Written by Ben Goren.
Long-term foreign residents of Taiwan who have a certain level of linguistic fluency in the nation’s official language of Mandarin, will likely at some point have found themselves stymied by conversations which are suddenly interspersed with words, idioms, and vernacular derived from some of Taiwan’s other languages. It is at this moment that they are faced with the reality of Taiwan as a multi-lingual society, the linguistic diversity of which being a result of contingent historical and political forces that have shaped who speaks what, when, and to whom. From personal experience of living in Southern Taiwan, I can attest that while Mandarin is a basic and necessary communication tool, speaking even a little Hoklo (also known as Taiwanese) is a key to forging closer relationships and friendships. Yet, this is not the case in every part of Taiwan and cannot be regarded as a steadfast rule. Indeed, there are occasions when speaking Hoklo rather than Mandarin can be counterproductive and even make a poor impression on the speaker’s audience. Much like a northern accent in the UK can have negative associational impacts when living in the southern Home Counties, language stereotypes in Taiwan can act as barriers and filters that incite and invoke subconscious and culturally embedded prejudices that divide according to class and political identities.
An examination of these linguistically embedded stereotypes was the subject of a recent experimental study by Chang Yu-tzung and Lu Jie.  The authors sought to demonstrate, using a bi-lingual comparison between Mandarin and Hoklo, the language stereotypes that exist in Taiwan and to draw out the implications for Taiwanese public opinion and democratic politics. What they found was that language divisions were not just proxies of Taiwan’s regional divisions, but that Taiwanese cognitively differentiate between the political and socio-economic implications of the spoken languages.
If our eyes are regarded as ‘windows to the soul’, then the language we speak, and how we speak it, is surely a window to our culture, education, socialisation, the location of our upbringing, and an indicator of how we wish others to perceive us. It also signifies how we perceive the world, and by extension, is a clue to the ideological matrix we employ to derive meaning and construct strategies for managing relationships and our position in society. Chang and Lu’s paper fills a gap in academic discussion of how substantively meaningful cues are attached to spoken languages in Taiwan, cues which are then decoded and activated during socio-economic and political interactions, in the process shaping the formation of attitudes and behaviour. Importantly, they provide a concise and extremely useful overview of how the retreating Republic of China polity sought to ‘Sinocize’ Taiwan after arriving as exiles in the country. While there has been substantial analysis of the Japanese policies of cultural assimilation, also known as the ‘Kominka Movement’ (皇民化運動) less effort has focused on how the Chinese Nationalist Party (KMT) carried out exactly the same kind of policies with the intention of installing Mandarin and Chinese culture as the dominant ‘currency’.
According to mainstream ROC narratives, post-war Taiwan was an impoverished island whose subjects were dangerously close to losing touch with their Chinese heritage and identity. Such was the level of oppression and indoctrination that occurred under the Japanese ‘occupation’ most Taiwanese had come to a ‘mindless accommodation’ with speaking and writing Japanese, and Japanese was the normalised and ubiquitous language of education, administration, and business. To a KMT that had suffered heavily at the hands of the Japanese, this was an unacceptable manifestation of Japanese influence that had to be erased. The narrative that claimed Taiwan had ‘retroceded’ to China could only maintained if the physical legacies of the Japanese Colonial Period were ‘cleansed’ from the landscape, and if the lingua franca of Taiwan became Mandarin – in all forms and in all spaces. Chang and Lu’s observation that “language use or language choice is not naturally produced, but the result of political engineering“ is therefore a critical component in understanding how the policies of the colonial ROC regime changed the linguistic landscape in Taiwan to serve as a foundation for both the ideological and economic legitimacy and authority of the ROC, the KMT, and the new immigrant ruling class. The paper reminds the reader that the significance of linguistic cleavages for political mobilisation in Taiwan cannot be overlooked, “…given that Taiwan’s linguistic cleavages reinforce, rather than crosscuts, existing cleavages among its ethnic groups.”
Between 1946 and 1987, the KMT instituted language polices that banned the publication of magazines and newspapers in Japanese and relegated Hoklo to a local dialect in the hope that it would eventually die out. It banned use of Hoklo in schools and public places, restricted Government positions to people who could or would speak Mandarin only, banned bibles written in Roman alphabets, and prohibited the use of Hoklo subtitles in film and TV. The net effect of these policies was to institutionalise discrimination based on language use, which in turn excluded Taiwanese from full participation in their own political economy and “planted the seed for future eruption of negative sentiment against Mandarin among Taiwanese-speakers.” By the time Taiwan democratised, Mandarin was spoken and written by 90% of the population. Not only did these language policies harm and possibly permanently undermine the sustainability of the Hoklo language, they also seriously eroded the cultural bases and identities of Aboriginal and Hakka communities, preventing their transmission of cultural experience and knowledge. The KMT’s Mandarin policies have arguably been more severe and more destructive than equivalent Japanese attempts at colonial assimilation. Both the Japanese and KMT language programmes were designed to achieve a kind of ethnic linguistic cleansing and both had the effect of creating clear socio-economic and political cleavages based on education, language, occupation, class, and political affiliation (though the KMT policies by virtue of being instituted via bloody Martial Law led to a clearer political divide based on language than had existed during the Japanese Colonial Period). Chang and Lu succinctly note the depth and impact of the language policies:
“In summary, before Taiwan’s democratization, local languages and culture (e.g., Taiwanese and Hakka folk songs and traditional opera) were deliberately discriminated against and denigrated by the KMT regime as “inferior.”At the same time, the regime established and implemented an efficient reward/punishment system (with penalties as the main element) to promote the use of Mandarin and discourage the use of any local dialects including Taiwanese. Moreover, complementary to its Mandarin Promotion Policy, the KMT government also established a variety of socioeconomic, bureaucratic, and political institutions that intentionally and systematically discriminated against Taiwanese speakers in economic activities, bureaucratic recruitment and promotion, and political participation.”
Post-democratisation, the internalisation of these policies, and their reproduction by the media have, Chang and Lu argue, led to a situation where “residents of Taiwan and even scholars of Taiwanese politics and society have taken the politically engineered and socially reproduced socioeconomic and political implications of various spoken languages, particularly Mandarin versus Taiwanese, for granted.” As such they posit that a set of socioeconomic and political attributes have been systematically associated with various spoken languages. This should be of prime concern to anyone interested in ethnic and identity politics in Taiwan, especially those discussing the viability and direction of Taiwanese democracy. Given a dearth of existing studies in this field, the authors argue that it is important to examine if and how language plays a role in shaping Taiwanese perceptions of themselves and each other, how they engage with their democratic institutions, and how they express their identity.
A major hurdle would be to design a means to measure language stereotypes where the subjects themselves would not consciously or subconsciously invalidate the results. To this end, they asked subjects to listen to three different speakers reading the same passage in different languages – A ‘mainlander’ speaking Mandarin and Hoklo Taiwanese, a native Taiwanese speaking Mandarin and Hoklo Taiwanese, and a young male speaking Mandarin as a filter. The subject matter of the passage was the nutritional value of popcorn, chosen for its political and socioeconomic neutrality. The subjects were then asked via questionnaire to evaluate the speakers’ socioeconomic status, political preferences and other personality features.
The results of the experimental study confirm to a large extent the existence of the language stereotypes Chang and Lu had suspected were prevalent in Taiwan. They note that:
“…language stereotypes do exist in contemporary Taiwan; and they are widely received and consistently decoded by many Taiwanese citizens… Basically, the salience of and consistency in decoding such language stereotypes vary across issue domains. Regardless of the origin of the bilingual speakers, the switch between Taiwanese to Mandarin, ceteris paribus, significantly altered the experimental subjects’evaluations of their possible political preferences in a consistent way.”
The Mainlander speaking Mandarin was perceived to have a higher educational attainment and social class, to be more honest and self-confident, and slightly more capable as a leader. At the same time he was regarded as less generous. His political preferences were perceived to be pro-KMT, more likely to have voted for the KMT candidates in the 2008 Presidential elections, more likely to approve of President Ma’s performance, and more enthusiastic about unification with China.
The Taiwanese speaker was evaluated more positively in terms of his educational attainment, social class, and monthly income when he spoke Mandarin. In terms of personality traits, he was perceived as less honest, less generous, but more confident in himself and his leadership ability. Again, when using Mandarin the Taiwanese speaker was assessed to be more loyal to the KMT, more likely to have voted for the KMT candidates in the 2008 elections, and again more amendable to President Ma’s performance, and to the possibility of unification.
In contrast, when the speaker used Hoklo Taiwanese, they were consistently perceived to favour independence, have a lower monthly income, and lower social status.
In short, Chang and Lu found that while changes in the evaluation of the speaker’s personality traits were mostly statistically insignificant, regardless of the actual origin of the bilingual speakers, when they switched to Hoklo Taiwanese, it produced “significant and positive changes in our experimental subjects’ evaluations of the speakers’ socioeconomic status and political preferences”. The results of the study, Chang and Lu argue, suggest that language stereotypes in Taiwan contain implicit information which is “consistently decoded for political issues, less so for socioeconomic issues, and almost insignificant for personality features.” They conclude by saying that:
“the switch from Taiwanese to Mandarin, ceteris paribus, sends some important information that indicates a speaker’s possible higher monthly income and social class, and places the speaker closer to the KMT in contemporary Taiwan’s political spectrum. And such implicit information is widely received and consistently decoded by many Taiwanese citizens.”
This implicit information, Chang and Lu explain, is embedded in daily discourses. In terms of understanding how Taiwanese communicate, this study provides substantive empirical arguments that Taiwanese judge, and are judged, by the language they speak in terms of their socio-economic status and even more so their political affiliations. This finding is reinforced by electoral analysis demonstrating how KMT politicians such as President Ma and former Vice-President Lien have adopted Hoklo Taiwanese during election campaigns, and particularly when canvassing voters in the south of the country where Hoklo Taiwanese is more prevalent in its day to day usage. Interestingly, the authors take their analysis a hesitant step in this direction by claiming that it is possible language stereotypes could be activated and strategically mobilised under a wide range of conditions, thereby affecting political attitudes and behaviour amongst Taiwanese, perhaps playing a role in Taiwan’s democratic consolidation. Although this is not the main remit of this paper, and an argument can be made here that the authors have somewhat overreached, Chang and Lu should still be commended for seeking wider implications to their study, especially given the clear results regarding language stereotypes and political affiliation. If anything, it is a pity that their study did not expand on this issue.
A good example of language playing a role in democratic consolidation could be the popularity of the Hoklo Taiwanese song ‘Island’s Sunrise’ by southern Taiwanese band Fire Extinguisher, which shot to national prominence and popularity when it became the ‘theme song’ for the recent 318 Movement protests and occupation of the Legislative Yuan. This is an example of how pro-Taiwan and pro-democracy activists choose to utilise Hoklo Taiwanese to express their dissent against an imported Mandarin speaking KMT elite who they feel have constrained Taiwanese democracy for their own benefit.
Chang and Lu note that their analysis has strong implications for the vast majority of surveys conducted in Taiwan which that collect data through face to face surveys or by phone interviews. They ask whether, if language contains implicit codes and political salience, would that not affect the results garnered especially on sensitive subjects such as racial attitude and political identity? Could it be the case that most data gathered on subjects such as political identity, party affiliation, cross-strait relations, political trust, and democratic performance in Taiwan to date has been unwittingly contaminated by the researchers and is therefore biased and misleading? If this is the case, this has serious implications for Taiwan’s democracy in so far as such surveys are used as critical tools by political parties and social movements, without and outside of Taiwan, to justify and legitimate their actions / viewpoints. In fact, it could be that Taiwanese opinion on a range of issues has been categorically and systematically misrepresented for a long period of time, rendering even respected longitudinal studies on national and party identification, such as that carried out by the NCCU Election Centre, hopelessly inaccurate.
Herein lies the great contribution of this paper to the study of language and methodology of researchers in Taiwan. In reminding us that language is not neutral, it asks us to cast a critical eye over the data academics use to discover patterns or understand developments in the evolution of Taiwan’s democratic, political, and linguistic landscapes. It is a reminder that our maps don’t always fit the territory and that unless we are careful, our maps can become tools used by others to shape the territory rather than serve as accurate representations of what is actually there.
 Chang Yu-tzung and Lu Jie (2014) Language Stereotypes in Contemporary Taiwan: Evidence from an Experimental Study. Journal of East Asian Studies 14(2): 211-248.