1. Introduction
English spelling has a reputation for being notoriously difficult to learn, whether for native speakers of English or for those who learn it as an additional language. An additional problem for EFL learners is that there are two somewhat different systems to choose between: the British system and the American one. As Bondesen (Reference Bondesen2004: 4) points out, ‘although the two spelling systems are much more similar than they are different, there is variation between them.’ (For surveys of some of the main differences see Trudgill & Hannah, Reference Trudgill and Hannah1994; Carney, Reference Carney1997.) The spelling discrepancies between the two varieties are systematic, and originate in large part from American spelling reform as a symbol of ‘linguistic independence’ (Knowles, Reference Knowles1997), and from the work of early American linguists such as Benjamin Franklin and Noah Webster. The aim of this article is to investigate the spelling preferences of English users in a few regions outside Britain and the United States, and to establish whether any clear patterns of regional variation are discernible.
In the Greater China context, many English varieties have been recognized, including Chinese English (or even Chinese Englishes, e.g. Bolton, Reference Bolton2003), Hong Kong English (e.g. Bolton, Reference Bolton2000, Reference Bolton2002), and Taiwan English (e.g. Jian, Reference Jian2004). However, the actual Englishes used in Mainland China, Hong Kong and Taiwan have their own regional characteristics since social, cultural, economic and political factors all play a part in shaping English forms. Wang (Reference Wang1991) notes that both the spoken English and the written English in China are a linguistic mixture and products of the two standard English varieties, which will be referred to in the present article as Standard British English (BrE) and Standard American English (AmE) respectively.
Over the years, the linguistic features of the different varieties in Greater China have been described in great detail, focusing on such aspects as pronunciation, lexis, sentence structure and discourse patterns. However, the subject of English spelling has thus far been inadequately explored. Which spelling system does each of these varieties favor? The BrE model or the AmE model? Few empirical studies have been conducted in this field, and no statistical evidence has been offered to disclose English spelling practices and preferences in these regions. To our knowledge, there have hardly been any studies of international variation in English spelling preferences since the pioneering work of Bondesen (Reference Bondesen2004), who investigated regional variation in English spelling in Canadian newspapers. Bondesen found that British spelling conventions were used, overall, at a higher rate than the American ones by journalists from seven Canadian regions in the year 2000. The present study will investigate whether similar or different trends are observed in the Greater China context.
Almost two decades ago, Fishman posed the following question: ‘[a]re there forces or processes which transcend the English mother-tongue world itself and which also contribute to the continued spread and entrenchment of English in non-English mother-tongue countries?’ (Bolton, Reference Bolton2003: 15). Fishman identified multinational corporations and elementary and tertiary-level education as key contributors to this process. In addition to these, the present study will propose that the news media are also playing an important part in precipitating the continued spread of English (or rather a prestigious English variety) and its underlying ideologies.
In order to understand recent developments in English orthography preferences in Greater ChinaFootnote 1 in the domain of the media, this article reports on a comparative study of the English spelling practices of China Daily (Mainland China), The Standard (Hong Kong) and Taipei Times (Taiwan) from 2000 to 2010, and reveals regional orthography variations and diachronic changes in their spelling preferences. The specific research tasks are:
1. To examine whether there was any systematic regional variation in actual English spelling practices among the three newspapers during the period 2000–2010;
2. To reveal whether there were any diachronic changes in the orthographical preferences of each of the three newspapers, and to identify any observable changes or trends over this period;
3. To determine when any spelling preference shifts occurred, and to consider possible causes and consequences.
2. Method
Three popular and representative English-written newspapers were used as newspaper sources for this research. China Daily, established in 1981 and published Monday through Sunday, has the widest print circulation of any English-language newspaper in China. The Standard, distributed Monday through Friday, boasts that it is ‘Hong Kong's biggest circulation English daily newspaper.’ It is now a powerful, influential medium in Hong Kong with a diverse audience and a broad reach since becoming a free newspaper in September 2007. Taipei Times, launched on June 15, 1999, assumes ‘the mission of presenting a voice long absent in the Taiwanese media: an English-language journal of record for national and international readers, presented from a Taiwanese perspective.’
A small corpus was built for the research, consisting of 1080 passages from the abovementioned newspapers, written by local journalists. Three passages per month on average were chosen from each newspaper, and news articles of the first, the 15th and the last day of each month were selected regardless of topic, genre and style. Usually, the first passage in the online archive of every issue was chosen if it was written by a local journalistFootnote 2. Accordingly, 36 passages were selected per newspaper per year, resulting in a total of 1080 passages over the 10 years from the 3 newspapers.
The next step was to identify representative spelling categories in the corpus and obtain occurrence frequencies for both BrE and AmE spelling variants. Trudgill & Hannah (Reference Trudgill and Hannah1994) list 16 major spelling differences between the two varietiesFootnote 3. This research draws upon this categorization. However, only 6 of these pairs (-our/-or, -ise/-ize, -ll/-l -re/-er, en-/in- and -mme/-m) occur in statistically significant numbers in the corpus. Accordingly, the article will focus exclusively on these pairs.
The corpus analysis software AntConc (Anthony, Reference Anthony2014) was used to obtain and calculate the data in this study. AntConc's Concordance tool was used to search for valid words for this research. For example, in order to establish how many words are ended with -or instead of -our, the ‘Search Term' *or was used. The asterisk means that there may be letters before or. The number of Concordance hits tells us how many eligible words are found under the condition, but some words needed to be eliminated because not all of them were valid words for this research (e.g. or, for, doctor, director, etc.) The process of retrieving words for other spelling categories was the same. The percentages of BrE spellings vs. AmE spellings of the six spelling pairs were then calculated, and the inter-paper spelling variations and intra-paper spelling changes were revealed by making year-by-year comparisons.
3. Results
Altogether, 2119 valid words were found in the corpus for the six differently spelled pairs. In the following two subsections, the spelling practices of the three newspapers will be described from both synchronic and diachronic perspectives.
3.1 Regional variation in English spellings
The following six tables present the occurrence frequencies of each item in each of the six spelling pairs from 2000 to 2010. In each pair below, the first item is the British spelling form, while the second is the American form. Each set of data in the tables has two numbers: the one on the left refers to the occurrence frequency of the BrE spelling form in a given year, while the one on the right refers to that of the AmE form. To reveal regional variation, these sets of data need to be compared vertically year by year.
-our vs. –or
The examples containing -our/-or are honour/honor, habour/habor, labour/labor and favourite/favorite. BrE ends relevant words with -our while AmE simplifies it into -or. In total, 295 words ending with -our/-or are found in the corpus. The data in Table 1 indicate that the American form -or (55.93%, n = 165) was used at a higher rate than the British counterpart (44.07%, n = 130) over the 10 years in all three papers.
Table 1: Analysis of -our vs.-or
For this pair, striking regional spelling variation used to be observed among the three papers. All local Taiwanese journalists at the Taipei Times chose to spell the concerned words exclusively with the American -or throughout the whole decade. However, local journalists with China Daily and The Standard overwhelmingly chose to spell relevant words the British way before 2007 and 2006 respectively, but switched to the American form afterwards. In short, variation among the three newspapers for this pairing ceased to exist from 2008 onwards.
-ise vs.-ize
Words ending with the morphemes -ise (preferred in BrE) and -ize (preferred in AmE) are socialise/socialize, publicise/publicize, modernise/modernize, industrialise/industrialize, etc. The Standard was more likely to adopt -ise before 2005 while China Daily and Taipei Times were consistently inclined to use more -ize at the end of words, Taipei Times in particular. However, this striking inter-newspaper variation between The Standard and the other two existed only before 2005, from which year on the American spelling was universally adopted, as is shown in Table 2. Overall, -ize was used with an overwhelmingly higher frequency (90.04%, n = 705). Only 9.96% of the words (n = 78) had the British ending -ise, and most of them occurred in The Standard.
Table 2: Analysis of -ise vs.-ize
-ll vs. –l
For the spelling pair -ll/-l, the practices of Mainland China and Hong Kong were again similar. Before 2007 and 2008 respectively, China Daily and The Standard tended to use -ll more when words ending with -l are added suffixes -ed, -ing, -er or -ment. Accordingly, words such as ‘travel’ and ‘label’ were inflected into ‘travelled’ and ‘labelling’ following the British spelling convention. These two newspapers then both experienced a preference switch from BrE to AmE. By contrast, Taipei Times remained loyal to the American English model throughout the whole decade. Of all the occurrences (n = 242) for the three papers in Table 3, -ll accounts for 48.35% (n = 117) while -l 51.65% (n = 125).
Table 3: Analysis of -ll vs.-l
-re vs. -er
Typical words with -re and -er are centre/center, fibre/fiber, metre/meter and litre/liter. Spelling variations among the three regions were also revealed for this pair. The BrE -re played a dominant role in China Daily and The Standard in the first few years of the decade. The Standard then began to use -er more after 2004, and China Daily used more -er from 2007. By contrast, Taipei Times persisted in the use of the American -er. Overall, -er frequency (53.71%, n = 152) was higher than that of -re (46.29%, n = 131) for the three papers during the whole decade (see Table 4).
Table 4: Analysis of -re vs.-er
en- vs. in-
The pair en-/in- occur in words like enquiry/inquiry, ensure/insure, etc. The spelling preferences of the three papers for en-/in- were also different, as can be seen in Table 5. From 2001 to 2010, China Daily used British en- with a higher frequency except for the first two years. In 2009 and 2010, the American spelling in- completely disappeared from China Daily.
Table 5: Analysis of en- vs. in-
By contrast, both en- and in- feature in The Standard, although the American spelling was somewhat preferred. Once again, the Taipei Times remained AmE-loyal and used more in- persistently over the decade. Overall, -in (64.91%, n = 222) was used at a higher rate than -en (35.09%, n = 120) in the three papers.
-mme vs. -m
Examples for this category include programme/program, gramme/gram, kilogramme/kilogram, etc. BrE spells these words with -mme while AmE uses a single -m at the end.
China Daily tended to prefer -mme before 2006, but -m afterwards. A similar pattern can be observed for The Standard, with 2005 being the year of its preference switch. After 2005, -mme completely died out, and the authority of the American convention -m was firmly established in the paper.
By contrast, Taiwanese journalists exclusively used -m at the end of related words. Not a single such word was spelled the British way in Taipei Times during the whole decade. Overall, the data in Table 6 also demonstrate the usage of the simpler -m (59.77%, n = 104) outnumbers that of -mme (40.23%, n = 70) in the three papers.
Table 6: Analysis of -mme vs. -m
Overall, the spelling practices of China Daily and The Standard displayed similar trends over the decade, and both witnessed a preference shift from BrE spelling to AmE spelling around the years 2005-2007, broadly speaking. In contrast, Taipei Times adhered to the American spelling system throughout the period. Its spelling scenario has been much simpler and more consistent than that of the other two papers.
As for the overall variation in the usage of British and American conventions, the American ones were used at a higher rate of frequency over the decade for all the six pairs without any exception, and this undoubtedly projects the dominant status of the AmE spelling system in these media despite the fact that the BrE conventions -our, -ll, -re, -en, -mme were all favored in both China Daily and The Standard in the first years of the decade.
3.2 Diachronic changes of English spelling preferences
Horizontally embedded in the data in each of the tables above are diachronic changes of English spelling preferences over the decade. These can be revealed when every year's percentages of BrE or AmE spellings are compared.
Spelling preference shifts in China Daily
Figure 1 shows the percentages of BrE spellings in China Daily from 2001 to 2010. Overall, with the exception of the en-/in- pair, whose usage tended towards BrE, there was a general shift from BrE to AmE spellings. Among four other pairs, three shifts occurred in the year 2007, and one in 2006. For the pair -ise/-ize, the paper's overall position remained relatively stable, following American usage throughout the period.
Figure 1. Frequencies of BrE Spellings in China Daily 2001–2010
However, the general BrE-to-AmE spelling preference change trend in China Daily is characterized by more instability. Before truly following the American system, this newspaper oscillated between the two spelling systems, in particular in the first half of the decade. The -ll/-l pair is the most typical example for such a state of flux, BrE spelling often being around 50% per year before 2008, after which it ultimately stabilized with a clear preference for single -l. The example of en-/in- exhibits a similar degree of unpredictability. Contrasting with the other five lines, the en-/in- pair in China Daily displayed a distinctly non-mainstream direction. Such states of instability and unpredictability in spelling choices suggest that China Daily probably lacked clear editorial policies with regard to English spelling conventions.
Spelling preference shifts in The Standard
The Standard showed a clear pattern in its spelling preference shift. Spellings of the six pairs all gradually moved away from BrE towards AmE. Three preference shifts occurred in the year 2005, two in 2006, and one in 2008. By 2010, all six spelling pairs preferred AmE usage, with the BrE alternatives totally absent. Contrasting with the more chaotic patterns for China Daily, the tidy graph lines in Figure 2 are more gradual, smooth and predictable despite minor fluctuations, and clearly indicate a steady decrease in BrE spelling usage.
Figure 2. Frequencies of BrE spellings in The Standard 2001–2010
Spelling preference shifts in Taipei Times
As shown in Figure 3, the spelling preferences of Taipei Times have remained firmly pro-American throughout the period of the study. For the four pairs of -our/-or, -re/-er, -ise/-ize, -mme/-m, the percentages of British spellings were consistently approaching 0% from 2001 to 2010; all words were spelled in accordance with the American spelling conventions except for one insignificant occurrence of -ise in 2002.
Figure 3. Frequencies of BrE spellings in Taipei Times 2001–2010
Though BrE spellings of the other two pairs (-ll/-l and en-/in-) did occur in Taipei Times, their frequencies never exceeded 50%, and even these spellings still moved towards a complete AmE usage by the end of the decade. Indeed, by 2009 no trace of BrE spellings could be found in Taipei Times anywhere in the corpus. It should be noted that no data were found in the corpus for the -ll/-l pair in 2003 or for the en-/in- pair in 2007 for Taipei Times, so their respective use frequencies in these two years have to be regarded as zero, but are better not compared due to the lack of data which might have resulted from the sampling. It is for this reason that the two lines in Figure 3 break at these two points on the timeline.
To sum up, English spellings of China Daily and The Standard have been noticeably Americanized over the decade. Though a few British spellings (such as en-) were still found in 2010 in China Daily, the trend towards American spellings is clear and inexorable. In contrast, Taipei Times experienced fewer spelling changes from 2001 to 2010 because it persisted in using more AmE spellings throughout the decade, although even here it is the case that English spellings in this paper have become even more Americanized during the period.
3.3 Discussion
The regional variations and changes identified in the current research might have different causes and consequences. English teaching materials that have been used in Mainland China in the past may have contributed to the preference for BrE spellings observed at the beginning of the decade. In the 20th century, most of the English teaching materials used in China originated from Britain. For example, the Linguaphone textbooks in the 1970s, the BBC television English series Follow Me in the 1980s, and New Concept English afterwards all had a strong BrE flavor, and exerted a great influence on English users in Mainland China.
For Hong Kong, its history of British colonization must have resulted in the prevalence of linguistic elements from BrE in the past, and thus is likely to have been an important factor contributing to The Standard’s now bygone preference of BrE spellings. Taiwan and the U.S.A have maintained close commercial, cultural and political ties for many years, and it is thus natural that Taiwan English spelling preferences have had a consistently American flavor.
Against this background of inter-newspaper variations, the diachronic spelling preference shifts took place between 2005 and 2007 for most pairs in China Daily and The Standard. As Crystal (Reference Crystal1997b: 359) points out, ‘[m]any factors contribute to the gradual spread of a language – chiefly political and military might, economic power, and religious influence.’ Given this complexity, the identification of possible causes may well prove elusive. The changes observed in the current study might have been the result of the interplay of many factors occurring simultaneously. The first important cause might be the possible changes of newspapers’ in-house editorial policies, which can play an important role in shaping and determining English formsFootnote 4. For example, it has been observed that ‘[M]ainland English-language newspapers such as the China Daily conform to an American copy-editing style.’ (Danielewicz-Betz & Graddol, Reference Danielewicz-Betz and Graddol2014: 26) In addition, closely related with editorial policies are the economic advantages of AmE spellings in saving writing effort, time, ink and paper. Such advantages might have helped perpetuate the spelling change direction in the three newspapers since simpler spellings save the cost of printing, which is a particular concern of newspapers. In addition to the contrasts of -mme/-m, -our/-or, and -ll/-l, opposites of -ou-/-o-, -oe-/-e-, -dgement/-dgment and the lack of diacritics in words like cafe and the lack of hyphens for many compounds also contribute to AmE's advantage in saving the learning and writing effort as well as the cost of printingFootnote 5.
Moreover, the accumulating effect of the United States as a superpower over the years and the spread of AmE in the regions studied here might be a fundamental cause. In many cases, AmE is a symbol of fashion. Hu (Reference Hu2004, Reference Hu2005) and Jin (Reference Jin2005) mention that the pedagogic model of English in China is currently dominated by native speaker-based models, especially American English. This domination seems to be reflected in the English spelling preference shifts identified here. This new trend has also been noted by some scholars.
The findings of the current study support the predictions and observations of previous scholars such as Bondesen, who predicted that ‘the changes will be a move from using British English spelling to using American English spelling because to date, this is the pattern that one sees happening all over the world, including England.’ (Bondesen Reference Bondesen2004: 114; see also Crystal, Reference Crystal1997a; Bolton, Reference Bolton2003) The current study also suggests that the situation described by Danielewicz-Betz & Graddol (Reference Danielewicz-Betz and Graddol2014: 26), in which ‘English spelling and usage in mainland China tends to be American-based, whereas in Hong Kong, a former British colony, it is British,’ is now rapidly changing towards a similar preference for AmE norms.
4. Conclusion
This study addressed an issue that had not been adequately treated before, i.e. the English spelling variations and changes in three newspapers in Mainland China, Hong Kong and Taiwan. The data indicate a general drift towards AmE spelling preferences in these newspapers.
The study also supports the claim that a region's English spelling variation is correlated with its historical context, and the consequent English orthography change is perpetuated by social changes, not only local changes, but also global ones - the elevated status of AmE as the global ‘prestige’ variety of English in this case. This kind of change has potential benefits. For example, if regional variations are eliminated by the most prestigious variety, this might serve to help ‘tidy up some of the anomalies [and] give greater consistency to the whole system.’ (Carney, Reference Carney1997: 67) This in turn may avoid possible confusion for learners of English as a foreign or second language. In the meantime, it seems reasonable to conclude that the rising prestige of AmE will inevitably continue to erode the global status of BrE, and cause a shift of language loyalty in a growing number of regions.
Acknowledgments
This study is supported by University Research Project (No.14Ba015) and by the Special Items Fund of Beijing International Studies University.
YONGHOU LIU is an Associate Professor at School of English, Beijing International Studies University, Dean of Center for Linguistic Semiotics. He obtained his Ph.D in Linguistics from Beijing Normal University with honours and did his post-doctoral research at the School of Anthropology at the University of Arizona. His primary research interests are Chinese sociolinguistics and applied linguistics, particularly the sociolinguistic study of address terms and language policies. Email: liuyonghou@bisu.edu.cn
YE ZHAO earned her BA in English Language & Literature from Beijing International Studies University and an MA in Linguistics from University of North Carolina at Chapel Hill. Currently, she works as a Chinese Medical Interpreter in US. Her main area of interest is syntax and endangered language documentation, with a particular focus on the syntax of Mandarin Chinese and the phonetics of an endangered Southeast Asian language Sgaw Karen. Email: zhaoye1025@live.cn
