Digression and indirectness in Japanese writing

Implications for education in the Asian century

Dilhara Darshana Premaratne, Australian National University [About | Email]

Volume 13, Issue 4 (Article 23 in 2013). First published in ejcjs on 15 December 2013.

Suggested citation: Dilhara Darshana Premaratne, "Digression and indirectness in Japanese writing: Implications for education in the Asian century." electronic journal of contemporary japanese studies volume 13 issue 4 (15 December 2013). Available at http://www.japanesestudies.org.uk/ejcjs/vol13/iss4/premaratne.html.


Rhetorical conventions used in a text could vary from culture to culture. Linearity, relevance and explicitness are important conventions in English writing. However, the rhetorical conventions observed in English may not necessarily be followed or valued in other cultures. Rhetorical differences could be particularly pronounced between Asian and non-Asian cultures. Such differences could have a significant impact on students in the Asian century due to the increased interaction in the field of education between Asia and the West, especially between Asia and the Anglosphere (United States, United Kingdom, Canada, Australia, New Zealand). This paper looks at the extent to which culture-specific rhetorical conventions are used in Japanese writing and discusses their implications for students from two perspectives: that of native English-speaker students learning Japanese and international students from Japan pursuing education in English-speaking countries.

Keywords: Japanese language, students of Japanese, Japanese international students, rhetorical conventions, digression, indirectness.


Rhetorical conventions used in a text could vary from culture to culture. Linearity and relevance are important conventions in English writing. While linearity expects the writer to move logically in a straight line towards an explicitly stated conclusion, without departing from the main line of thinking, relevance stringently limits the area to be covered, excluding everything that falls outside the limits of the topic. Along with linearity and relevance, the practice of signposting in English, used to guide the readers by making the order of the discourse explicit, reflects a rhetorical tradition where the responsibility falls on writers to make their texts readable (Clyne 1987; Duszak 1997).

The rhetorical conventions observed in English may not necessarily be followed or valued in other cultures (Connor 1996; Duszak 1997; Hinds 1990; Kaplan 1966; Nisbett 2005; Panetta 2001). For example, the reader-friendly rhetoric used in English contrasts with the contemplative rhetoric that is found in the German intellectual tradition. Therefore, science writing in German, Polish and Czech involves creative thinking which is used for the sake of truth instead of reporting research as in the Anglo-American tradition (Duszak 1997). Rhetorical differences could be particularly pronounced between Asian and non-Asian cultures. Oliver (1971:3) argues that rhetoric is culturally-based, and therefore, the nature of rhetoric in Asia cannot be conceptualised in terms of conventions proved appropriate in the West:

The East is not the West. Cultures differ, and minds, feelings, and intentions in differing societies intermesh in differing ways. Discourse occurs, or is constrained, under different circumstances and has different styles for different reasons. The standards of rhetoric in the West which have had a unitary development since their identification by Aristotle are not universals. They are expressions of Western culture, applicable within the context of Western cultural values.

Research shows that knowledge of the rhetorical conventions used in a text aids readers in comprehending the text and also in producing a similar text later. Of particular importance here is the rhetorical structure or superstructure of a text which refers to global text structure. Different text types, such as exposition, argumentation and narration, have different rhetorical structures. Regular confrontation with a particular text type determines reader assumptions of the rhetorical structure of that text type. It is this stored knowledge that is supposed to get activated when required to read or write the same text type later. Second language reading difficulty may therefore be due partly to mismatches in rhetorical structure, over and above language proficiency factors (Carrell 1984b; Connor 1996; van Dijk 1983). Similarly, differences in rhetorical conventions between languages are likely to pose challenges when writing in a second language. More importantly, rhetorical conventions that are valued and taken to signify sophistication and erudition in written discourse in the first language may signify quite the opposite when used in second language writing. Therefore, such culture-specific rhetorical conventions could disadvantage writers when required to function in the second language (Farrell 1996; Bliss 2001).

Rhetorical conventions used in Asian cultures are likely to have a significant impact on education in the Asian century. Due to the rapid rise of Asian economies and their growing global influence, interaction with Asia has become increasingly important for English-speaking countries in the new millennium. Similarly, due to the internationalisaton of scholarship and the widespread acceptance of English as the high-profile medium of education and publication, Asian students are keen to pursue education in English-speaking countries. These developments have not only created an interest in English-speaking countries in learning Asian languages, but have also dramatically increased the number of Asian students who go to these countries for education purposes. The number of students enrolled in tertiary education outside their country of citizenship increased more than threefold, from 1.3 million in 1990 to nearly 4.3 million in 2011. Asian students made up 53% of the total in 2011. In Australia alone, the number of Asian students engaged in higher education is expected to increase from 952,000 in 2005 to 1,755,000 in 2025, with the number of East Asian students making up the majority. (IDP Education Australia 2007; OECD 2013). In this context of international scholarship, exploring the challenges faced by students due to mismatches in rhetorical conventions used by Asian and Anglo-American cultures has become timely and pertinent.

This paper focuses on the rhetorical conventions used in the Japanese language due to the popularity of the language in English-speaking countries today, and the high number of Japanese students who pursue higher education in these countries. The paper looks at the extent to which culture-specific rhetorical conventions are used in Japanese writing and discusses their implications for students from two perspectives: that of native English-speaker students learning Japanese and international students from Japan pursuing education in English-speaking countries. To this end, ten newspaper editorials written in Japanese were examined using van Dijk’s (1996) framework for editorials, which is based on his observations of English editorials. Van Dijk’s framework was chosen as it consists of several categories that define the functions of the respective parts of the text, which facilitates the comparison of rhetorical conventions used in English and Japanese editorials.

Culture-Specific Rhetorical Conventions

The idea that rhetorical conventions used in a text could be influenced by cultural writing norms first came to light when analysing the writing problems of overseas students in an American university. It was found that the rhetorical structures (text organisation patterns) found in the native languages of learners may be negatively transferred to their English language compositions (Kaplan 1966). Analysing 600 examples of foreign students’ English compositions representing three basic language groups, Kaplan found that Arabic language paragraph development was based on a complex series of parallel constructions, both positive and negative. He found that oriental writing, on the other hand, was marked by indirection, and that paragraph development was characterised by “circles or gyres” that turned around the subject and showed it from “a variety of tangential views” without ever looking at it directly (Kaplan 1966:10). Kaplan found that in the Romance languages there was much greater freedom to digress or to introduce extraneous material than in English. Although Kaplan’s findings are based on student compositions and writing samples of one paragraph, they are significant because he was the first to find that different rhetorical structures are used in different languages.

Hinds (1983) examined Japanese expository prose and found that it had a specific rhetorical structure which did not exist in English expository prose. The primary data for his analysis were taken from a popular newspaper column which appears daily in the Asahi Shimbun, entitled Tensei Jingo, consisting of articles on a range of topics current in Japan. Hinds found that the texts displayed the common rhetorical structure for Japanese compositions known as ki-shō-ten-ketsu, which originated in classical Chinese poetry. Analysing the news texts, Hinds found that the topic was introduced in the ki stage, and continued in the shoo stage, but that an abrupt change occurred in the ten stage where a new topic was introduced, and was followed by a conclusion that was not decisive. Hinds perceived the ten stage as the intrusion of an unexpected element into an otherwise normal progression of ideas. By surveying reader opinions, Hinds found that this rhetorical structure was highly valued in Japan. This was a remarkable discovery that revealed the use of a unique rhetorical structure in modern Japanese prose. However, in spite of the significant contribution made by Hinds, his study is not wholly reliable as it was based on translated texts. Although the English texts used in the study had been translated sentence by sentence, Hinds noted that the quality of the translation may be a source of difficulty for this type of study. The difficulty was particularly noted in the case of cohesive devices which could be left out of a translation.

Kobayashi (1984) compared US and Japanese students’ use of general statements in essays. The study examined a total of 678 writing samples obtained from 226 students from four groups; US college students, Japanese advanced ESL students in the US, English-major Japanese students in Japan, and non-English-major Japanese students in Japan. Each student wrote three compositions involving narration and exposition. The first three groups wrote in English, and the non-English-major Japanese students in Japan wrote in Japanese. The results showed inter-group differences. While US students favoured the general-to-specific pattern, placing the general statement at the beginning, Japanese students writing in Japanese favoured the specific-to-general pattern, placing the general statement at the end. The two Japanese groups that wrote in English differed from each other as the Japanese students in the US wrote more like the US students, and the Japanese students in Japan wrote more like the non-English-major Japanese students in Japan. Kobayashi’s study is noteworthy because of its design, which included writers of L1 and L2 both in the native language and the second language settings. The results confirmed that cultural variation exists in rhetorical structures used in written discourse.

Matalene (1985) found in her students’ writing, features of a rhetorical structure that is unique to Chinese written discourse. The structure was the one characteristic of the classical ‘eight-legged essay’ or ba gu, wen, which had been the central genre of the public service examinations in China, from the fourteenth to the twentieth century. This rhetorical structure consisted of eight stages—the exposition of the theme, the discussion of its significance, the argument, a turn in the argument, and the carefully balanced concluding paragraphs. Although now considered as “stereotyped writing,” Matalene, found a structure similar to this in her students’ persuasive writing. It consisted of an opening description of a specific incident, a look back at the usually unfortunate history of the issue or practice, an explanation of the current much improved state of affairs, and a concluding moral exhortation.

Coe and Hu (1989) confirmed Matalene’s findings that Chinese students have difficulty keeping their English compositions coherent due to the difference between the direct linear structure of English discourse and the spiralling structure typical of Chinese discourse. They concluded that an awareness of the differences in Chinese and English rhetorical structures can assist Chinese students become more proficient in their English writing.

The impact of culture-specific rhetorical structures has also been examined in relation to reading comprehension. Eggington (1987) investigated the impact of rhetorical structures on comprehension and recall of native Korean speakers. The study used two texts, both written in Korean, one using the linear rhetorical structure and the other using the traditional Korean non-linear rhetorical structure. The study revealed that the participants were able to recall information equally well immediately after reading the texts, whether the text presented information using the linear or non-linear rhetorical structure; however, when they were asked to recall the same information a week later, their ability to recall information from the text that used the traditional non-linear rhetorical structure proved to be significantly better than from the linear text. The study showed that rhetorical structures have an impact on comprehension and memory even when reading in one’s first language.

Spyridakis and Fukuoka (2001) assessed the impact of rhetorical structures on a group of American and Japanese readers. The subjects of the study read and recalled either an inductively or deductively organised text that contained the thesis revealing the author’s point of view. Results revealed that Japanese subjects were able to recall more information when an expository text is organised inductively. Spyridakis and Fukuoka noted the relevance of these findings for document designers of Japanese texts, and suggested that document designers should consider using an inductive organisation when localising documents for Japanese readers. Although American readers in the study did not seem to be influenced by the rhetorical structures of the text, the researchers noted that the findings could turn out to be different with texts that were difficult and unfamiliar to the reader. The study confirms the findings of Eggington (1987) that the rhetorical structure of texts has a considerable impact on comprehension and recall.

In the meantime, expanding on his earlier study of 1983 and Kaplan’s study of 1966, and looking at East-Asian texts in terms of coherence, Hinds (1987, 1990) produced two significant works on East-Asian writing style. Hinds (1987) noted that the degree of communicative responsibilities between text writer and reader could vary across cultures. He observed that in many East-Asian cultures the responsibility in eliciting meaning generally rests with the reader, while in English speaking cultures the responsibility rests with the writer. Hinds identified the unity of a text as the area in which speaker and reader responsibility operates. Therefore, while unity is an important element in English writing, it also requires writers to use transition statements to help the reader piece together the thread of the writer’s logic which binds the composition together.

Hinds (1990) found a disparity in the logical order and rhetorical organisation of texts between East Asian and English writing. He explained that native English speaker writers and readers generally expect expository writing to use either deductive or inductive reasoning although the deductive style is the more common. In deductive writing, the main idea is explicitly stated at the beginning followed by supporting details, while in inductive writing, the main idea is explicitly stated at the end preceded by supporting details. However, the rhetorical organisation pattern followed in East Asian writing is neither deductive nor inductive, because the main idea is not explicitly stated at the beginning or the end. Instead it is merely implied towards the end, leaving the readers to interpret the thesis using their knowledge of culture and society. Hinds coined the word ‘quasi-inductive’ to describe this pattern of organisation. Texts written using the quasi-inductive style may appear disorganised and unfocused to native English speaker readers as they are unlikely to be able to understand the main idea conveyed by the text.

In addition to compositions of English as a second language students, the cultural variation in East Asian writing has been investigated in a number of different genres. Of these, some focused on academic genres used in higher education. Taylor and Chen (1991) analysed the introductory section of science papers written by Chinese and Anglo-American researchers and found that the Chinese scholars, writing both in English and Chinese, omitted or truncated the step of “reviewing previous research.” The study was conducted using a macro analytical framework known as Swales’ Move Structure Analysis (MSA) based on research articles written in English (Swales 1984). This finding was confirmed by a number of other studies conducted on the research article genre and doctoral prospectuses written in Thai, Indonesian and Chinese (Basthomi 2009; Jogthong 2001; Loi and Evans 2010).

Morales (2012) compared sixteen research articles written by Japanese and Filipino writers, eight from each culture, with special focus on how they handled the conclusion section of the articles. He used the Conclusion Model of Yang and Alison (2003 cited in Morales 2012) to make comparisons between the conclusion sections of the articles. The model consists of three moves: summarising the study; evaluating the study; and deductions from the research. Of these, the Filipino writers considered Move 2 as mandatory, while the Japanese writers considered Move 1 summarising the study as mandatory. Therefore, while the Filipino writers indicated the probable contributions that their research might have made to the field, the Japanese writers merely provided a brief summary of the main points of their research. Morales observed an affinity between this type of conclusion and the concluding stage in the Japanese ki-shō-ten-ketsu rhetorical structure.

In addition to academic genres, some writers analysed the rhetorical conventions used in business genres. Kong (1998) analysed thirty authentic business request letters from a variety of sources and companies in Hong Kong, using Swales’ Move Structure Analysis (MSA) and Rhetorical Structure Theory (RST). Swales’ Move Structure Analysis showed that Chinese letter writers show a preference for the Justification + Claim pattern while the English letter writers show a preference for the Claim + Justification pattern. The Rhetorical Structure Theory analysis also revealed a more complicated structure in the Chinese letters with a higher average number of propositions than in the English letters. The results showed that the Chinese and English business letters have a different rhetorical structure even though they express the same communicative purpose.

Connor (1988) examined the written correspondence between a Japanese and an American manager responsible for marketing training of a large US pharmaceutical company during a two year period. In the study, forty seven documents of written correspondence (letters, faxes and electronic mail messages) were analysed to assess directness and concern for interpersonal harmony. The findings showed that while the American manager suggested changes and even improvements through open criticism, the Japanese manager used more subtle ways of expressing his opinion.

Connor also did a detailed analysis of the argumentative strategies used by the Japanese manager in an important business report. She found that his strategies led to a misunderstanding when he made a funding request for a marketing training session for new salespeople. Instead of stating the claim directly, the Japanese manager only suggested extra funding and facilities, using tentative expressions along with hedges and modifiers. The American manager reacted negatively to his request which resulted in the Japanese manager not receiving support for the suggested improvements.

Since the early study of Hinds (1983) on rhetorical conventions used in Japanese print media, other scholars have investigated their use in both Japanese and Chinese newspapers. Maynard (1996) did a study of Japanese mass media texts to explore the rhetorical structure of persuasion employed by Japanese writers. To this end she analysed thirty-eight newspaper columns entitled Koramu Watashi no Mikata (“Column, My View”). The articles were written by different reporters and writers who worked for Asahi Shimbunsha, “the publisher of Asahi Shimbun,” a major daily newspaper in Japan. Maynard used the concepts of commentary strategy to examine how the writer presents his/her own view as it relates to the central message of the column. The study focused on four discourse levels—the location of the headline paraphrase or where the central message reflecting the writer’s view and or opinion appears within the column, the characteristics of column-initial paragraphs, the overall sequencing of commentary sentences in Column, My View and the sequencing of commentary sentences within each danraku (“paragraph”). Maynard found that the preferred rhetorical structure in Japanese newspaper texts places commentary sentences toward the conclusion in multiple levels of discourse. However, this study did not aim to find the overall rhetorical structure of the news texts.

Scollon and Scollon (1997) did a comparative analysis of eleven newspapers from Hong Kong and three from the People’s Republic of China. The data consisted of a Xinhua release about Li Peng’s meeting with George Bush that appeared in all fourteen newspapers. Four of the newspapers were in English while the remaining ten were in Chinese. The results revealed that the traditional qi-cheng-zhuan-he structure (consisting of introduction, elucidation of the theme, transition to another viewpoint and summing up) was found in both the Chinese and English newspapers.

Ramsay (2001) examined three news e-journal texts of considerable length (5-6 pages)—two written by L1 speakers of Mandarin, one in Modern Standard Chinese and one in English, and the third written by an L1 speaker of Australian English. Using Rhetorical Structure Theory (RST), Ramsay found that there was less emphasis on corroborating claims and on debates and contentions in the Chinese news texts. Ramsay further noted that this would pose a problem for Australian readers of Chinese news texts.

Among the text features identified as showing cultural variation between East Asian and English writing, three stand out as dominant: direct and linear style of writing versus indirect and digressive style; inductive versus deductive style; and reader versus writer responsible rhetoric. However, some scholars have refuted the concept of cultural variation in rhetorical conventions used in texts. Kubota (1997) challenged the view that Japanese texts are unique and are different from English texts by pointing to studies that revealed that inductive structures in Japanese essays were not generally preferred by Japanese academic staff. Kubota explained the contradiction as a fault in the way that some studies viewed language and culture. She pointed out that language and culture need to be viewed as dynamic, rather than as exotic and static. In addition, she drew attention to the influence of the West that Japan experienced during the nineteenth century as a factor that challenges the typical characterisation of Japanese prose.

Mohan and Lo (1985) disputed the indirectness that previous studies had observed in Chinese writing and the influence of the “eight-legged essay.” They claimed that the eight-legged essay is an ancient style of writing used by government officers for centuries in imperial China. According to Mohan and Lo, this pattern was replaced by a style based on spoken language in the early twentieth century. Also, they argued that both classical and modern Chinese styles taught in schools favour a direct rather than an indirect expressive mode.

Although Kubota is correct in viewing language and culture as dynamic, whether language and culture can be greatly changed within a short period of time is questionable. What is also questionable is whether rhetorical traditions in Japan would dramatically change under the influence of English writing as suggested by Kubota. Given that the Japanese rhetorical tradition was formed over centuries, principally influenced by China, it is not possible to say that this long-term historical tradition, deeply rooted in its culture, would change in a short period of time (a little over a hundred years) under the influence of English. Similarly, Mohan and Lo’s claim that modern Chinese writing may not have the culture-specific features of indirectness and digression due to the disappearance of the ancient Chinese style of writing and the teaching of a direct style of writing in contemporary Chinese schools may not be completely true given the veneration with which classical Chinese literature is still regarded in China.

As two positions have emerged regarding the cultural variation in the rhetorical conventions used in texts, it is not possible to conclude whether the traditional Japanese rhetorical framework, the ki-shō-ten-ketsu structure, is still observed in modern Japanese writing. Also, as this rhetorical framework has been identified so far in Japanese expository prose, it is important to find out whether it is also found in other types of Japanese prose. Therefore, the purpose of the current study was to examine whether modern Japanese writers use this rhetorical structure in opinion discourse.

Current Study

The current study is based on Japanese newspaper editorials. Editorials were chosen for the study for two important reasons. Firstly, previous studies had examined the rhetorical organisation of Japanese expository prose but not opinion discourse. Secondly, opinion discourse plays an important role in the writing required by students in higher education.

The editorials were extracted from the electronic versions of the Asahi Shimbun and the Yomiuri Shimbun. These newspapers were chosen as they are reputed, well established national newspapers in Japan, and can, therefore, be taken as the national standard for writing styles, including the rhetorical organisation of texts.

The editorials were analysed using van Dijk’s (1996) framework for editorials, which is based on his observations of English editorials. Van Dijk’s framework consists of three categories that define the functions of the respective parts of the text: summary of the event; evaluation of the event; and pragmatic conclusion. The summary of the event is a brief description of what happened and is generally factual. The evaluation of the event spells out what was good or bad, right or wrong about the event. The pragmatic conclusion is a response to the newspaper readers’ implicit question ‘What next?’, What are we going to do about this?’. Therefore, in this category, the writer is expected to give recommendations, advice and warnings. This category is action oriented and describes what the authorities should or should not do.

The editorials were analysed to find out the categories that formed the rhetorical structure of the text. This was done by determining the functions of the respective parts of the text and chunking the text into smaller units or categories based on those functions. Both linguistic and non-linguistic clues employed to demonstrate internal coherence were used in the process.

The Uniquely Japanese Way of Writing Editorials

The analysis of the Japanese editorials showed that the three categories specified in van Dijk’s framework—summary of the event, evaluation of the event and pragmatic conclusion—were used in all the Japanese editorials. However, two major variations were made to the framework by the editorial writers. The most important and prevalent variation was the insertion of a new category between the evaluation of the event and the pragmatic conclusion.

This category reported on a problem, situation or issue different to the topic that was introduced earlier and thereby shifted the main focus of the editorial. The new category was named the focus shift category based on its function. Due to the addition of this category, these editorials consisted of a four-part structure instead of the three-part structure observed by van Dijk in English editorials.

The other variation was the change made to the pragmatic conclusion in van Dijk’s framework. This category is supposed to be action-oriented and specify what the authorities should or should not do by offering recommendations, advice and warnings. In the examined editorials recommendations were made but without identifying the entities that should implement the measures, leaving the responsibility instead to a large political system

The results indicate that there is a uniquely Japanese way of writing editorials. This is reflected firstly in the four-part structure that emerged from the editorials which consists of the summary of the event, the evaluation of the event, the focus shift category and the pragmatic conclusion in that order. It is different from van Dijk’s framework in terms of the number and type of categories.

The Japanese four-part editorial structure appears to have a strong affinity with the ki-shō-ten-ketsu structure unique to Japanese prose. This structure which has its origins in the four-line Chinese poetry (Maynard 1998), is considered to be the most fundamental organisational structure in Japanese prose (Hosaka 1978). Hinds (1983) observed this structure in the expository prose he examined in Japanese newspapers. He found that in the news texts that followed this structure, the topic was introduced in the ki stage, and was continued in the shoo stage; an abrupt change occurred in the ten stage where a new topic was introduced, and was then followed by the ketsu stage that loosely tied up the previous topics without being decisive.

Similar to the traditional ki-shō-ten-ketsu structure, the focus shift category, which is the third category in the Japanese four-part editorial structure, shifts the focus from the central issue introduced and developed in the two previous categories. The focus shift category is therefore, similar to the ‘ten’ part in the traditional Japanese structure.

The uniquely Japanese way of writing editorials is also reflected in the way opinion is expressed in the conclusion of the editorials. As the solutions that are offered lack specificity and implementability, they give the impression of a hazy conclusion. This confirms the finding of Hinds (1983) that the traditional Japanese four-part structure has a conclusion that is not decisive.

The Japanese writer’s indirectness in allocating responsibility to authorities can be explained by the general reluctance for criticising others, especially those who are older or who have a higher social or economic status that is observed in many Asian cultures. Particularly, most East Asian cultures such as the Japanese, Chinese and Korean consider group harmony and collective values very important, and so even prefer to keep silent to criticising others (Gudykunst and Ting-Toomey 1988; Nisbett 2005). Although singling out individuals or institutions as responsible for implementing recommended measures does not amount to criticising others, it can be considered as a closely-related activity.

Gudykunst and Ting-Toomey attribute the above quality to the two key dimensions on which cultures vary in interpersonal communication—individualistic versus collectivistic and low-context versus high-context communication. In individualistic cultures, emphasis is placed on individuals’ goals, whereas in collectivistic cultures, group goals have precedence over individuals’ goals. This dimension plays a key role in Western and Eastern analyses of culture. While the individual-collectivism dimension defines broad differences between cultures, the low-and high-context dimension deals with cultural differences in communication processes. A high-context message is defined as one in which very little of the information is in the coded, explicit, transmitted part of the message. A low-context message in contrast, is one in which most of the information is in the explicit code. Most Asian cultures, such as Japanese, Chinese and Korean fall toward the high-context end of the continuum, while most Western cultures, such as the Anglo-American, German, Scandinavian and Swiss cultures fall toward the low-context end of the continuum.

The findings of the present study are significant because they show that digression is a part of Japanese writing not only in expository prose but also in opinion discourse. The traditional Japanese ki-shō-ten-ketsu structure had been observed before only in Japanese expository prose (Hinds 1983, Matalene, 1985). The fact that it was also found in opinion discourse shows the prevalence of the structure and its deep entrenchment in Japanese prose. It suggests that digression may be a characteristic feature in other types of Japanese prose as well. The more frequently Japanese people see this structure in their day-to-day reading the more likely they are to follow it in their own writing.

The findings of the present study are also significant because they show that digression and indirectness are still present in Japanese newspaper editorials of the twenty-first century. Kubota (1997) surmised that the traditional rhetorical features of Japanese prose are likely to have been influenced by the Westernisation process Japan underwent in the Meiji period (1868-1912). As editorials are a common newspaper genre it has the potential to be influenced by Western media. What this shows is that Kubota’s argument that language and culture need to be viewed as dynamic, rather than as exotic and static does not seem to apply to writing in all situations. Although editorials are an internationally well-known genre and are therefore likely to be internationalised, editorials in Japanese newspapers are written for the consumption of Japanese readers. Therefore, the kind of Westernisation that Kubota expected does not seem to take place in Japanese editorials which are targeted for a national readership.

The differences in rhetorical conventions observed in Japanese writing are likely to stem from the literary principles and social values that influence Japanese culture. Firstly, the Japanese rhetorical style which is derived from the composition of Chinese poetry is based on aesthetic principles unlike the English rhetorical style which is based on analytical principles. Also, the preference for indirect communication over explicitness and the moderate stance adopted in expressing opinion by the Japanese editorial writers could spring from the value attached to collective goals, humility, modesty and social harmony in Japanese society. Due to the moderate position taken by Japanese writers, they are able to avoid open criticism and finger-pointing at particular authorities. This strategy facilitates social harmony which is an essential feature in inter-personal relationships in Japanese society.

Implications for students

The Japanese rhetorical conventions observed in the examined news texts—digression from the main topic of discussion as reflected in the four-part structure of the Japanese editorials and the indirectness observed in expressing opinion—could pose different types of pedagogical challenges for students. Therefore, they are likely to affect native English speaker students who learn Japanese as well as international Japanese students who study in English speaking countries.

The observed rhetorical conventions are likely to pose challenges for native English-speaker students because reading comprehension is not solely based on the learning of vocabulary and grammar, or the understanding of the local effects of sentences or paragraphs. Research shows that reading comprehension is also determined by the understanding of organisational forms and rhetorical structures of written texts. This is more so when reading in a second language because there are culture-specific ways of organising texts (Carrell 1984a, 1984b, 1987; Nishida 1999; Shingal 1998; Stott 2001). Therefore, even when readers have adequate vocabulary and sound knowledge of the grammar and syntax of the target language, unfamiliarity with the rhetorical structures used in texts could hinder full comprehension of the texts.

Research shows that readers become familiar with rhetorical structures used in texts through regular reading. While each type of text—e.g. stories, fables, expository and scientific texts—has its own conventional structure, regular confrontation with a particular text type determines reader assumptions of the rhetorical structure of that text type. Knowledge about rhetorical structure thus stored in the memory of readers gets activated when they come across the same text type again (van Dijk 1983) and aids them in comprehending the text as well as in recalling it later (Carrell 1984b). Knowledge about text structure is therefore grounded in past reading experience (Ramsay 2001).

If so, it can be assumed that second language comprehension may become difficult due to mismatches between the rhetorical structure presumed by the target language text and that already internalised by the second language reader. Hudson (cited in Carrell 1984b) notes that second language reading difficulty may be due partly to mismatches in rhetorical structure, over and above language proficiency factors. Maynard (1998:32) stresses the importance of “grasping the overall framework of the text” when reading in Japanese as a second language. The reader’s ability to anticipate differences in how thoughts progress and “knowing what a Japanese text is likely and not likely to provide, helps in one’s overall comprehension of the text.”

From the above perspective it can be assumed that readers in a second language may expect to find in target-language texts the conventional text structures they have learnt from their first language reading experience. What this means for native English speaker students is that when rhetorical structures they are accustomed to in English discourse are not used in the corresponding Japanese language texts, coming to grips with those texts would be difficult for the students. It can therefore be anticipated that the focus shift category will pose problems for English native speakers who attempt to read Japanese texts in which this category is used. The reason for the surmise is that the addition of the focus shift category gives the general impression that in Japanese editorials, actors, events and issues change through the text, and that the focus does not remain with the main issue till solutions are presented to resolve that issue. The difficulty this poses is heightened as it may also be the longest category in the texts as observed in the Japanese editorials.

Duszak (1997) points out that texts that include digressions could appear complex, incoherent or redundant to readers because digressiveness places extra demands on the processing abilities of readers. The difficulties are augmented when the readers are not accustomed to that kind of rhetoric. They not only have to invest extra effort into the reading of the texts, but also find that they have still not grasped the main line of the writer’s thought. Duszak makes the discerning observation that it is not easy to translate digressive texts into languages which prefer linearity. What this means for native English students is that even when they find the meaning of all the words in a Japanese text that uses the ki-shō-ten-ketsu structure, they may still be unable fully to comprehend the text.

Based on the observations made in relation to the pragmatic conclusion category in the Japanese editorials, it can be anticipated that the conclusions of Japanese texts would also pose problems for native English students reading Japanese. The pragmatic conclusion category in the Japanese editorials shows that it ends the editorial without addressing the expectations of readers accustomed to English editorials. In van Dijk’s framework, this is where the writer’s final opinion is expressed and where the editorial writer is expected to provide an answer to the reader’s implicit question ‘What next? What are we going to do about this?’ by spelling out what the authorities should or should not do. Japanese texts with conclusions that are not decisive as in the examined editorials would therefore seem to end abruptly to native English speakers, without bringing the discussion to a proper closure. As both the current research and past research show that conclusions appear to be hazy in Japanese prose, coming to grips with a Japanese text may be challenging for native English speaker students.

The results of the present study also indicate that variations observed in text organisation and expressing opinion in Japanese writing can make it difficult for international students from Japan to enter the academic discourse community in English speaking countries. Swales (1990) defines a discourse community as a group that is founded on shared discoursal patterns and expectations. According to Swales, these may involve the appropriateness of topics and the form, function and positioning of discoursal elements, such as the global organisation schemata that determines how content is organised within texts. Duszak (1997) observes that discoursal phenomena could also relate to aspects such as degrees of explicitness and redundancy used by writers, and the amounts of background that is considered acceptable to establish relevance between ideas. These observations suggest that international Japanese students would be familiar with the shared discoursal patterns and expectations of the Japanese discourse community and are likely to follow discoursal phenomena that are valued in that community. Therefore, they need to familiarise themselves with the discourse styles and conventions used in the corresponding English discourse community, and use them appropriately in their writing. Inadequate knowledge of such communication styles and conventions could have a negative impact on their education (Leki and Carson, 1997; Raimes, 1991; Duszak 1997b).

Swales (1990) observes that people enter discourse communities through practice and training. This applies also to native English speaker students who are not familiar with the English academic discourse community. It can be argued that entering this discourse community would be more challenging for international Japanese students as they would be trained in discoursal phenomena valued in the Japanese academic discourse community. Therefore, while they need to learn the new conventions as any novice member, they also need to unlearn the familiar conventions they have been using in their own cultural settings. Farrell (1996) notes that culture-specific rhetorical conventions are not simply matters of style that can be easily altered, but an intrinsic part of established literary practices in the first language. Therefore, a first step towards entering the new discourse community is to understand how the conventions differ and re-learn what is appropriate for the new discourse community. For example, Japanese students should be aware that digression in text organisation and reticence in expressing opinion are not universal discourse phenomena, but rather discourse phenomena observed and valued in their own discourse community. This would help them acculturate themselves into English discourse conventions.

In addition, it is also necessary to raise the awareness of teachers about the variations that can occur across cultural discourse communities. In western education contexts, particularly those in the Anglosphere, the focus of teaching is mainly on helping students acquire subject-specific skills to advance in their particular academic fields. While students’ writing difficulties may not be given adequate attention by subject-specialist teachers, even when they are, the underlying problems may not be clear to such teachers. Consequently, writing difficulties of students are likely to be attributed solely to inadequate linguistic competence, with no recognition given to the inadequacy of their discourse competence.

Distinguishing linguistic issues from discourse issues is an important first step in helping Japanese students with their writing. It will help teachers understand why Japanese students sometimes do not express their opinion where it is required, or use factual information instead of opinion in such contexts, or weaken their argumentation by the insertion of seemingly extraneous or redundant material. Being sensitive to such underlying problems instead of merely labelling a piece of writing as incoherent or irrelevant would help them help students advance from novice to expert member within their new discourse community in a short time.


Basthomi, Y. 2009. Examining research spaces in doctoral prospectuses. TEFLIN Journal 20, no. 2. http://journal.teflin.org/index.php/teflin/article/viewFile/137/123.

Bliss, A. 2001. Rhetorical structures for multilingual and multicultural students. In Contrastive rhetoric revisited and redefined, ed. C. Panetta, 15-30. Mahwah, N.J.: L. Erlbaum Associates.

Carrell, P.L. 1984a. The effects of rhetorical organization on ESL readers. TESOL Quarterly 18: 441-69.

———— 1984b. Evidence of a formal schema in second language comprehension. Language Learning 34: 87-112.

———— 1987. Content and formal schemata in ESL reading. TESOL Quarterly 21, no. 3: 461-81.

Clyne, M. 1987. Cultural differences in the organisation of academic texts: English and German. Journal of Pragmatics 5: 211-47.

Coe, R.M. and S. Hu. 1989. Yinhan duibi xiuci yanjiu chutan. Waiguoyu (Shanghai waiguoyu xueyuan xuebao) 2:40-6.

Connor, U. 1996. Contrastive rhetoric: Cross-cultural aspects of second-language writing. Cambridge; New York: Cambridge University Press.

Duszak, A. 1997. Culture and styles of academic discourse. New York: Mouton de Gruyter.

Eggington, W.G. 1987. Written academic discourse in Korean: Implications for effective communication. In Writing across languages: Analysis of L2 text, eds. Connor, U. and R.

Kaplan. Reading, Mass.: Addison-Wesley.

Farrell, L. 1996. A case study of discursive practices and assessment processes in a multi-ethnic context. Journal of Pragmatics 26, no. 4: 267-89.

Gudykunst, W.B. and S. Ting-Toomey. 1988. Culture and interpersonal communication. Newbury Park, Calif: Sage Publications.

Hinds, J. 1990. Inductive, deductive, quasi-inductive: Expository writing in Japanese, Korean, Chinese, and Tai. In Coherence in writing: Research and pedagogical perspectives, ed. U. Connor and A. M. Johns. Alexandria: TESOL.

———— 1983. Contrastive rhetoric: Japanese and English. Text 3, no. 2: 183-95.

———— 1987. Contrastive rhetoric: Japanese and English. In Writing across languages: Analysis of L2 text, ed. Connor, U. and R. Kaplan. Reading, Mass.: Addison-Wesley.

Hosaka, H. 1978. Repooto, shooronbun, sotsugyooron no kakikata. Tokyo: Koodansha.

IDP Education Australia Limited. 2007. Global student mobility: An Australian perspective. IDP Education Australia Limited, Sydney.

Jogthong, C. 2001. Research article introductions in Thai: Genre analysis of academic writing. PhD diss., Department of Educational Theory and Practice Morgantown, West Virginia. http://wvuscholar.wvu.edu:8881//exlibris/dtl/d3_1/apache_media/L2V4bGlicmlzL2R0bC9kM18xL2FwYWNoZV9tZWRpYS82MDM1.pdf.

Kaplan, R.B. 1966. Cultural Thought Patterns in Inter-Cultural Education. Language Learning 16: 1-20.

Kobayashi, H. 1984. Rhetorical patterns in English and Japanese. TESOL Quarterly 18, no. 4: 737-38.

Kong, K.C.C. 1998. Are simple business request letters really simple? A comparison of Chinese and English business request letters. Text 18, no. 1: 103-41.

Kubota, R. 1997. A reevaluation of the uniqueness of Japanese written discourse: Implications for contrastive rhetoric. Written Communication 14: 460-79.

Leki, I., and J. Carson. 1997. Completely different worlds: EAP and the writing experiences of ESL students in university courses. TESOL Quarterly 31, no. 1:39-70.

Loi, C. K., and M.S. Evans. 2010. Cultural differences in the organization of research article introductions from the field of educational psychology: English and Chinese. Journal of Pragmat-ics 42: 2814-25.

Matalene, C. 1985. Contrastive rhetoric: An American writing teacher in China. College English 47, no. 8: 789-808.

Maynard, S.K. 1996. Presentation of one’s view in Japanese newspaper columns: Commentary strategies and sequencing. Text 16, no. 2: 391-421.

———— 1998. Principles of Japanese discourse: A handbook. New York: Cambridge University Press.

Mohan, B. A. and W. Lo. 1985. Academic writing and Chinese students: Transfer and developmental factors. TESOL Quarterly 19, no. 3: 515–34.

Morales, R. 2012. Conclusions in research articles: A Filipino-Japanese contrastive rhetoric study. Philippine ESL Journal 8. http://www.philippine-esl-journal.com/V8_A4.pdf.

Nisbett, R.E. 2005. The geography of thought. New York: Free Press.

Nishida, H. 1999. Cultural schema theory. In Theorizing about intercultural communication, ed. W.B. Gudykunst, 401-18. Thousand Oaks, CA: Sage Publications.   

OECD. 2013. Education indicators in focus. OECD Publishing.

Oliver, R.T. 1971. Communication and culture in ancient India and China. New York: Syracuse University Press.

Panetta, C. 2001. Understanding cultural differences in the Rhetoric and Composition Classroom: Contrastive Rhetoric as answers to ESL dilemmas. In Contrastive rhetoric revisited and redefined, ed. C. Panetta, 3-14. Mahwah, N.J.: L. Erlbaum Associates.

Raimes, A. 1991. Out of the woods: Emerging traditions in the teaching of writing. TESOL Quarterly 25, no. 3: 407-31.

Ramsay, G. 2001. Rhetorical styles and newstexts: Contrastive analysis of rhetorical relations in Chinese and Australia news-journal text. ASAA E-journal of Asian Linguistics & Language Teaching 1. http://www.arts.unsw.edu.au/languages/asaa_ejournal.

Scollon, R. and S. Scollon. 1997. Point of view and citation: Fourteen Chinese and English versions of the ‘same’ news story. Text 17, no. 1: 83-125.

Singhal, M. 1998. A comparison of L1 and L2 reading: Cultural differences and schema. The Internet TESL Journal 4, no. 10. http://iteslj.org/Articles/Singhal-ReadingL1L2.html.

Spyridakis, J. and W. Fukuoka. 2001. The Effect of Text Structure on Text Comprehension of Japanese and American Readers. STC Proceedings. http://www.stc.org/confproceed/2001/PDFs/STC48-000114.PDF

Stott, N. 2001. Helping ESL students become better readers: Schema theory applications and limitations. The Internet TESL Journal 7, no. 11. http://iteslj.org/Articles/Stott-Schema.html.

Swales, J. 1984. Aspects of article introductions. In Common ground: Shared interests in ESP and communication studies, ed. R. Williams, and J. Swales. Oxford: Pergamon Press.

Taylor, G., and Chen, T. 1991. Linguistic, cultural, and subcultural issues in contrastive discourse analysis: Anglo-American and Chinese scientific texts. Applied Linguistics 12, no. 3: 319-36.

van Dijk, T.A. 1983. Discourse analysis: Its development and applications to the structure of news. Journal of Communication 33, no. 2: 20-43.

———— 1996. Opinions and ideologies in editorials. http://www.discourses.org/Unpublished/Opinions%20and%20ideologies%20in%20editorials.htm.

About the Author

Dilhara Darshana Premaratne, PhD (Sociolinguistics), MA (Higher Education), is currently undertaking further postgraduate studies in translation at the Australian National University. His research interests are tertiary teaching and learning, contrastive rhetoric, language change and language policy and planning.

Email the author

Back to top