Is the Chatting between Students Spoken Form, Written Form or

Is the Chatting between Students Spoken Form,
Written Form or Something Else?
-Analysis of KWCCDLP (Korea University Waseda
University Cross-Cultural Distance Learning Project) DataNari Lee
Korea University
This paper deals with chatting data of Korea University Waseda
University Cross-Cultural Distance Learning Project (KWCCDLP). The main
concerns are the analysis of synchronous chatting data and showing its style. It
can be spoken or written form or something new. It starts to show the definitions
and categories of spoken and written styles from the previous studies. Then these
will be compared to the analysis of KWCCDLP chatting data.
The results show that the chatting data has spoken, written and chatting
forms which are well mixed. The ones prefer to use spoken forms also have
tendency to use more chatting forms than those of preferring to written forms.
1. Purpose
As the computers and internet access become common in schools and
homes, the concern of Computer-Mediated Communication (CMC) is increased.
It has also brought many changes into education, especially English learning. It
helps the students who learn English as a foreign language to have more chances
to use English in their communication.
CMC is consisted of synchronous CMC (e.g. chatting or Net meeting)
and asynchronous CMC (e.g. e-mail). The methods and effect of CMC have been
widely studied. Many researchers (Park, Nakano & Lee, 2003; Harris &
Wambeam, 1996; Hubbard, 1997; Takayoshi, 1996) have already compared
CMC based education with traditional one. Yet in synchronous CMC, the analysis
132
of chatting style has hardly been studied.
The main purpose of this paper is synchronous CMC discourse analysis
from KWCCDLP data. And it will be categorized into three sections: spoken,
written and chatting.
2. Literature Review
According to Park, et al. (2001), discourse is the result of
communication including conversation, interview, written text and so on.
Discourse analysis is the study of spoken or written communication.
Discourse is divided into spoken and written discourse (Kaplan &
Grabe, 2002; McCarthy, 1991; Stenström, 1994). The pauses, in other words,
tone unit boundaries mark the spoken discourse while punctuation marks do
the written discourse (Stenström, 1994). Chafe & Danielewicz (1987) use
lexical usage. They claim that spoken discourse has less and limited lexicons
than written one. Similarly, Corson (1997) states the lexicons used in written
discourse are more various and newer than those in spoken discourse. In
aspect of lexical density, written discourse has a higher lexical density than
that of spoken (Ure, 1971). Halliday (1992) also has similar study of lexical
density. He argues that lexical density of written discourse has twice more
than that of spoken. And written discourse has more lexical items, while
spoken discourse has more functional items.
Spoken discourse is here-and-now activity. Speakers take turns and
collaborate. To show collaboration, they use backchannels, feedback and
hesitation phenomena such as silent, filled pause, verbal fillers, false start,
repetition and incomplete utterance (Stenström, 1994). In addition,
paralinguistic features like facial expressions and gestures are also used in
spoken discourse.
Written discourse has relatively longer and complex sentences and
its grammar is more complicated than spoken discourse (Chafe &
133
Danielewicz, 1987). As Stenström (1994) mentions written discourse had
punctuation marks. Using capital letters is another distinctive feature of
written discourse.
On the other hand, some claim that it is hard to distinguish these two
(Kaplan & Grabe, 2002; Biber, 1988). Many of features in spoken and written
discourse are overlapped. For example, writing a letter to a friend is similar to
spoken discourse like face-to-face conversation despite using pen and paper
with writing skills. Having a lecture in a class, however, is more like written
discourse despite using verbal expressions. It uses more complicated words
and professional terms than every day conversation.
CMC discourse also has features of both of spoken and written
discourse. In synchronous CMC discourse, participants type messages using
keyboard like written discourse and they interact simultaneously like face-toface spoken discourse. Graddol (1989) and Johanyak (1997) claim that
synchronous CMC is hybrid forms of communication having both features of
spoken and written discourse. On the contrary, Collat & Belmore (1996)
mention CMC discourse:
“Messages delivered electrically are neither ‘spoken’ nor ‘written’ in
the conversational sense of these words. There is an easy interaction
of participants and alternation of topics typical of some varieties of
spoken English. However, they can not be strictly labeled as spoken
messages since the participants neither see nor hear each other. Nor
can they be considered strictly written since many of them are
composed directly on-line, thereby ruling out the use of planning and
editing strategies which are at the disposal of even the informal
writer.”
But Collat & Belmore (1996) only consider the medium of communication
not patterns used in the discourse. For qualitative analysis, Doell (1998),
134
Werry (1996) and Lim (2003) suggest four subsections of analyzing CMC
discourse. These are lexis, orthography, grammar and discourse. Lexis section
includes analyses of abbreviation, onomatopoeia, emoticon, slang and jargon.
In orthography section, analyses of misspelling, varieties, alternative spelling,
capitalization, punctuation, omission and contraction are included. The
omission of subject or verb, disagreement of tense, mismatch of third person
singular and verb and use of ‘ain’t’ are the components of grammar section.
Discourse section includes turn-taking and repetition of short sentences.
3. Method
3.1 Subject
Among twenty seven Korea University (KU) students who enrolled
Global English trough Internet I, the chatting data of ten students who had
regularly chatted with Waseda University (WU) students were chosen. Two
students from KU were consisted in one Korean group and two students from WU
were consisted in one Waseda group. And each group was matched for a chatting
pair. Simply if ten Korean students’ chatting data were collected there were 40
participants. But some students were in the same group and consequently 37
students were participated in. All KU students were native Korean speakers and
all WU students were native Japanese speakers. There were 18 Koreans and 19
Japanese students. There were seven males and eleven females in Korean group,
eleven males and eight females in Japanese group. The range of age was 18 to 26
years. Basically four students who were pre-matched by TA of each university
were supposed to have a chat in a group. However some chats were held with
only two students or different students from other groups every week. Among ten
chatting data, nine KU students whose chatting data were not overlapped were
chosen and numbered from K1 to K9. Following table shows the participants’ sex
and the number of partners.
135
Table 1 Participants’ sex and the number of partners
number of partners
participants
sex
KU
total
WU
male
female
male
female
K1
M
3
0
0
1
2
K2
F
1
0
0
0
1
K3
F
2
0
1
1
0
K4
M
3
1
1
0
1
K5
M
2
1
0
1
0
K6
F
1
0
0
0
1
K7
M
10
1
3
4
2
K8
F
1
0
0
1
0
K9
F
6
1
1
3
1
total (M:F)
4:5
29
4
6
11
8
(M: male, F: female)
3.2 Procedure
All students who enrolled Global English through Internet I had
participated in KWCCDLP. There were video conferencing every week during the
class from May to June between KU and WU. Conference topics were decided by
students from both universities in the beginning of video conferencing session.
After video conferencing, all students were required to have a chat at least once in
a week with their partners. Chatting topics were as same as those of video
conferencing. Students were required to have a chat for five weeks at multimedia
education room where special programs and equipments for chatting were set up.
There was more than one TA who helped students participated in chat session. In
136
the beginning of a semester, there was a trial session for students how to use
chatting program, BizMate. With web camera, participants see their partner on the
screen. The message are typed with keyboard and word files, presentation files,
pictures, graphs and web site can be displayed on the screen while interact.
Students were asked to save all chatting data after their chatting. At the end of the
semester, students had to submit their chatting data which were downloaded
diskette to TA.
4. Result and Discussion
4.1 General analysis of chatting data
For quantitative approach, the total number of turns and words used in
chatting data were calculated. Then the total number of words was divided by the
total number of turns. It will be called ‘Ave’ in this paper to explain which had
denser and longer sentences. According to Ure (1971) and Halliday (1992),
written discourse has higher lexical density than spoken discourse. Like
studies of Ure (1971) and Halliday (1992), if one’s chatting data has higher
Ave than that of the other, it may be relatively close to written style.
Calculating Ave is for comparing quantitative analysis to qualitative analysis.
Table 2 and figure 1 show the total number of turns, words and its Ave. The
average Ave of all participants was 5.40.
Table 4 Total number of turns, words and Ave of the participants
participants
Turn
Words
Ave
K1
651
3431
5.27
K2
835
5719
6.84
K3
890
5344
6.00
K4
1504
8703
5.78
137
K5
177
1207
6.81
K6
987
4924
4.98
K7
1882
8289
4.40
K8
587
3996
6.86
K9
1443
6731
4.66
total
8956
48343
5.40
average
218
1179
5.40
Figure 1 Participants’ Ave
8
7
6
5
4
3
K1
K2
K3
K4
K5
K6 K7
K8
K9
(average Ave of all participants: 5.40)
Having qualitative analysis, it can be assume that the chatting data
which has higher Ave than the average Ave is close to written discourse. On the
contrary, the chatting data below the average Ave will show more spoken forms in
their data. From table 2 and figure 1, there is an assumption that the data of K2,
K5 and K8 have more written forms than the others. K1, K6, K7 and K9 use more
spoken forms in their chatting data. For qualitative approach, the spoken, written
and chatting forms in chatting data were examined.
4.2 Criteria of chatting data
Doell (1998), Werry (1996) and Lim (2003) mention the subsection
138
of synchronous CMC discourse. And the features of the subsections are
categorized to spoken forms, written forms and chatting forms.
Firstly, spoken forms are consisted of slang, jargon, onomatopoeia,
capitalization, overuse of punctuation marks (exclamation, question mark),
omission of subject (I, it), omission of verb (be, have), mismatch of tense,
number, third person singular verb, turn-taking and repetition of short
sentences.
Secondly, written forms are consisted of capital letters (proper nouns,
titles, first person singular I, interjection O, the first word of every sentence)
and punctuation mark (comma, period).
Lastly, in synchronous CMC discourse, participants need to type their
messages as speedy as possible to cope with their interact speed like face-to-face
interaction. The chatting forms are consisted of abbreviation, computer jargon,
emoticons, symbol and number, varieties and alternative or omission of
spelling. There are two types of abbreviation. One is using first letters of each
word like ‘BTW’ which stands for ‘by the way’. The other is only using
consonants of a word like ‘PLS’ for ‘please’. Participants use computer jargon
like ‘spam mail’ or ‘web cam’. The most distinctive feature of synchronous
CMC is emoticon. To express participants’ emotion and facial expression,
they use letters and symbol keys and it is called emoticon. It gives
paralinguistic help during the chat session. Participants also use symbol and
number to reduce time of typing. For example, ‘at’ is replaced to ‘@’, some
preposition such as ‘to’ or ‘for’ are changed to ‘2’ or ‘4’. Some use number
and letters to type a word. For instance, ‘tonight’ becomes ‘2nite’ and ‘skater’
becomes ‘sk8er’. Alternative spelling is changing some spellings in a word as
being pronounced or in a simple way. People use ‘boi’ as ‘boy’ and ‘dunno’
instead of ‘don’t know’. Varieties are very similar to alternative spellings but
they are changed intentionally. Most of varieties are simple forms of longer
and complicated words. For example, ‘because’ is changed to ‘cos’ or ‘cuz’,
‘you are’ to ‘u r’ and ‘people’ to ‘ppl’. Omission of spelling happens when the
139
omitted letter cannot affect the pronunciation or meaning. It is often shown at
the end of the words like ‘nothin’’ or ‘goin’’. Or it happens in the middle of
the words like ‘wat’ for ‘what’ and ‘havnt’ for ‘haven’t’. But in this paper,
varieties include alternative or omission of spellings. Because they are
displayed in very similar ways and it is hard to distinguish participants’
intentions in every changed forms.
Based on these categories, the criteria for spoken, written and
chatting forms are made. Since written form has only two features, the
analyzing scope of features was settled for two characteristic features in each
criterion for logical counting and comparing. Two features are selected which
are frequently shown in the chatting data. Table 3 suggests two distinctive
features of each criterion.
Table 3 Distinctive features of criteria
Spoken Form
-onomatopoeia
(sounds of laugh,
Written Form
-capital letters
Chatting Form
-emoticons
(proper nouns, titles, -varieties
interjection,
first person singular (including alternative or
filled pause,
I, interjection O, the
repetition of vowels in
first word of every
the middle of a word
sentence)
omission of spellings)
or consonants at the -punctuation mark
end of a word, ~mark)
(comma, period)
-overuse of punctuation
marks (…/???/!!!/??!!)
4.3 Spoken form
Two features of spoken form are use of onomatopoeia and overuse of
140
punctuation marks. They are frequently used to express sounds and intonation.
The onomatopoeia includes sound of laugh, interjection, filled pause such as
‘um…’, ‘hmm…’, use of ~ mark and the repetition of vowel in the middle of a
word and consonant at the end of a word like ‘hiiiiii’ for ‘hi’ and ‘verrrrry’ for
‘very’. Farfelder (2000) insists the repetition of specific letters in a word be
considered as spoken discourse for showing emotions. Using ~mark during the
chat is very unique form in KWCCDLP data. It is used for illustrating the
intonation of words and sentences. The more use of ~mark, the stronger and the
longer intonation is made. Another feature is overuse of punctuation marks. It is
also used to demonstrate the emotions of participants. The more punctuation
marks are used, the stronger emotions are expressed. One unique pattern in
KWCCDLP data is over use of period. It makes filled pause, implies interacts are
kept going and illustrates participants’ opinion indirectly. The followings are the
examples of spoken forms.
Example 1 Overuse of onomatopoeia and punctuation mark
korea006 : cool~~~~~ but appearance is not coolooooool~~ ha?
korea001 : inlineskate+_+wow cool~
korea001 : nono~
korea006 : my nick name is ... you can say just "yong"
korea006 : yong~~
waseda edu006 : Are you freshman?
korea006 : wow!!!!
korea006 : thanks alot!!!!!!!!!!!!!!!
korea001 : kk
korea006 : that's really reaallly~~ nice complementary to me~
korea006 : yes...that's what we know...
waseda edu013 : Yes......Next week is our last chat.......
141
korea006 : i guess time is up for today...
Two Korean students in the first example use various spoken forms. They repeat
specific letters in words like ‘coolooooool’ or ‘reaallly’ to show their strong emotion.
Next, they use ~mark to demonstrate intonation, it is mostly used at the end of the
sentences or after the words they want to emphasize. There are overuse of punctuation
marks on line 7, 8 and the second example. Repeated exclamation marks illustrate
amazement and pleasure. Repeated periods show sadness for closing chat session.
Figure 2 demonstrates the frequency of spoken forms used by each
participant. It is calculated that the total number of spoken forms including use of
onomatopoeia and overuse of punctuation marks is divided by the total number of
words.
Figure 2 Frequency of spoken forms used by each participant
14%
12%
10%
onomatopoeia
punctuation mark
spoken forms
8%
6%
4%
2%
0%
K1
K2
K3
K4
K5
K6
K7
K8
K9
According to figure 2, K7 has the most frequent use of spoken form and K1 has
the second most. It is corresponded to the Ave analysis. As the Ave is below the
average, it has more spoken forms. In the previous chapter, K1, K6, K7 and K9
are below the average; K1 and K9 have higher use of spoken forms. In contrast,
K2, K5 and K8 have higher Ave and their frequencies of spoken form use are
relatively very low. Even though K4 and K6 are below the average of Ave, they
don’t have many spoken forms. It is drawn that there is a tendency that the data
with lower Ave are likely to have more spoken forms but not always.
142
4.4 Written form
Starting each sentence with capital letter and ending with punctuation
mark are features of written forms. Figure 3 shows the number of used written
forms in chatting data. Being guessed, K5 and K8 which have higher Ave
demonstrate more use of written forms. On the contrary, K7 has the least use of
written forms. K1, K6, K7 and K9, the ones which have lower Ave usually use
less written forms. Although K2 has higher Ave, its use of written forms was low.
K2 use less both of spoken and written forms. Except K2, there is a tendency that
chatting data with higher Ave has more written forms and lower one has less.
Figure 3 Ratio of using written forms to the total number of turns
100%
80%
60%
capital letter
punctuation mark
40%
20%
0%
K1
K2
K3
K4
K5
K6
K7
K8
K9
4.5 Chatting form
The main characteristic features of chatting forms are emoticons and
varieties. Emoticons help participants have paralinguistic cues such as facial
expressions or gestures. In KWCCDLP, the participants can see their partners on
the screen by web camera. The frequency of using emoticons might be less than
the chatting without seeing partners on the screen. Following example
demonstrates the use of emoticons.
143
Example 2 the use of emoticons
korea008 : ah, did you send?? thanks thans!! ^---------------------^*
waseda law006 : yourwelcome
waseda law006 : ^^
waseda edu013 : (@[email protected])
korea012 : have you ever eaten korean food?
The participants use symbol keys to show their feelings. They use various
types of emoticons according to the situation and their feelings.
The next feature is varieties including alternative or omission of
spellings. To save time of typing and interact simultaneously, participants use
varieties. One way of varieties is typing words as being pronounced like ‘c’ for
‘see’ and ‘foriners’ for ‘foriegners’. Another way is changing spelling of a word
like ‘wuz’ for ‘was’ and ‘hab’ for ‘have’. In this case, the pronunciations of words
are changed. The last one is omission of punctuation mark apostrophe ’ and
changing some spellings like ‘thaz’ for ‘that’s” and ‘dunno’ for ‘don’t know’.
Example 3 the use of varieties
korea004 : i hab to go to class and u?
korea004 : c ya next time~
korea004 : hab a nice day~~
korea004 : Thatz okay~
korea004 : because Korean ppl think Nationalism is very important
korea011 : really it wuz.!
waseda edu012 : always foriners think
korea012 : +_+ dunno @[email protected]
korea012 : yes, you r right
144
In figure 4, there is frequency of using chatting forms. K6 and K7 have higher use
of chatting forms than the others. Both of K6 and K7 use many of emoticons and
varieties. Generally the use of emoticons is more than that of varieties. The
affective filter can be one reason why the participants prefer to use emoticon to
varieties. All participants are non native English speakers and they might have
tendency to avoid misspelling which shows lower English level. And this happens
to the ones who prefer to use written forms in their chatting. For example, K4, K5
and K8 never use varieties and K5 doesn’t have any chatting forms. The ones
whose Ave is lower than the average of Ave are K1, K6, K7 and K9. The three
fourth of these participants have higher frequency of using chatting forms. Among
the ones whose Ave is higher, K5 and K8 hardly have chatting forms. But K2 has
relatively higher frequency of using chatting forms than those of spoken or
written forms in her chatting data.
Figure 4 Frequency of using chatting forms
7%
6%
5%
emoticon
varieties
chatting form
4%
3%
2%
1%
0%
K1
K2
K3
K4
K5
K6
K7
K8
K9
5. Conclusion
According to data analysis, several tendencies of chatting styles are
found. Firstly, the chatting data with higher Ave have more written forms than
spoken or chatting forms. The other ones with lower Ave have more spoken and
145
chatting forms. Not always these relations are proven, there is a higher Ave
chatting data which has neither spoken nor written forms predominantly but
chatting forms.
Secondly, the chatting data which has more spoken forms also have
more chatting forms. But the ratio between use of spoken and chatting forms are
different. The frequency of using chatting data is relatively lower than that of
spoken. It can be explained by KWCCDLP chatting situation and affective filter.
The participants of KWCCDLP can see each other on the screen and
paralinguistic cues like facial expressions or gestures were delivered on-line
situation. With this help, the need of using emoticons is relatively reduced. The
other reason is affective filter to use varieties. As mentioned, all participants are
non native English speakers and they might hesitate to use varieties. Since
varieties change or omit the spellings which are considered as lower English
proficiency.
Lastly, the spoken, written and chatting forms are displayed in all
chatting data except K5. It can not be concluded that chatting data is spoken style,
written style or chatting style by one dimensionally. All features are well mixed
and there are only tendencies that one chatting data is close to spoken style,
another is close to written style and the other is close to chatting style.
This study is only dealt with the chatting data of 37 non native English
speakers for six weeks. For further study, more chatting data are examined and
compared to those of native English speakers. The relationship between chatting
style and pedagogical implication is needed for the next step.
Reference
Biber, D. (1988). Variation across speech and writing. Cambridge: Cambridge
University Press.
Chafe, W. & Danielewicz, J. (1987). Properties of spoken and written
146
language. In Horowitz, R. & Samuels, S.J. (Eds), Comprehending
oral and written language. New York: Academic Press.
Collot, M., & Belmore, N. (1996). Electronic Language: A new variety of
English. In Herring. S. (Ed), Computer-Mediated Communication:
Linguistics. Social and Cross-Cultural Perspectives (pp.13-28).
Amsterdam: John Benjamins.
Corson, D. (1997). The learning and use of academic English words.
Language Learning, 47. 671-718.
Doell, W. (1998, May 11). The internet relay chat (IRC): Linguistic
perspectives. Wernfrid Doell's personal homepage: Germany.
Retrieved
June
2nd, 2004
from the
World
Wide
Web:
http://home.allgaeu.org/ndoell/work/ircpaper.htm.
Farfeleder, S. (2000). The analysis of chat communication. The website of
English and American Studies at University of Salzburg: Austria.
Retrieved
June
2nd, 2004
from the
World
Wide
Web:
www.sbg.ac.at/ang/projects/ps_s00/webfiles/farfeleder.htm.
Graddol, D. (1989). Some CMC discourse properties and their educational
significance. In Robin, M. & Anthony, K. (Eds), Mindweave:
Communication, computers, and distance education (pp.236-241).
Oxford, Pergamon Press.
Halliday, M.A.K. (1992). Spoken and writing language (3rd ed.). Oxford:
Oxford University Press.
Harris, L.D. & Wambeam, C.A. (1996). The internet-based composition
classroom: A study in pedagogy. Computers and Composition, 13.
353-371.
Hubbard, D. & Walberg, H.J. (1997). Student view of computer-composition
effects on writing. Computers and Composition, 15. 59-71.
Johanyak, M.F. (1997). Analyzing the amalgamated electric text: Bringing
cognitive, social, and contextual factors of individual language
users into CMC research. Computers and Composition, 14, 91-110.
147
Ablex Publishing Corporation.
Kaplan, R.B. & Grabe, W. (2002). A modern history of written discourse
analysis. Journal of second language writing, 11, 191-223, Pergamon.
Lim, J.K. (2003). Study of English patterns on Internet, Master’s thesis,
Graduate school of education, Sungshin Women’s University.
McCarthy, M. (1991). Discourse analysis for teachers. Cambridge:
Cambridge University Press.
Park et al. (2001). Dictionary of Applied Linguistics. Seoul: Kyungjin
Publisher.
Park, K.J., Nakano, M. & Lee, H.K. (2003). Cross-Cultural Distance
Learning and Language Acquisition. Seoul: Hankook Publishing Co.
Stenström, A-B. (1994). An Introduction to spoken interaction. London and
New York: Longman.
Takayoshi, P. (1996). The shape of electronic writing: Evaluating and
assessing
computer-assisted
writing
processes
and
products.
Computers and Composition, 13. 245-257.
Ure, J. (1971). Lexical density and register differentiation. Applications of
linguistics: Selected papers of the second international congress of
applied linguistics, Cambridge, 1969, Cambridge: Cambridge
University Press.
Werry, C.C. (1996). Linguistic and interactional features of internet relay chat.
In
Herring.
S.
(Ed),
Computer-Mediated
Communication:
Linguistics. Social and Cross-Cultural Perspectives (pp.13-28).
Amsterdam: John Benjamins.
148