Decline of The Native Speaker
Decline of The Native Speaker
Decline of The Native Speaker
Chapter 11
157
In England it doubles in 50 years; therefore, in a century (in 1970) it will be 124,000,000. In the United States, in Canada, in Australia, it doubles in twenty-five; therefore it will be 736,000,000. Probable total of English-speaking race in 1970, 860,000,000. (Bailey, 1992: 111) One can detect, even a century or so later, the sheer exhilaration in all those zeros. Figure 11.1 shows the logarithmic algorithm employed by Candolle, and projects the numbers of English speakers a century before and after the 100-year period he describes. It suggests that in 1770 there were only about seven million speakers of English, but by 2070 the projected number would be nearly 10 billion: rather more than the expected total population of the planet.
We may now smile at nineteenth century confidence in forecasts for English, but there undoubtedly remains a fascination with speaker numbers despite the fact that there are many other aspects of global English which might better capture its growing importance in the world. The issue of How many speakers? remains, for example, top of the list of frequently asked questions about global English, according to the British Councils FAQ.1 This paper examines what we know about
158
trends in English language usage and provides a principled basis for estimating and forecasting the numbers of people speaking English in the world.
159
Figure 11.2 Estimated growth of numbers of English native speakers since Anglo-Saxon times In The Future of English?, a research document published by the British Council (Graddol, 1977), I put forward a model for the growth of native speakers of English which provides a basis for projecting future trends. The model relies on two kinds of data: full demographic forecasts for populations of countries in which English is spoken as a first language and estimates of the proportion of those populations for whom English is a mother tongue. In principle, all the major world languages can be modelled in this way, but English is perhaps the most difficult. McArthur (1998: 42) lists 113 territories which currently use English on a widespread, sustained basis. The Ethnologue 2 database of world languages maintained by the Summer Institute of Linguistics lists 83 countries and territories in which English is spoken. Ideally, any model forecasting numbers of speakers requires data for each of these countries. Country by country population data is available from the United Nations Population Division. The figures reported in this article draw on the Sex and Age Quinquennial dataset for 19502050 (1998 revision). This dataset provides detailed population forecasts, including numbers for males and females in each of 21 age-bands, at five-year intervals through to 2050. Such detailed population forecasts are insufficient on their own. We also need to know the population of each countrys population who speak English as their mother tongue. This is a surprisingly difficult statistic to obtain. It raises some of the most problematical issues in linguistic surveys, such as What counts as a native speaker? How do you collect such information systematically in many countries? How do people interpret questions about language usage? Do people answer such questions honestly? Even in those countries such as Britain and the US where the
160
linguistic culture is often described as monolingual, it is clear that a significant proportion of the population speak languages other than English as their first language. Furthermore, the assumption of low linguistic diversity in such countries means that language questions are rarely included in national census returns. We are left with an unsatisfactory rag-bag of sources from partial studies in different years, often based on very different criteria. The problem is demonstrated by the Ethnologue itself which estimates the global number of English first-language speakers at 322 million. They cite two sources for this estimate: a 1984 study of the UK, and the World Almanac (1995). The latter provides a half-page table of language statistics below a list of pen names of famous writers. The language data are compiled by a retired professor of psychology with an interest in Esperanto, but there appears to be no published account of how the figures were arrived at. Despite the problems associated with estimating native speaker numbers, approximate figures for present-day usage can be derived. David Crystal (1997), for example, provides estimates of L1 speakers in 56 countries, arriving at a total of over 337 million speakers worldwide. These figures can provide a starting point for modelling long-term trends. Since the size of population of each country in 1995 is known, we can estimate the proportion of the population who, according to David Crystals estimates, speak English as their first language. In practice, we can model only 24 of the 56 countries in this way these are the ones with large enough populations to be included in the UN dataset. The areas excluded from the model are mainly island territories, such as St Kitts and Nevis (with an estimated 39,000 speakers) and Guam (56,000). It is clearly important to include these areas in any comprehensive description of English in the World, but their exclusion from the statistical model makes almost no difference to global estimates. Figure 11.3 shows the projections provided by the model for native speaker numbers during the century 19502050. It shows that the growth of L1 speakers will slow in about 20 years time, leading to a likely total population of around 433 million in 2050. Despite the continued gradual growth in absolute numbers of native speakers, the proportion of the worlds population who speak English as their first language has, in fact, declined sharply. Figure 11.4 shows the same numbers as Figure 11.3, but expressed as a proportion of the worlds population calculated year-on-year. It shows the impact on English of the rapid population increase in parts of the world where English is not spoken as a first language. This represents a significant demographic shift away from English to other, non-European languages.
161
Figure 11.3 Projections for native speaker numbers to 2050, and estimates 19501995
Figure 11.4 Decline of native English speakers as a proportion of the worlds population
In 1950, over 8% of the worlds population spoke English as their first language; by 2050 it will be less than 5%.
162
163
1995 to 278 million. A similar allowance can be made for Latin America, particularly Argentina, Costa Rica, Honduras, Nicaragua and Panama. This addition brings the estimated total of L2 speakers worldwide in 1995 to 351 millions. By 2050 this may have grown to 668 millions (Figure 11.5, Line C).
164
dramatically, according to census data, between 1971 and 1981 from 25 million to 3 million. This reflected a change in the census questions on language rather than actual patterns of English usage. Mahapatra (1990: 9) suggests that there has been a deliberate suppression of linguistic data on the extent of Indian multilingualism. Certainly, there seems to be a tradition of talking down the extent of Indian English usage. Traditionally, almost no Indians are regarded as using English as their first language (a notion which will undoubtedly come under further scrutiny in the future). Crystal (1997) suggests a figure of around 0.03% of the population. Srivastava (1990: 50) claims English is spoken and understood by only 2% of the population. Most linguists, however, seem to agree that around 4% of the Indian population speaks English as a second language. Kachru (1984), for example, suggests 4% as a conservative estimate; Crystal (1997) presents a similar figure. This suggests there were a little more than 37 million L2 users in 1995. This figure is represented in Line B of Figure 11.5. There is evidence, however, that the number of English speakers in India is higher than this. The magazine India Today, commissioned a survey in mid 1997 which claimed: contrary to the census myth that English is the language of a microscopic minority, the poll indicates that almost one in three Indians claims to understand English, although less than 20 per cent are confident of speaking it. (India Today, August 18, 1997). If the figure of 20% is accurate then there must be around 186 million English speakers in India. Of course, it is unlikely that India is the only country in which data is inaccurate. The global number of L2 speakers of English would be significantly affected if similar revisions were made to estimates in Bangladesh and Pakistan. Line D on Figure 11.5 shows the effect of upgrading the numbers for India and Pakistan to 20%, and Bangladesh to 10%. This trajectory assumes there were around 418 million L2 speakers worldwide in 1995 and indicates that by 2050 there may be 790 million L2 speakers. Although this projection is higher than Line B, the starting point (of 418 million speakers in 1995) is not much higher than the consensus figures. The model allows us to understand better the different projections which result from taking the lower or higher estimates.
165
geolinguistic space. Eurobarometer surveys of the populations of European Union (EU) countries have collected data about language usage for some years, and it is clear that in some countries English is now widely spoken: 77% of Danish adults and 75% of Swedish adults for example, say they can take part in a conversation in English. Eurobarometer 50, based on a survey made in late 1998, concludes that nearly one third of the citizens of the 13 non English-speaking countries in the EU can speak English well enough to take part in a conversation. Furthermore, it seems that the function of English is changing: European citizens learn and use English in order to communicate with European nationals from all countries and not just with native speakers. English, in other words, is fast becoming a second language in Europe. Eurobarometer 41, based on data collected at the end of 1994 and analysed by Labrie and Quell (1997), found large country variations in English usage, ranging from Spain (with 13% of adults able to hold a conversation in English) to the northern European countries such as Sweden (75%) and The Netherlands (71%). Figure 11.6, Line E shows the projections for European L2 use, based on the 13 relevant member countries, together with estimates for Norway and Switzerland. These suggest that in 1995 there were around 95 million English speakers in these 15 countries. The relatively static population figures for Europe mean that this number will soon peak, and then gradually reduce during the next 50 years.
166
167
The shift in Europe towards English which reflects increased biand multilingualism rather than abandonment of other languages undoubtedly has several causes, including significant changes to language curriculums in school; exchange programmes encouraging young people to move between EU member countries; the growing importance and utility of English as a second language in Europe; and greater exposure to English cultural products. The problem in modelling such shifts is that the rate of shift itself changes (Figure 11.8). Line F of Figure 11.6 shows the consequences of making conservative assumptions about the rate of shift in Europe. It assumes that ongoing shifts, as estimated from successive Eurobarometer surveys, are now at their most rapid point of development, and that in the future the rate of change will decline according to the curve shown in Figure 11.8. Line F should therefore safely underestimate the future trend towards L2 English usage in Europe. Throughout the EU as a whole, the increase in the number of people speaking English seems now to be in the region of just over 3% per year. (So, if 31% of the EU population spoke English as a second language in 1998, we would expect that to increase to 31.9% in 1999 and 32.9% in the year 2000.) It is clear, however, that the trend towards English is occurring
168
more rapidly in some countries than others. Eurobarometer surveys suggest that of all EU countries, shift is slowest in Portugal and fastest in Spain. The projections in Line F suggest that the present one third of the EU population who speak English will grow to roughly two thirds to 2050. But the EU is likely to acquire new members during this time, such as Hungary and Poland. This will alter the relative positions of major European languages within the EU in less predictable ways. A similar modelling of language shift can be applied to all countries in the world where English is spoken as an L2. Figure 11.9, Line G shows the effect of adding the European numbers of L2 speakers to the global figure (i.e. Line D and E added together). This provides a grand total of 518 million L2 speakers in 1995, rising to 880 million by 2050. Line H includes conservative estimates of language shift in all countries with significant numbers of L2 speakers. This starts from the same point of 518 million speakers in 1995, but projects 1.2 billion speakers by 2050. Figure 11.10 compares the lowest estimates in this paper (Line B) and the highest (Line H). The future is likely to lie somewhere between these two extremes. The lowest variant (B) suggests that L2 speakers will not outnumber native speakers until the 2030s. The highest variant (H) suggests that native speakers lost their majority in the 1970s.
169
Figure 11.10 High and low estimates for global use of English as a second language
170
The status of native speaker The decline of the native speaker in numerical terms is likely to be associated with changing ideas about the centrality of the native speaker to norms of usage. There has been a lively debate about the cult of the native speaker: do native speakers have privileged access to an understanding of the language, and are they therefore more reliable informants and teachers? The special status accorded to native speakers is long standing. In theoretical linguistics, it is associated with the rise of Chomskyan linguistics and the special role assigned to introspection and intuition in theory-building. The native speaker is claimed to have access to a much richer source of data (in judgements of grammaticality, for example) than a researcher could discover by studying a speakers actual utterances. This theoretical turn has had political repercussions. As the US was the centre of research and publishing the new paradigm, English became a privileged object of study. Theoretical linguistic studies of other languages lagged behind those of English. Although it was maintained that the project with Transformational Generative Grammar was a universalist one (more interested in universal properties of the human mind than the characteristics of particular languages), English remained the centre of attention. It would be wrong, however, to think that the importance of native speakers began with Chomsky. Traditional dialectologists, as well as anthropologists, drew on similar ideas of good speakers (often old, male, non-mobile). But the discourse of the native speaker is even older it is a product of modernity. Since the European Renaissance, identities have been constructed according to a particular model of perfection: unified, singular, well-ordered. Language has played a major role in the construction of modern European identities from the level of nation states and standard languages, to the subjectivities of individual speakers. Multilingualism, both in individual and social terms, does not fit well with concepts of modernity: multilingual speakers and societies have long been regarded with suspicion, and have caused problems for public policy. During the late twentieth century, much of the modernity project has unravelled. The construction of a new Europe has resulted in new formations of identity; greater mobility; and a new hierarchy of languages which places most European citizens in a plurilingual context. In many countries in which English is learned and spoken as a second or foreign language, the centrality of the native speaker is being challenged. At one time, the most important question regarding global English seemed to be will US English or British English provide the world model? Already
171
that question is looking dated, with the emergence of New Englishes, and dictionaries and grammars that codify new norms. But the tantalising question still remains. Large numbers of people will learn English as a foreign language in the twenty-first century and they will need teachers, dictionaries and grammar books. But will they continue to look towards the native speaker for authoritative norms of usage?
Notes
1. http://www.britishcouncil.org/english/engfaqs.htm 2. http://www.sil.org/Ethnologue/
References
Agnihotri, R.K. and Khanna, A.L. (1977) Problematizing English in India. New Delhi: Sage. Crystal, D. (1995) Encyclopedia of the English Language. Cambridge: Cambridge University Press. Crystal, D. (1997) English as a Global Language. Cambridge: Cambridge University Press. Graddol, D. (1997) The Future of English? London: The British Council. Kachru, B.B. (1984) Standards, codification and sociolinguistic realism: The English language in the outer circle. In R. Quirk and H.G. Widdowson (eds) English in the World: Teaching and Learning the Language and Literatures. Cambridge: Cambridge University Press for The British Council. Labrie, N. and Quell, C. (1997) Your language, my language or English? The potential language choice in communication among nationals of the European Union. World Englishes 16: 326. McArthur, T. (1998) The English Language. Cambridge: Cambridge University Press. Mahapatra, B.P. (1990) Multilingualism in India: A demographic app raisal. In D.P. Pattanayak (ed.) Multilingualism in India. Clevedon: Multilingual Matters. Srivastava, A.K. (1990) Multilingualism and school education in India: Special features, problems and prospects. In D.P. Pattanayak (ed.) Multilingualism in India. Clevedon: Multilingual Matters.