii] here: firstly, he contends that “the idea of Sri Lankanness is identified in practice with the dominant political ideology in Sri Lanka, “namely, that of “Sinhala Budddhist hegemonism;” and, secondly, the type of idealism embodied in Sangakkara’s sentiments, he says, believes that “ethnic, religious and other communitarian groups are bad for the individual;” “[but represents a position that] rests on a poor understanding of the contribution that a communitarian setting makes to the well being of an individual.”

In opposition Aachcharya insists that “it is important to move from the particular to the universal and not to dichotomise the relationship between the two.” We must, therefore, “first understand the reality of value pluralism.” Thus, “the plurinational Sri Lankan identity can never be constructed without acknowledging and recognising (institutionally) the significance of other identities that evoke a sense of belongingness and community.”[xiv]

Summing up, therefore, in my argument the ecumenical approach presented so wonderfully by Kumar Sangakkara generates a notion of Sri Lankan-ness that is a melange of many categories, a pot pourri dish so to speak, one that could be depicted in indigenous metaphor as an achchāru (pickle). In itself that kind of category and identity is not to be discarded or disallowed. It can be a fertile and liberating perspective for individuals and families. The box “Sri Lankan” in my listing will enable those so minded, even those of mixed lineage, to mark that category if they wish to adopt such a cosmopolitan perspective.

But it is both illiberal and impolitic to deny others who are attached to their Sinhalese, Tamil or other specific communitarian identities the right to specify this affinity as one step in their being-ness as “Sri Lankan.” Patriotism and nationalism can be built up like a pyramid formation in gymnastics. By way of example I insist here that for many Burghers residing today in Sri Lanka it is through their life-world and being as Burghers (their Burgherness in other words) that they become Sri Lankans and adhere to Sri Lanka in subjectivity and loyalty. So, too, with necessary adjustments, can this be said of many Sinhalese, Muslims, Malays, Sri Lankan Tamils, Malaiyāha Tamils, et cetera — though I do not think that any social scientist could even guess at the proportions in each category which are so constituted.

The idealism that impelled Sangakkara to present what I have called “an ecumenical Asokan perspective” in such a majestic way[xv] should be encouraged via the category “Sri Lankan.” However, the census categories must – and I insist here on the MUST — permit other avenues of ethnic subjectivity, categories of being that have built, and can continue to build, Sri Lankan-ness on their communitarian pillars. Among the several such building blocks should be that collective identity which recognises and values the subjectivity of those sankara, namely, those with a mixed pedigree. In this manner one can strike a blow against casteism and any emphasis on racial purity. But, more vitally, it will remind people of Sri Lanka of the variety and complexity of their history. The island state and society has to be construed, like the skies above us, as a constellation of diversity.

[i] See Roberts, “Ethnicity after Edward Said: Post-Orientalist failures in comprehending the Kandyan period of Lankan history,” in Roberts, Confrontations in Sri Lanka: Sinhalese, LTTE and Others,”Colombo, Vijitha Yapa Publications, 2009, ISBN 9789556650358 (

[ii] See and http://

[iii] See Michael Roberts, Incursions and Excursions in and around Sri Lankan Cricket (details in

[iv] In a word file document “Say we are Tamils” (sent to me by a Tamil friend).


[vi] I first presented this suggestion during the Marga Workshops in the years 2000-03 devoted to “A History of Ethnic Conflict in Sri Lanka: Recollection, Reinterpretation & Reconciliation;” but the idea was not greeted with any enthusiasm by the liberal moderates who were the predominant element in the small seminar audience.

[vii] See In this Frontline article I used the term “Samkara Lanka” (samkara can also be written as sankara). Sankara is an erudite Sanskrit-Sinhala word for mixed lineage or mixed blood. It was a term used occasionally by Piyadasa Sirisena. I have, however, been informed that this term would confuse contemporary Sinhala speakers and that it carries a pejorative connotation today. Thus, I now consider it best to use mishra sinhala in the Sinhalese language and “mixed Lankan” in English.

[viii] I spotted this tendency initially in some of the Anagarika Dharmapala’s writings in the early twentieth century. This finding was a critical ingredient in my pessimistic reading of the Sri Lankan political scene in the 1970s which contended that the island would go the way of Cyprus, Lebanon and Northern Ireland [embodied in an article drafted in Germany in 1976 and appearing in print in 1978: viz., “Ethnic Conflict in Sri Lanka and Sinhalese Perspectives: Barriers to Accommodation,” Modern Asian Studies, 1978, 12: 353-76, which can be found in Roberts, Exploring Confrontation, (Reading, Harwood, 1994]. Dharmapala was a Sinhala Buddhist ideologue and can be considered the patron saint of what political analysts call “the 1956 ideology,” which in turn inspires the “Mahinda Chintanaya” of contemporary times. For my subsequent studies of Dharmapala, see the two essays,For Humanity. For the Sinhalese. Dharmapala as bosat Crusader,” and “Himself and Project. A Serial Autobiography. Our Journey with a Sinhala Zealot, Anagarika Dharmapala,” which are reprinted as chaps 8 and 10 respectively in Roberts, Confrontations in Sri Lanka: Sinhalese, LTTE & Others, Colombo, Vijitha Yapa Publications, 2009.

[ix] As also stressed by Aachcharya in groundviews: “But when those who take part in the Sinhala Buddhist hegemonic project assert ‘Sri Lankanness’, as Rajapaksha did in his end-of-the-war speech (the no-minorities in Sri Lanka speech) many quite rightly suspected the motives behind such an agenda” ( back-to-basics-the-need-for-an-honest-conversation-about-%E2%80%98sri-lankanness%E2%80%99-and-%E2%80%98sri-lankan-identity%E2%80%99/)..


[xi] and

[xii] Senaratne, “Our Sri Lankan Identity: Another Case of ‘Being Nobody, Going Nowhere’? in  nobody-going-nowhere%e2%80%99/


[xiv] However, I would place caveats around Aachcharya’s claim that “[the assertion of] Sri Lankanness within the present political status quo (without challenging it) is to sign up to this agenda of Sinhala Buddhist hegemonism.”
[xv]Roberts, “Kumar Sangakkara’s sentiments as an ecumenical Asokan Lankan,”