Textual Features of Intercultural Internet Chat between Learners of Japanese and English


Recent studies have shown that Information Communication Technologies such as chat can create a natural context for learner independence, and a means of extending opportunities for interaction, seen as important in major theories of Second Language Acquisition. However, as a developing variety of interaction, Internet chat has unique linguistic and interactional features that are distinct from either oral or written communication. This study aims to examine the textual features of Japanese? English intercultural Internet chat through one specific software application, MSN Messenger, and also to explore the ways in which non-native speakers use this medium as an opportunity for communication with native speakers, and language learning.


This study aims to examine the textual features of Japanese-English intercultural Internet chat through one specific software application, MSN Messenger, and also to explore the ways in which non-native speakers use this medium as an opportunity for communication with native speakers, and for informal language acquisition. It appears that there have been very few studies of chat examining naturally occurring "authentic interaction" where participants are motivated by reasons other than teacher-directed language learning to participate in chat conversations. Tudini (2003) found that some students engaged in learner-only chat complained that they wanted "someone they could learn from". All of the participants in this study utilized online chat for social interaction, of their own volition, and in their own homes, with native speakers of the target language.

Based on the goals of this research, the following research questions were formulated:
  1. What are the features of Japanese-English intercultural Internet chat?
  2. How does intercultural Internet chat provide opportunities for communication with native speakers and informal language acquisition?


Hata (2003) identifies three main areas of current research within CALL:

  1. Characteristics of Computer Mediated Communication
  2. Applying SCT using IT in language classrooms
  3. The negotiation of meaning using CMC

A number of studies, such as Hanna and de Nooy (2003), Payne and Ross (2005), Fitze (2006), Thorne (1999), Kitade (2000), Belz, (2002), Spiliotopoulos and Carey (2005), Jepson (2005), Toyoda and Harrison (2002) and Smith (2004), have investigated the effects of CMCs on second language learning, especially conversations between non-native speakers in a classroom setting, yet very few examine conversations between native and non-native speakers outside the classroom. A large percentage of the body of recent research on CMC has concentrated on the features of the different varieties of synchronous communication, such as chat, and asynchronous communication, such as bulletin boards, including Hanna and de Nooy (2003), Payne and Ross (2005), and Fitze (2006). Features documented include code switching, typing errors, non-linguistic cues, inappropriatesegmentation of sentences, and strategies for expressing emphasis.

Research Methodology

In order to examine the textual features of Japanese-English intercultural Internet chat, ten participants (five Australian student volunteers and their five Japanese chat partners) were asked to record their naturally occurring chat conversations for the period of approximately one month. All ten participants used the popular instant messenger (chat) program MSN Messenger, created by the Microsoft Corporation. To shed light on the histories and motives of participants, immediately following their final chat conversation at the end of the month, the Japanese participants completed a brief follow-up interview via chat with the researcher, while the Australian participants completed a face-to-face follow-up interview. This study utilizes three primary sources, the chat logs, the chat follow-up interviews, and the face-to-face follow-up interviews. A table summarizing the characteristics of the participants is given below. Participants have been assigned pseudonyms to protect their identity. All of the Japanese participants studied an advanced level of English, and all of the Australian participants were advanced students of Japanese.

Characteristics of the Participants

      Language Study Chat Background Pair Characterization
Nadia Lower Advanced Beginning light user Education Students
Tomi Upper Advanced Experienced light user
Kaylene Upper Advanced Experienced light user Language Exchange Partners
Ruriko Lower Advanced Beginning light user
Evelyn Lower Advanced Beginning heavy user Former Host Sisters
Azusa Lower Advanced Experienced light user
Jacob Lower Advanced Experienced heavy user Social Chatters
Miku Upper Advanced Experienced heavy user
Phia Upper Advanced Experienced medium user Former Classmates
Soichiro Lower Advanced Experienced medium user


Input Method and Orthographic Switching

Many participants, especially those new to bilingual chat, experienced difficulty switching between the different input methods for English and Japanese writing. Typing in Japanese on a standard keyboard requires the use of software called an Input Method Editor, or IME. The IME allows the user to type the word, using the roman alphabet (romaji), which will then appear in the phonetic symbols hiragana. If desired, the user may convert the hiragana to Chinese characters (kanji) by selecting the appropriate candidate from a list of compounds that are a phonetic match. This selection is achieved by using the mouse or hitting Enter, and is called the kanji henkan. There are a number of ways to switch between Japanese and English, including keyboard shortcuts and using the mouse to click on the IME icon in the toolbar, and participants used a surprisingly wide variety of strategies to manage their orthographic switching, with mixed success.

Australian Participant's Orthography Switching

Participant Strategy Success(based upon number of related errors and participant’s self-evaluation)
Nadia Mouse Beginning light user
Kaylene ALT+~ Limited
Evelyn CAPITALS Medium
Jucob F10 High
Phia Avoidance High

Kaylene was not only very successful at orthographic switching, but was able to assist her Japanese language exchange partner Ruriko in one of their chats, suggesting that it is not only non-Japanese users who have difficulty navigating the IME. Their conversation is reproduced below, with secondary topics removed.

Language Exchange Partners Kaylene and Ruriko, 04/05/2006

Time From Message
9:57:08 PM 瑠璃子 I am just looking for
9:57:27 PM 瑠璃子 how set Japanese!!
9:57:33 瑠璃子 I forgot!!
9:57:45 瑠璃子 tyottomattetene:-○
          ((Just wait a little while:-○))
9:59:07 PM kay miró ローマ字不便だね...
          ((Romaji is inconvenient))
9:59:37 PM 瑠璃子 うん、セッティングが難しい!!
          ((Yes, the setting is difficult!!))
10:00:51 PM kay miró You know if you press the ALT and~key that will shift between Direct input and ひらがな
          ((You know if you press the ALT and~key that will shift between Direct input and hiragana))
10:01:19 PM 瑠璃子 oh!
10:01:26 PM 瑠璃子 that's easy!!

Evelyn used a different orthographic switching strategy in her chat conversations with former host sister Asuza, writing English in capitals rather than switching the IME to English, as in the example below. She explained in the interview "for some reason, when I'm typing in Japanese, if I do it in capitals, it will come up in English". Evelyn concluded, "it's just easier for me to write in capitals".

Time   From   Message
20:46:35   Wallaby   GENGAと言う所
              ((A place called GENDA))

Social chatter Jacob knew a keyboard shortcut to switch between the different orthographies in his chat with Miku. In the interview, he said he "hit F10 or something, to turn it over to English, and hit enter to get out of that". Jacob navigated the two input methods skilfully, employing orthographic-switching mid-sentence a number of times, as he explained in the interview, to compose words in English that he did not know in Japanese in a Japanese sentence, or type words in Japanese that he did know into an English sentence that he could not otherwise have written in Japanese. On the other hand, in her interview, Phia reported trying to avoid orthographic switching in her conversation with former classmate Soichiro. She said "if I knew the keyboard shortcuts, I'd use them, but I don't, which really annoys me" so "when I've switched into the Japanese IME, it's much easier just to write, go on in Japanese". Phia stated that switching frequently would "infuriate" her, and for this reason, she said, "I've got to try and stick to one language or the other".

Language Choice and Code Switching

In addition to orthographic switching issues, participants had many other reasons to code switch during their chats. In their predominantly Japanese chat, education student Tomi used English mid-sentence to write the names of the courses she was completing at an English-speaking university. She stated in the interview that she felt writing the name of the course in katakana would have been difficult to understand. In her interview, Nadia explained that she followed Tomi’s example, and wrote the name of her workplace and hometown in English as she thought "it would be easier for her".

Kaylene and Ruriko code switched frequently, which appears to be due to the nature of their relationship as language exchange partners. Ruriko explained in her interview that she felt using only Japanese was not good, which on one occasion, even prompted her to translate her response directly afterwards, 「なるほど!! I see!! 」 (I see!! I see!!). Ruriko also expressed feeling relief when her partner Kaylene employed mid-sentential code switching also.

Former host sister Azusa code switched at her chat partner Evelyn's request. Evelyn stated in the interview that she "wanted to speak Japanese", yet used English for words she did not know in Japanese, rather than using a dictionary, in order to maintain Azusa's interest in the conversation by not slowing the pace. Evelyn also code switched to English when she felt Azusa's attention was waning, to rekindle her interest in the conversation.

Social chatter Miku, who had lived in Australia for a number of years at the time of the study, switched between Japanese and English frequently. Miku was so accustomed to code switching that in the interview, she was unable to explain her reasons for doing so. On the other hand, her partner Jacob was not used to such rapid code switching, having only conversed with monolingual chat partners prior to his interaction with Miku. In the interview, Jacob said that he noticed himself "starting to pick up the like, half-Japanese, half-English sentence structure, and I think I tried to use it later on".

Phia and Soichiro, former classmates, predominantly used Japanese in their first chat log, and predominantly English in the second. Soichiro used English in parentheses after particularly difficult Japanese words, yet Phia said in the interview that she would only use English if she really came “to a standstill”, instead preferring to use a Japanese word that she did know. Phia’s efforts to use Japanese may have been due to her difficulties code switching.

As the discussion above shows, orthographic switching can be a useful tool for chat participants to effectively code-switch, and use the lexical items and grammar of the second language they know, in concert with elements of their first language. A limited knowledge of orthographic switching strategies, however, may seriously impede participants' willingness to attempt interaction in their L2, if the conversation is predominantly in their L1, even if language learning is their primary goal. Conversely, participants' willingness to attempt a discussion of more complex topics that require the use of their L1 may also be adversely affected if the conversation is predominantly in their L2.


Symbols were used to express both emotions and prosody in the chat logs. Five strategies to express emotion were identified, including Emoticons, text-based art, laughter, asterisks and font symbols. These strategies were of particular importance to participants, such as Kaylene, the language exchange partner, who used a wide range of emotion expressing strategies, and stated in the interview, "that's the one area I think chat really falls short on, because you don't have, mainly, the ability to actually have them see your facial expressions. To know that you really feel for them". ("Kaylene", Interview, 12/05/2006)


Participants only used 18 of the 75 standard Emoticons available with MSN Messenger. Emoticons are listed in order of usage below. Of the 18 standard Emoticons used, 16 were emotion-expressing Emoticons, predominantly "smiles" depicting facial expressions and feelings, and a red heart, representing love. One experienced participant, Miku, a social chatter, used a downloaded custom Emoticon of a superhero in her chat with Jacob.

In their interviews, education students Nadia and Tomi, language exchange partners Kaylene and Ruriko, former host sister Evelyn, and social chatters Jacob and Miku made explicit mention of half of the Emoticons used. Their comments highlighted the ways in which Emoticons were used in the chats to show the nature of the conversation, express emotions, represent one's facial expressions, reassure, perform acts, show enthusiasm, provide emphasis, and even show sympathy. On some occasions, Emoticons were used to emphasise emotions expressed, such as a red heart after a participant exclaimed 「やった!!楽しみ」(Yay!! Looking forward to it). On other occasions, emoticons were used instead of verbal reference, such as a confused smile on its own, to represent the participant’s facial expression after asking what they described as a "stupid" question.

Emoticons Used in the Ten Chat Logs

Keystrokes Description Emoticon Total Use
:) Smile 45
:( Crying Face 22
:D Open-Mouthed Smile 14
;) Winking Smile 10
:$ Embarrassed Face 10
(L) Red Heart 9
:( Sad Smile 7
:P Smile With Tongue Out 4
:ˆ) I Don't Know Smile 2
*-) Thinking Smile 1
:-○ Surprised Smile 1
:S Confused Smile 1
Party Smile 1
ˆo) Sarcastic Smile 1
+o( Sick Smile 1
|-) Sleepy Smile 1
(au) Auto 1
(@) Cat Face 1
*sh007 CUSTOM Superhero 1

Text Art

The only non-Japanese participant to use text art faces, Kaylene, the language exchange partner, commented in her interview that she liked text art because they did not fill up the screen as much as Emoticons, and she felt "silly for having that many Emoticons". Most of the Japanese participants used text art, often called kaomoji (letter faces) in Japanese. Essentially there are two kinds of text art, Western text art, in which smiles are rotated 90o and appear vertical, such as :), and Eastern text art, in which smiles are not rotated and appear horizontal, such as ^_^. In MSN Messenger, typing most Western text art combinations will result in the keystrokes :) appearing as the Emoticon . For this reason, the vast majority of text art used in the chat logs collected was Eastern kaomoji. The only example of a 90o-rotated text art was a non-standard 'Winking tongue-poking out face' created by Kaylene. The text art produced by participants in the current study is displayed below.

Text Art Use in the Ten Chats

     Participants' Description Total Use
^_^ Smile(Kaylene) 2
(^ ^) (Variation of smile) 2
^ ^ (Abbreviation of smile) 12
^^; Embarrassed smile(Kaylene) 9
^_^; (Variation upon embarrassed smile) 1
(′_`) Negative feeling(Ruriko) 1
(>Д<) You idiot(Kaylene's interpretation of Ruriko's face) 2
(>_<) Tired/Trouble/Bad/Stressed(Miku) 2
(><) Abbreviation of tired/trouble/bad/stressed) 2
f(^^ Hand Scratching a head with bitter smile sayingえ~
(;^^ (Abbreviation of cold sweat) 2
;P Winking tongue-poking out face(Kaylene) 4
Additional information regarding variations and abbreviations(Dunn, 2006;Takagi, 2006)

Kaylene displayed a thorough knowledge of the constituent symbols that make up text art faces, both Eastern and Western, in the interview. The meanings she gave for the different parts of the kaomoji ^_^; are outlined in the table below.

Text Art Constituent Symbols

Part Meaning
^ Curved eye
- Mouth
; Sweat drop(See diagram below)

The diagram below depicts the text art representation of a sweat drop. Kaylene explained the text art as "like in anime, when they have a sweat drop, coming down the side of their face". The original drawing was produced by Kaylene in the interview, and the translucent red circumflexes, and the semicolon, were overlaid on top of the original drawing by the researcher.

Text Art Drawing

Furthermore, Kaylene described how she had created her own text art, ;P, "It's a cross between a wink, with a smile, and the smiley face where you stick your tongue out. I basically use this to express sarcasm, but sometimes just to express the fact that I'm joking around. It's something that I think I made up, I don't think I've seen it anywhere else, but I doubt that I'm the only one that uses it." ("Kaylene", Interview, 12/05/2006)


Laughter was also represented using symbols or abbreviations. According to Morreall, laughter denotes a combination of physical events, including "the spasmodic expulsion of air from the lungs, accompanying sounds, characteristic facial distortions, and… shaking"(1987: 4). Accordingly, responses to humour in the chat logs attempted to reproduce one or more of these features. However, the main difference between laughter and laughter produced in chat, is that, with the exception of performed laughter, laughing is an "involuntary or semi-voluntary response to a stimulus"(Morreall, 1987: 4). In chat, the response must be typed, and therefore, cannot be labelled involuntary. As there are no specific standard emoticons in MSN Messenger to depict facial distortions or shaking of the body due to laughter, written approximations of the accompanying sounds made when air is expelled from the lungs were one of the most common ways of expressing laughter in the chats. These included "hehe", "haha", "hehehe", "hahaha", "hahah", and in Japanese, 「ははは」 (hahaha). Another popular way to express laughter was to explicitly state the fact that the participant was laughing. In English, the acronym "lol", standing for 'Laugh Out Loud' was preferred, although it was never written with capitals, and in Japanese, variations upon the verb ‘to laugh’ were used, such as 「(笑)」, 「笑い」, and 「笑」. Sometimes a facial expression was used to accompany these written approximations of laughter, namely ^^; (embarrassed smile), perhaps as an attempt to represent the characteristic facial distortions of laughter.


Asterisks to denote action were only used by one participant, language exchange partner Kaylene, who used asterisks on four occasions, to express two actions, hugging and sighing. In her interview, Kaylene stated that writing a word between two asterisks, for example, *hugs*, "denotes action", and that "it's sort of like short for hugs you". Kaylene used this emotion expressing strategy to say goodnight to Ruriko, and another time when Ruriko related a story about her parents leaving for Japan, and that she couldn’t stop crying. In this context, Kaylene said she used *hugs* to express sympathy to Ruriko, but as a strategy, "it's not doing it very well". She explained that "it feels insufficient when you do it", and commented, "I wished I could have been there and said more to her". However, Kaylene did mention that it was an accurate reflection of what she would have done had the pair been speaking face-to-face. Discussing her use of *sigh*, Kaylene stated, "it's like an action thing. But you can't just write sigh, you need to put in action symbols". If the asterisks were not utilized, Kaylene reported that it would be equivalent to saying "sigh" rather than performing the action of sighing.

Font Symbols

Participants occasionally used font symbols to enhance their messages and express feelings. These are entered in the same way as kanji, by typing the Japanese name of the symbol, and then selecting either the symbol from the list produced by the kanji henkan list. That only Japanese participants used font symbols appears noteworthy, as perhaps Australian participants were unaware of how to produce them. A list of the font symbols used and their names is provided below.

Font Symbols Used in the Ten Chats

Type... ...to create
ほし (star)
おんぷ (note)

The star was by far the most frequently used font symbol, and appears to have positive connotations, being used in conjunction with phrases such as 「楽しみ」 (look forward to), 「かっこいい」 (cool), 「がんばれ」 (good luck) and emphatic endings such as 「よ」 and 「ね」. One of the participants, education student Tomi, even used ♪ font symbols in her display name. In the follow-up interview, she explained that she used the musical notes because it made her name appear fun. The down arrow was the least frequently used, and appears to have been used by accident, as 「した」 is a past-tense verb ending as well as the word for down, which would produce the font symbol. This kind of mistaken usage of a font symbol may be problematic for students of Japanese.

Elongation and Trailing Off

Aside from symbolic mediation to express emotion, participants utilized a wide range of strategies to represent elongation of sound and trailing off. Overall, Japanese participants by far outweighed Australian participants in their usage of both elongation and trailing off strategies. Four main categories of elongation strategies, and four main categories of trailing off strategies were identified.

To express elongation, multiple letters and use of dashes were the most common. Many participants simply repeated alphabetic letters to represent elongation, eg. "youuuuuuuuuuuu", while repetition of kana syllables was used by Japanese participants only. Each hiragana symbol may represent either a vowel (such as あ= a), a consonant plus vowel (such as か= ka) or a nasal (ん= n). To elongate kana, the hiragana vowel あ was followed by a small hiragana ぁ, eg. 「あぁ」. In the case of ね (ne), え (e) is the vowel, so the elongation would be represented ねぇ, as is shown in the following example, 「すごいねぇ!」(Whoaa!).


~ and ー were commonly used also. The dash ~, called a tilde, swung dash or wave dash, was used both in the middle and at the end of lexical items, for instance, 「じゃ~あ」(We~ll), and「いいよ~」 (Fine~). Kaylene commented after her interview that the wave dash is sometimes used on to make a phrase look "cute" in a feminine or pleasant way. The ー, called a choon, is used to lengthen vowels, and like the wave dash, was used in the chats in both mid and end of word positions, eg.「えーと」(U-m), and「かわいいー!」 (Cute-!). Only one example of multiple, repeated usage of the ch?on was found, again by Miku, in the same chat on 07/06/2006, where she used 「うわーーーーーー」 (Ugh------). This particular usage was in the context of Miku expressing her exasperation over the amount of homework she had to do.

Elongation Used in the Ten Chats

Description Represented Example
Repeated letter eg.Aaa with youuuuuuuuuuuuuuuu
Small kana eg.あぁ すごいねぇ
Tilde/Swung dash Wave dash ...じゃ~あ、私は病気の時だけ、体重を計る。
Choon/Bosen えーと、souvenir shops!!

Trailing off was also represented graphically. The first three of the methods outlined below are produced using characters readily available on a standard keyboard, while the fourth is produced by typing three full stops and then using the kanji henkan function to transform the three individual characters into a single-width character depicting three dots. Again, these strategies to represent prosody were used in various situations, mid and end sentence, and even sentence-initially. Groupings of three full stops or commas were most common, but groupings of two to six full stops were also found. Kaylene even incorporated spaces into her trailing off and on sequences, possibly to indicate pausing, eg.「...  ...」.

Trailing Off Used in the Chat Logs

Description Represented Example
Three English full stops ... I see...
Three Japanese full stops 。。。 私本当に子供っぽい。。。
Three Japanese commas 、、、 あら、、、
Three point Japanese leader icic...


In MSN Messenger, bold, italics and underline cannot be applied to individual words or sentences in the one e-turn, thus users must find other forms of emphasis. Capitalization to express emphasis in the chat logs was relatively infrequent, with exclamation marks clearly preferred. Participants sometimes used capital letters while typing in Japanese to prevent the IME from transforming the inputted letters into Japanese characters, providing a simple way to code-switch without having to switch orthography input method. For this reason, only capital usage where lower case usage was also present, either in the e-turn in question, or in that user's surrounding turns, suggesting that the IME was set to English at the time, were counted as a use of emphasis.

Of the ten chat logs collected, only four showed clear usage of capitalisation for emphasis, and according to the participants, one of these uses was accidental, and hence excluded from analysis. One example of capitalisation for emphasis occurred in the language exchange partners Kaylene and Ruriko’s chat on 24/04/2006, in which Kaylene used capitalisation to show that she was surprised at how often Ruriko used to check her weight in Japan. Kaylene responded "ehhh, you what? ...EVERYday???". The other two usages were by Miku, the social chatter, in her conversation with Jacob. Miku used capitals to specify the topic of a turn, the title of a movie, "Have you ever watched LOST IN TRANSLATION?" and to write an entire sentence, "YOU ARE A JAPANESE". This final example occurred in a section of rather animated and lively chat, and use of capitalisation in this quantity, an entire sentence, is commonly referred to as the chat equivalent of shouting (Ito, 1996).

Exclamation marks were frequently used in the chats of many pairs. Former classmates Phia and Soichiro had the highest overall frequency of turns containing exclamation marks, and in one of their chats, used exclamation marks in over 73% of their turns. Ruriko used exclamation marks in 78 of her 134 turns, and Evelyn produced a total of 25 exclamation marks over 18 turns.

In the example below exclamation mark usage appears to be so commonplace for Ruriko that when she forgot to add exclamation marks to the end of a sentence, she composed a whole new turn consisting solely of the missing exclamation marks.

Language Exchange Partners Kaylene and Ruriko, 11/05/2006

Time From Message
8:32:49 PM 瑠璃子 OK,thanks
8:32:49 PM 瑠璃子 !!

Due to the high frequency of exclamation mark usage, it appears unwarranted to assume that the exclamation mark represents a strong emphasis in all chats, particularly when exclamation use is the norm. Phia and Soichiro's usage, as well as Ruriko's personal usage of 186 exclamation marks in a single chat, compared to 30 full stops suggests that for some of the chatters, using exclamation marks to end sentences was an unmarked form of language use.

The wide use of symbols and abbreviated chat slang in the chat logs often appeared to represent a third code, related to, but distinct from, the language choices of Japanese and English. As participants themselves cited a need to express emotion, in particular, sympathy, difficult on a number of occasions, it would seem important to utilize symbolic mediation to its fullest. Some Japanese participants found their Australian partners' usage of English chat slang such as 'lol' confusing, while alternatively, many of the Australian participants would have benefited from an increased understanding of Japanese text art, as several participants cited simply ignoring the text art faces they did not understand, and a knowledge of how to use font symbols may also have enhanced their chat usage.

Summary and Conclusions

The first research question proposed was "What are the features of Japanese-English Internet Chat?". A wide variety of orthographic switching methods were found, an area that has been under explored in the body of literature on CMC. The use of symbols and chat slang was found to represent a third code choice, related to, but distinct from, the language choices of Japanese and English.

The second and final research question was "How does intercultural Internet chat provide opportunities for communication with native speakers and informal language acquisition?". Most of the participants drew on previous relationships with their chat partners that extended beyond the virtual world. Chat was, in some cases, a vehicle for cementing or continuing these interactions despite geographic separation experienced by returning to the participant's home country, as in the case of the former classmates and the former host sisters. Chat was found to provide a positive environment for interaction with native speakers and informal language acquisition, and all of the participants stated that they enjoyed chat interaction overall.

It is important to note that participants in this study were engaged in authentic interaction, to achieve their own personal goals of keeping in touch with friends, language learning, and to combat boredom, not teacher-directed, pedagogically focussed contrived interaction or classroom role-play. In the cases examined in this study, it appears that chat can provide a positive environment in which to achieve goals of socializing and language learning in an intercultural context.


Sarah Pasfield-Neofitou is a Ph D student at Monash University, Australia, after completing her honours year of a BA Languages degree with a double major in Linguistics and Japanese (Advanced). She is currently teaching in the Japanese Studies department at Monash University.