KAMALIKA PIERIS

The first census of Ceylon of which there is a record is the census of 1824. That was an incomplete census which grouped the population by caste.  The first complete population census   of Ceylon was the census of 1871.  This census left out ‘caste’ and introduced ‘race’ and ‘nationality’,   two European   concepts which were in vogue at the time.   Race appeared for the first time in this census.  There were 72 nationalities and 24 races, including foreign ones such as Chinese, German and Irish. The local ‘races’ were Sinhalese, Tamil, Burgher, Chetty, Malay and Moor. Sinhalese and Tamils were ‘races’ as well as ‘nationalities’. 

The Sinhala, Tamil and Moor races were officially recognized for the first time in 1871. The proportions were Sinhalese 69.40%, Tamils 22.21% and Moor 6.79%. There was a census every ten years or so after 1871.  At the 8th International Statistical Congress 1872, it was agreed that a census must include language, religion, birthplace and nationality. The 1881 Census referred to 72 nationalities in the text but classified the population only by ‘race.’ The races were ‘Europeans, Sinhalese, Tamil, Moors, Malay, Veddahs and Other’. Race became the main category of classification thereafter. The 1911 Census    had ten races, ‘Low country Sinhalese, Kandyan Sinhalese, Indian Tamils, Ceylon Tamils, Indian   Moors, (also known as Coast Moors) Ceylon Moors, Malays, Burghers, Veddahs, Europeans and Other’.

The 1911 Census said Sinhalese and Tamils are distinct, clearly differentiated races. They have their own religion and speak different languages. ‘Their settlements are clearly defined.’ Intermarriage between them is very rare. Even a superficial observer could see, it said, that there are marked physical differences between the Sinhalese and the Tamils. However,  the 1921 census has photographs of the ‘races’. They all look alike. The   difference was in their costumes, not in facial or physical characteristic.

Ponnambalam Arunachalam in his introduction to the 1901 census used history to show that the Sinhala and Tamil races were genuine and long standing. Arunachalam added a chapter on the history of Ceylon to the 1901 census. This   was not there in the earlier reports.  He wrote that the Tamils and Sinhalese have lived in the island for two thousand years. He drew attention to Dutugemunu’s fight with Elara. Denham in 1911 said that the Sinhalese and Tamils had been in Sri Lanka for centuries, fighting with each other. However, he added that only the Sinhalese could ‘regard Ceylon as home’. It was the ‘shrine of their national traditions.’

Each district was   designated ‘Tamil’ district’ and ‘Sinhala district’ according to the racial percentage. This created geographically contiguous ‘Tamil areas’ and ‘Sinhala areas’.   Census of 1871 showed 50-90% Sinhalese in all provinces except Jaffna, Vanni, Mannar (5%) and Trincomalee, Batticaloa (10%). Tamils were 90% in Vanni and Jaffna, 70% in Mannar and Eastern province, 40% in Central province and 16% in Badulla. Moors were 35% in Batticaloa 30-20% in Mannar, Trincomalee and Puttalam.

The Census of 1891 has three separate maps indicating the ‘relative proportions of races’ for Tamil, Sinhala and Moor. Jaffna, Vanni, Mannar, Trincomalee and Batticaloa districts were assigned in bright pink to the Tamils. These figures ignore the important issue of population density.  The 1911 census showed the following population density. In Northern Province, Jaffna with 998 sq miles had 326,510 persons. Mannar had 904 sq miles and 25,603 persons, Mullaitivu had 1466 sq miles and 17,336 persons.  In Eastern province, Batticaloa had 2800 sq miles and 153,943 persons Trincomalee had 1048 sq miles and 29,374 persons.

Sometime after the Sinhala race” was invented, it got bisected.  Ponnambalam Arunachalam, Superintendant of Census, divided the Sinhalese into Kandyan and Low country for the 1901 census. Then he said that Tamils were equal to each of the Sinhala groups when taken separately. Census of 1921 stated that the Kandyan Sinhalese differed from the Low country Sinhalese in all respects except those of color, religion and language. Kandyan villagers saw the Low country Sinhalese as a separate race (pahata rata minissu), probably due to the impact of Portuguese, Dutch, British, Malay and Chinese influences (sic).

Census of 1911  however, stated ‘ the distinction between Kandyan and Low country Sinhalese is every year lessened, intermarriages are on the increase and in many parts   of the  ‘Up country’, it is difficult to distinguish  between Kandyan and Low country men and women’.  Then  in  1922   a  district court case dealing with property rights in marriage  went  into appeal and  Supreme Court ruled that  that Low country Sinhalese  and Kandyan Sinhalese were the same race. (NLR vol.24 p245)

But as late as 1946, persons from the Central, North central, Uva and Sabaragamuwa provinces, the Kurunegala and Puttalam districts and the ’Sinhalese divisions’ of the districts of Batticaloa, Trincomalee and Vavuniya continued to be Kandyan. Persons from Western and Southern provinces and Chilaw and Puttalam districts were Low country (Census 1946). Arthur Ranasinghe, Superintendant of census for 1946 cited the 1922 case and observed that this division made no sense. Kandyan and Low- country Sinhalese were treated as two separate census groups from 1901 to 1971.  They were combined in the 1981 census.

Ponnambalam Arunachalam said in the 1901 census that the ‘modern’ approach at that time was to identify ‘aggregations of persons believed or presumed to belong to the same stock and having a common language, character and political institutions.’  But no objective criteria  were devised for this. The answer given by the respondent was accepted by the enumerators in 1981. Denham had said in 1911 that it is inconceivable that any Sinhalese would enter himself as Tamil or vice versa.

In the Census of 1946, the enumerators were offered a list of races and had to slot people into one of them. The enumerators were instructed, ‘ enter the race of each as Low country Sinhala, Kandyan Sinhalese, Ceylon Tamil, Indian Tamil,  Ceylon Moor .Indian Moor, Malay, English, scotch, Irish etc. also state whether Malay, Parsee, Borah. Do not use general terms such as British , Ceylonese. The children were given their fathers’ race.

In 1946, the criteria for deciding if a person was a  Ceylon Tamil ,was ‘racial stock’. Ceylon Tamils are those who can trace their origins to a ‘Tamil district in Ceylon,’  said the Census. A Ceylon Tamil therefore  was a person who traditionally had his origins in a Tamil district. It was ‘stock’ not birthplace that decided.

For language, 1946 census said’ Enter in this cage the language of the race to which the father belongs. For Moor, Burghers, Eurasians the language used in the home.  The census asked for ‘mother tongue.’ But father’s tongue must be stated  for the mother tongue. Enter father’s language in the mother tongue column”, the Census instructed.

The Census asked whether   a person could speak Sinhala or Tamil. There was a separate question for literacy, which asked can they read and write Sinhala or Tamil.  Jennings said the census would have been more useful if it has asked, instead for mother tongue,  for languages spoken, literacy in mother tongue, literacy in other languages. The 1950 Whitepaper quietly dropped mother tongue and substituted the language of the home, noted Jennings.

The 1981 census classified the population into six ethnic groups: ‘Sinhalese, Sri Lanka Tamil, Indian Tamil, Sri Lanka Moor, Burgher, Malay and other.’  The 2012 census added ‘Sri Lanka Chetty’ and ‘Bharatha’ to the list. The ethnic pattern in 2012 was: Sinhalese (15,250,081) Sri Lanka Tamil (2,269,266) Indian Tamil (839,504) Sri Lanka Moor (1,892,638) Burgher (38,293) Malay (44,130) Sri Lanka Chetty (5,595), Bharatha (1,717) Other (18,215).  Total 20,359,439. (Census 2012). 

The Population Census, which is based on the Census Ordinance, has a great deal of authority and influence, observed Tudor Silva. Its statistics are used as the base for many economic activities, such as distribution of state resources.   Its ethnic categories, which are utterly artificial, are accepted as official. This has led to other complications. These artificial ethnic Identities then started to    develop a life   and history of their own   and become fixed.   Now they are claiming land rights and   sovereignty, Tudor concluded. (Continued)