Phenotypical, Linguistic or Religious? On the Concept and Measurement of Ethnic Fragmentation

Malaysian Journal of Economic Studies, Jun-Dec 2003 by Yeoh, Kok Kheng

"The real division stems from religion as a cultural force and a badge of ethnic identity ... For a Catholic, religion is an integral part of Irish nationalism, something inextricably joined with the history of a persecuted and oppressed people struggling for liberation. For a Protestant, religion is even more important because of a confusion over national identity that leaves him unsure whether he is British, Irish, or Ulsterman ... In the "black North" of Ireland, as in the Middle East, religion is what distinguishes "us" from "them", especially for Protestants."

Thus sectarianism can be conceptualised as a sub-type of ethnic diversity, and religion as a source of ethnic differentiation. This reorientation in perspective is not only applicable to Northern Ireland, but also possesses important theoretical implications for other cases where religion is perceived to be a principal source of conflict, e.g. Lebanon, BosniaHercegovina, Sri Lanka and Cyprus.

Therefore, the term 'racial' should more appropriately be used to describe group distinction on the basis of phenotypical (i.e. physical) characteristics, while 'ethnic' refers to those based solely or partly on cultural characteristics. The term 'ethnic' can also be generalised to be a blanket concept (Hoetink's attribute 'socioracial') to cover both the above distinctions. The term 'cultural' here mainly covers the ascriptive attributes 'ethnolinguistic' and 'ethno-religious'. The emphasis on language and religion in empirical research is due mainly to the fact that they are the relatively less vague factors in the fourfold categorisation of ascriptive loyalty (Hoetink 1975: 23-4). Despite examples such as the Jews and Judaism or the tendency to identify Arabs with Islam, the use of religion to define ethnicity is unsatisfactory. Turks, Kurds and Arabs have the Islamic faith in common but it is absurd to classify them as one 'ethnic group'. Similarly, the Muslim Bengalis in East Pakistan (later Bangladesh) had failed to identify with the Muslims in West Pakistan, neither have the ethnolinguistically diverse co-religionists in (West) Pakistan itself ever identified with one another. While the sharing of a common language has been the most frequently attested attribute of 'ethnicity', there are flaws with this definition too. The cases of Hindustani (Hindi-Urdu) and Serbo-Croatian are examples where religion overcomes language in defining ethnic identity. Thus there is reason to regard these two ascriptive criteria as largely complementary. Karpat (1985 : 96) gave the following example:

" ... today the Bulgarian government regards the Pomaks (Bulgarian-speaking Muslims) as ethnic Bulgarians but the Pomaks do not accept that view. They intermarry not with Christian Bulgarians but with Muslims. Turkey accepts as "Turks" the Bosnian Muslims and the Pomaks although these do not speak a word of Turkish and belong to the Slavic race. In other words, today, language and religion are assumed to go together, although they do not always do so in fact: Muslims attach more importance to religion while Balkan Christians emphasise language as the primary ethnic bond."