Universitatis Iagellonicae Cracoviensis 123 (2006)
Instytut Filologii Angielskiej UJ
TRANSFER IN L2 WRITING: LINGUISTIC
COMPETENCE VS. COMPOSING SKILLS
Despite the recent developments in the study of language transfer and numerous
studies confirming the complexity of this phenomenon, language teachers
understandably associate transfer with its inhibitive effect on the development of
learners’ L2 competence. This particularly concerns classroom settings where
the teacher as well as all students come from the same background, sharing not
only their L1 but also the experience of eliminating its interference in L2. Some
teachers are known to be extremely sensitive to calques, which negatively
affects their assessment of student papers, even if those calques are quite
acceptable in L2 and would neither be identified as errors or recognized as
calques by raters not knowing students’ L1. Such a response to L2 production
may be justified in the case of advanced students, where the vague notion of
“what sounds natural” is more of an issue than mere accuracy.
If accuracy is teachers’ primary concern, then obviously transfer is perceived
as L1 interference resulting in errors. To illustrate this point, a selection of
interlingual errors in Polish learners’ written English is presented in Table 1.
These errors were produced by the poorest candidates applying for entry to the
English Department of the Jagiellonian University in 2001. What must have
triggered language transfer in these texts was the writing task itself, which
involved producing an English (L2) summary of a Polish (L1) text. So even if
some errors may be interpreted as intraligual resulting from e.g.
overgeneralization, simplification (Richards 1971), or transfer of training
(Selinker 1972), it is quite likely that they are primarily transfer errors.
While researchers are concerned with error taxonomies, the teacher would
frequently be more concerned with error evaluation. The criterion that is perhaps
most commonly used is what Hughes and Lascaratou (1982) call ‘basicness’,
relying on the judgment that particular rules are ‘more fundamental’ than others,
which in the educational context is naturally related to the syllabus. In other
words, more serious errors are in the areas that have already been taught. So, for
example, the learner’s inability to form questions in the Simple Present Tense
(Error 8 in Table 1) would be viewed as particularly serious, because this
grammatical point is typically introduced very early in the syllabus. However, in
the case of such interlingual errors the criterion of basicness may be extended to
the use of strategies and thus contribute to the overall effect of error gravity. The
strategy employed by the poor candidates, i.e. literal word-for-word translation,
is to be eliminated at very early stages of second language acquisition.
170 EWA WITALISZ
Table 1. Sample errors in written English produced by Polish EFL learners
Error Reconstruction in L1 Reconstruction in L2
1. Efect is one. Efekt jest jeden. There is one effect.
2. They not can go. Oni nie mogą pójść. They cannot go.
3. They not nice smell. Oni nieprzyjemnie pachną. They don’t smell nice.
4. ... live these children is ... życie tych dzieci jest ... these children’s life is
very hard. bardzo ciężkie. very hard.
5. Situation children is bad. Sytuacja dzieci jest zła. The situation of children is
6. In our world always will Na świecie zawsze będą In our world there will
poor children. biedne dzieci. always be poor children.
7. People imagine from Ludzie (?) zastanawiają People wonder how it
where did takes babies in się, skąd się biorą happens that there are
trash. noworodki na śmietniku. babies found in dustbins.
8. Cam from the wors Skąd są gorsze dzieci? Where do worse children
children? come from?
9. What exactly he meant? Co dokładnie miał na What exactly did he mean?
10. Children steal clothes in Dzieci kradną ubrania, aby Children steal clothes in
order to have what wear. mieć co włożyć. order to have something to
11. smiled children uśmiechnięte dzieci smiling children
12. No one not trying Nikt nie próbuje No one is trying.
13. They will grow on people Wyrosną na ludzi, którzy They will grow up to be
who ... ... the people who ...
14. It’s very possible that at Jest bardzo możliwe, że ze It’s very possible that bad
the bad children will grow złych dzieci wyrosną źli children will grow up to be
up bad people. ludzie. bad people.
15. They shame for the worse Wstydzą się gorszych They are ashamed of the
clothes. ubrań. worse clothes.
They shame worst cloths.
16. One people are getting Jedni ludzie się bogacą, Some people are getting
rich, another loosing saves inni tracą oszczędności rich, others are losing
of whole their lives. całego życia. savings of their whole
17. observators obserwatorzy observers
18. licvidation of schools likwidacja szkół closing down schools
In the case of this university entrance examination, there are other factors
that add to the general impression of error gravity. In socio-psychological terms
the candidates taking a highly competitive exam in the area of their choice
should display high motivation, positive attitude and high self-esteem. In
linguistic terms, there are two aspects: first of all, the candidates are expected to
have advanced language skills (FCE – CAE level), so obviously such errors
Transfer in L2 Writing: Lingustic Competence vs. Composing Skills 171
should not appear at all. But what is even more striking is that a number of the
poorest texts display vast discrepancies between different features, e.g. evidence
of advanced lexis and ‘basicness’ of grammatical errors, so in those cases it is
the text features themselves that point to the gravity of the error.
Studying these texts, one cannot help feeling that what the poorest candidates
could have benefited from is the commonly advocated strategy of “thinking in
L2” rather than their attempts to translate the original advanced text in L1. It
was quite possible to attempt that particular writing task with very low language
skills (e.g. lower intermediate), so if those candidates whose language skills
were very poor had tried to formulate their ideas in English, their texts would
have been simple but not affected by language transfer to such a high degree and
resulting in such serious errors.
However, the strategy of “thinking in L2”, i.e. generating the content and
organizing it into a text, may lead to unexpected problems which I would like to
illustrate with a film review (Text 1) produced by a serious and intelligent
student, Kasia, whose level of English could be described as post-FCE. (She was
a Kraków Polytechnic student attending a CAE course; her scores in CAE tests
were around 50%.)
“Przedwiośnie” – Kijów Cinema
If you feel weak and old you should see this film.
The movie is directed by well-known Filip Bajon. He added some romantic
plots to make watching more exciting. Nevertheless, it’s not a romance. Never.
This is a picture of young and mutinous man involved in martial movements.
Cezary Baryka, played by Mateusz Damięcki, on one hand seems to be
childish. On the other hand, he’s determined and matured. I’m impressed by
his acting. His mother, successfully played by Krystyna Janda, and his father,
brilliantly played by Janusz Gajos, are marvellous. But as the film progresses I
can see Maciej Stuhr as Cezary’s best friend. His starring is absolutely
stunning. I love the scene with horses. He’s so impulsive. I’m also thrilled by
the soundtrack. Especially, when Baku is the setting for some action with its
The big problem of the movie is lack of clear and main plot. While Cezary
is achieving his peak level of engagement in politics he’s rebellious and eager
to change the world. Then, suddenly he’s calm and rational. It’s a bit irritating.
I reckon Cezary should be more creative.
To sum up, “Przedwiośnie” is a highly-recommended film. The
professional photography and most of the actors make it esthetic and magnetic
for your eyes.
Text 1. Film review in L2 (214 words).
As can be easily seen, the text is highly incompatible with the writer’s profile
both in terms of the content and the language. It strikes the reader with its
172 EWA WITALISZ
immaturity (a group of raters estimated the writer’s age to be between 15–16),
its chaotic structure, lack of cohesion and poor paragraphing as well as poor
language skills. What is interesting about the language, though, is that it is not
really marred by errors and, apart from perhaps one lexical error (majestatic), it
is practically not affected by language transfer. However, there is a noticeable
discrepancy between two text features at a micro-level of discourse, i.e. syntax
and lexis. According to Cumming’s (2001: 1–23) comprehensive review of
empirical research conducted in the last two decades, various studies show that
L2 learners improve the complexity and accuracy of the syntax and morphology
and use a greater range of vocabulary in their writing as their L2 proficiency
develops. Kasia shows very different levels of the development of these
features. The simplicity of her syntax contrasts with the advanced lexis, which
can be illustrated using the frequency bands from Collins COBUILD English
Dictionary (1995). In her text of 214 words she used 12 words belonging to the
lowest frequency bands and one from the least frequent 5% of all written and
spoken English (Table 2).
Table 2. Frequency of advanced vocabulary from text 1.
Collins Cobuild English Dictionary
advanced vocabulary from text 1
most frequent band 1 ca. 700 words
band 2 ca. 1200 words
band 3 ca. 1500 words
band 4 ca. 3200 words romance, martial, marvellous,
stunning, soundtrack, engagement,
rational, esthetic, magnetic (9)
least frequent band 5 ca. 8100 words thrilled, impulsive, rebellious (3)
total ca. 14700 words
95% of all spoken
and written English
remaining 5% mutinous (1)
At a macro-level of text structure, Cumming (2001: 3) concludes, learners
“become more adept at signaling a hierarchy of related ideas at the beginning,
end, or throughout a text, specifically by using cohesive, functional-semantic, or
various stylistic devices”. What might be surprising about this text is that it
makes an overall impression of lack of cohesion despite as many as 9 cohesive
devices dutifully used by the writer (nevertheless, on one hand, on the other
hand, but, also, especially, while, then, to sum up).
Puzzled by the incompatibility between the text produced by the student and
her language skills measured in other tests as well as her background, I asked
her to attempt the same writing task in her L1 (Polish) in order to determine
Transfer in L2 Writing: Lingustic Competence vs. Composing Skills 173
whether the problems in her text resulted from the lack of competence in writing
or from the lack of linguistic competence (Text 2).
„Przedwiośnie” – Kino Kijów
Jest to film dla poszukujących wiosny, witalności i entuzjazmu
w codziennym życiu, dla zmęczonych i smutnych.
Reżyserem jest znakomity artysta Filip Bajon, znany z raczej luźnych
adaptacji powieści, na podstawie których powstają jego filmowe dzieła. Stąd
odnajdziemy tutaj dodatkowe sceny, zwłaszcza te romantyczne, które
emocjonalnie koloryzują całą fabułę nie zaciemniając jednak jej prawdziwego
charakteru. W główną postać Cezarego Baryki wcielił się młody, a jakże
przekonywujący w swej roli Mateusz Damięcki. I z jednej strony wydaje się
być nazbyt dzieciny, aby pretendować do miana bohatera o buntowniczym
nastawieniu do życia. A jednak jego zdecydowany głos, znakomite wyczucie
sytuacji sprawiają, że potrafi, jak prawdziwy mężczyzna, stanąć na wysokości
zadania. Oprócz niego na ekranie zobaczymy dużo nowych, młodych twarzy,
które dominują w obsadzie aktorskiej. Jak choćby Maciej Stuhr, który gra
Hipolita, przyjaciela Baryki. Jest on dla mnie bardzo autentyczny w swej roli,
a najbardziej ujął mnie sceną z pędzącymi końmi, nad którymi on zdaje się
mieć całkowitą kontrolę. Na ten film warto też iść dla dwóch ról: Krystyny
Jandy i Janusza Gajosa, które choć epizodyczne, to nadają romantyczny
charakter Cezaremu, czyli ich filmowemu synowi. Zdjęcia, zwłaszcza te
wykonane w Baku też zasługują na wyróżnienie. Obraz jest estetyczny
i klasyczny, tzn. nie szokuje udziwnieniami, nie razi nowoczesną techniką.
A czymże jest obraz bez dźwięku? Niedoskonałością. Muzyka sączy się
przyjemnie dla ucha wybuchając tylko raz po raz różnorodnością
i intensywnością barw.
Jedynym mankamentem filmowej adaptacji wydaje mi się być dwoistość
głównego bohatera. Mam tu na myśli te momenty, w których Cezary staje się
z nieustraszonego rewolucjonisty zwykłym cywilem. Nie odczuwam żadnego
wpływu jego pasji do historii na życie prywatne.
Na zakończenie powiem tylko tyle, że jest to film zasługujący na najwyższą
uwagę z naszej strony. Warto więc poświęcić te 3,5 godziny, aby dać się
porwać wartkiej i ekscytującej akcji w tak cudownych plenerach.
Text 2. Film review in L1 (302 words).
This text shows a mature writer, whom the same group of bilingual raters
estimated to be at least 5 years older (definitely a high school graduate). It is
well organized; the elements that did not seem to be connected at all (His
starring is absolutely stunning. I love the scene with horses. He’s so impulsive.
I’m also thrilled by the soundtrack. Especially, when Baku is the setting for
some action with its majestatic oil-wells.) are now well linked. The ambiguity of
the opening sentence is removed. The topic sentence in paragraph 4 (The big
174 EWA WITALISZ
problem ...) is changed (paragraph 3 in text 2) so that it suits the development of
The Polish text is obviously not flawless. There might be some unfortunate
lexical choices and some inconsistent imagery. It is also worth noticing that in
both texts the writer uses emphatic rhetorical devices inadequate for the text
type (a film review):
Text 1: Nevertheless, it’s not a romance. Never.
Text 2: A czymże jest obraz bez dźwięku? Niedoskonałością.
However, the Polish text simply reads well, and this effect is primarily
achieved by collocations or even clichés, which on closer examination may be
identified as longer chunks of language imported from professional samples of
this text type.
Naturally, the writer’s L1 proficiency is responsible for the linguistic quality
of the Polish text. Still, her L2 linguistic competence would have allowed her to
produce a cohesive text even with the poor range of devices she displayed in her
text (cohesive devices, relative clauses). What seems to have happened while
she was engaged in composing her English text is a kind of simplification or
reduction at the stage of generating her ideas in L2. Contrary to the university
candidates discussed earlier, who paid a high price for formulating their ideas in
Polish, which resulted in serious transfer errors, this writer would have
tremendously benefited from generating and organising her text in L1. The
comparison of her two texts shows that her composing skills did not transfer
from L1 to L2.
This case is by no means unique. Frequently L2 texts of advanced and
mature learners give evidence of unexpected lack of composing competence,
which brings up the question of how L1 writing compares with writing in L2.
Earlier it was argued that composing processes are similar across L1 and L2, but
more recent studies have found a number of differences between L1 and L2
writers. In her overview of second language writing process research, Krapels
(1990: 37–56) provides an extensive commentary on the major studies and
compares the research findings pointing to the contradictions among them.
A number of studies confirm the similarities:
1. A lack of competence in writing in English results more from the lack of
composing competence than from the lack of linguistic competence (Jones
1982; Zamel 1982; Raimes 1985a).
2. The composing processes of “unskilled” L2 writers are similar to those of
“unskilled” L1 writers; similarly, the composing processes of “skilled” L2
writers are similar to those of “skilled” L1 writers, which indicates that the
differences between L1 and L2 writers relate to composing proficiency
rather than to their first languages (Zamel 1983).
3. One’s first language writing process transfers to, or is reflected in, one’s
second language writing process (Edelsky 1982; Gaskill 1986; Jones and
However, there are also studies showing the differences between L1 and L2
writers. Raimes (1985b, 1987, 1991) found that L2 students spent much more
Transfer in L2 Writing: Lingustic Competence vs. Composing Skills 175
time rehearsing what they wanted to write, were not as bound to local contexts
or a concern for making errors, and were not inhibited by teachers’ efforts to
correct their work (Grabe and Kaplan 1996: 240). Campbell (1990: 211–230),
who examined both the products and the composing processes of L1 and L2
students to compare the way in which they used a background reading passage
in their academic writing, found that the L2 writers planned less and depended
more on the reading than the L1 writers.
A very interesting study was conducted by Silva (1993) who, in an attempt to
understand the distinct nature of L2 writing, examined 72 reports of empirical
research comparing L1 and L2 writing. The subjects, predominantly
undergraduate college students with fairly advanced levels of English,
represented at least 27 different L1s. His findings suggest that in general terms
adult L2 writing is distinct from and simpler and less effective than L1 writing.
Although general composing process patterns are similar in L1 and L2, it is clear
that L2 composing is more constrained, more difficult and less effective. L2
writers did less planning and had more difficulty with setting goals and
generating and organizing material. Their transcribing (producing written text)
was more laborious, less fluent, and less productive. They reviewed, reread, and
reflected on their written texts less; they revised more but with more difficulty
and were less able to revise intuitively. On the whole, L2 writers’ texts were less
fluent (fewer words), less accurate (more errors), and less effective (lower
holistic scores). In linguistic terms their texts were stylistically distinct and
simpler in structure; they contained more but shorter T units; they displayed
distinct patterns in the use of cohesive devices and less lexical control, variety,
and sophistication overall.
If L2 writing is seen as distinct from L1 writing and composing skills and
processes cannot be expected to transfer from L1 to L2 in a natural, intuitive
way, then it is important to consider a possible role of L1 use in L2 writing.
A number of studies established a positive influence of L1 use. Friedlander
(1990: 109–125) concluded that planning on certain topics seems to be enhanced
when writers use the language of the topic-area knowledge and that translation
from L1 into English appears to help rather than hinder writers when the topic-
area knowledge is in their L1. According to Cumming (1990), L1 allows the
writer to access appropriate lexical items and phrases and to consolidate
ideational relations across languages. L1 also permits more sophisticated
thinking on the writing topic (Cumming 1989; Leki 1992). Krapels (1990) notes
that studies demonstrate various use of L1 in L2 writing (cf. Martin-Betancourt
1986; Cumming 1987; Friedlander 1990), but some studies actually offer
contradictory findings on this issue (Krapels 1990: 40).
Given the contradictions between the findings as well as individual
differences between writers (Raimes 1985b; Arndt 1987), it is obviously not
easy to draw any definite conclusions. But coming back to the classroom setting
and the teacher’s perspective, there are certain pedagogical implications
concerning expectations of transfer in L2 writing. It appears quite plausible that
although the results of negative language transfer are unquestionable and clearly
visible in the language produced by learners, the teacher’s extreme fear of any
cross-linguistic influence may in fact inhibit the learner’s progress. While some
176 EWA WITALISZ
transfer errors can be just surface errors easily corrected at the stage of editing,
discouraging the learner from L1 use at the initial stages of text production
(generating, planning, organising) would frequently result in global problems in
the text, which are more difficult to solve and, in the case of assessment, lower
the holistic score much more than occasional language errors. Dealing with
educated, advanced, adult learners, the writing teacher should take advantage of
the learner’s writing competence in their L1 and enhance the transfer of their
composing skills, even if it involves the use of L1, because this does not
necessarily have to result in transfer errors. As Arabski’s (1979: 29) analysis of
a corpus of 4263 errors showed, both compositions in L2 and translations of L1
texts into L2 elicited the same types of errors.
Advanced writing is frequently misunderstood as being synonymous with
very complex syntax and rare lexis with hardly any emphasis on task fulfilment,
which requires generating relevant content, organizing it appropriately and
putting it in the language adequate for the text type and the communicative
event. This misconception about L2 writing can be seen in the written English
produced by advanced L2 writers, e.g. college graduates majoring in English.
While their language displays certain features of advancement, it may be used to
express irrelevant content in a very poorly organized text. Text cohesion is
misunderstood as simply using numerous cohesive devices, following some
prescribed patterns apparently as a result of training, with hardly any attention
paid to the logical transition between the ideas.
An extreme example of this approach is an MA student who once confessed
to me that when she was to produce some English, she learned to think first of
some difficult structure and vocabulary and then about the idea she might
possibly express with it. So in our informal chats she would often make an effort
to produce some really impressive complex structure with very advanced lexis,
completely inadequate for the current discourse, and then, somewhat
disappointed with the result, she would immediately admit, “Oh, this is not what
I really wanted to say!”, as if advanced language skills were to inhibit
communication. Naturally, this is the most undesirable effect of any language
training and, hopefully, it can be avoided if learners are trained to benefit from
their L1 competence rather than fear its possible negative effects.
Arabski J. (1979): Errors as Indications of the Development of Interlanguage, Katowice.
Arndt V. (1987): Six Writers in Search of Texts: A Protocol Based Study of L1 and L2
Writing, „ELT Journal”, t. 41, s. 257–267.
Campbell C. (1990): Writing with Others’ Words: Using Background Reading Text in
Academic Compositions [w:] B. Kroll red. Second Language Writing, Cambridge, s. 211–
Collins Cobuild English Dictionary (1995), London.
Cumming A. (1987): Decision Making and Text Tepresentation in ESL Writing Performance
[Paper presented at the 21st Annual TESOL Convention, Miami, April].
Cumming A. (1989): Writing Expertise and Second Language Proficiency, „Language
Learning”, t. 39, s. 81–141.
Transfer in L2 Writing: Lingustic Competence vs. Composing Skills 177
Cumming A. (1990): Metalinguistic and Ideational Thinking in Second Language
Composing, „Written Communication”, t. 7, s. 482–511.
Cumming A. (2001): Learning to Write in a Second Language: Two Decades of Research,
„International Journal of English Studies”, t. 1, s. 1–23.
Edelsky C. (1982): Writing in a Bilingual Program: The Relation of L1 and L2 Texts,
„TESOL Quarterly”, t. 16, s. 211–228.
Friedlander A. (1990): Composing in English: Effects of a First Language on Writing in
English as a Second Language [w:] B. Kroll red. Second Language Writing, Cambridge,
Gaskill W. (1986): Revising in Spanish and English as a Second Language: A Process
Oriented Study of Composition [Nieopublikowana praca doktorska, University of
California, Los Angeles].
Grabe W., Kaplan R. (1996): Theory and Practice of Writing, London, New York.
Hughes A., Lascaratou C. (1982): Competing Criteria for Error Gravity, „English Language
Teaching Journal”, t. 36/3, s. 175–182.
Jones S. (1982): Attention to Rhetorical Form While Composing in a Second Language [w:]
C. Campbell et al. red. Proceedings of the Los Angeles Second Language Research
Forum, t. 2, Los Angeles, s. 130–143.
Jones S., Tetroe J. (1987): Composing in a Second Language [w:] A. Matsuhashi red. Writing
in Real Time: Modelling Production Processes, Norwood, NJ, s. 34–57.
Krapels A.R. (1990): An Overview of Second Language Writing Process Research [w:]
B. Kroll, Second Language Writing, Cambridge, s. 37–56.
Leki I. (1992): Understanding ESL Writers: A Guide for Teachers, London, Portsmouth, NH.
Martin-Betancourt M. (1986): The Composing Processes of Puerto Rican College Students of
English as a Second Language [Nieopublikowana praca doktorska, Fordham University].
Raimes A. (1985a): An Investigation of the Composing Processes of ESL Remedial and
Nonremedial Students [Paper presented at the 36th Annual CCCC Convention,
Minneapolis, Minn., March].
Raimes A. (1985b): What Unskilled Writers Do as They Write: a Classroom Study of
Composing, „TESOL Quarterly”, t. 17, s. 535–552.
Raimes A. (1987): Language Proficiency, Writing Ability, and Composing Strategies: A Study
of ESL College Student Writers, „Language Learning”, t. 37, s. 439–468.
Raimes A. (1991): Out of the Woods: Emerging Traditions in the Teaching of Writing,
„TESOL Quarterly”, t. 25, s. 407–430.
Richards J. (1971): A Non-Contrastive Approach to Error Analysis, „English Language
Teaching”, t. 25, s. 204–219.
Selinker L. (1972): Interlanguage, „International Review of Applied Linguistics”, t. 10,
Silva T. (1993): Toward an Understanding of the Distinct Nature of l2 Writing: The ESL
Research and Its Implications, „TESOL Quarterly”, t. 27/4, s. 657–677.
Zamel V. (1982): Writing: The Process of Discovering Meaning, „TESOL Quarterly”, t. 16,
Zamel V. (1983): The Composing Processes of Advanced ESL Students: Six Case Studies,
„TESOL Quartely”, t. 17, s. 165–187.
178 EWA WITALISZ
Transfer w pisaniu w języku obcym: kompetencja językowa a umiejętność
Transfer jest kojarzony przez nauczycieli języka obcego z negatywnym wpływem języka
ojczystego na rozwój kompetencji w języku obcym. Świadectwem takiego transferu są liczne
błędy, które wynikają ze stosowania przez uczących się nieprawidłowej strategii dosłownego
tłumaczenia struktur językowych. Jednakże w przypadku rozwijania sprawności pisania
w języku obcym, transfer dotyczy nie tylko kompetencji językowej, ale także umiejętności
komponowania tekstu. Wyniki badań w tej dziedzinie nie są jednoznaczne: niektórzy
językoznawcy podkreślają różnice pomiędzy komponowaniem tekstu w języku pierwszym
i drugim, inni potwierdzają podobieństwa, dzięki czemu studenci sprawnie piszący w swoim
języku również biegle komponują tekst w języku obcym, a więc w tym zakresie można się
spodziewać transferu pozytywnego.
W niniejszym artykule kwestie transferu językowego oraz umiejętności komponowania
tekstu zostały zilustrowane przez autorkę przykładami z badań własnych nad rozwojem
sprawności pisania na poziomie zaawansowanym u Polaków uczących się języka
angielskiego jako obcego.