421890

890Maxwell-ReidWritten Communication
2011 SAGE Publications

Reprints and permission:

WCX28410.1177/0741088311421

sagepub.com/journalsPermissions.nav

Written Communication

The Challenges of 28(4) 417435

2011 SAGE Publications
Reprints and permission:
Contrastive Discourse sagepub.com/journalsPermissions.nav
DOI: 10.1177/0741088311421890
Analysis: Reflecting on http://wcx.sagepub.com

a Study into the Influence

of English on Students
Written Spanish on
a Bilingual Education
Program in Spain

Corinne Maxwell-Reid1

Abstract
This article discusses challenges involved in contrastive discourse analysis
that emerged while carrying out a follow-up study into a Content and
Language Integrated Learning (CLIL) program in Spain. Reversing the focus
on English of much contrastive rhetoric work, the study investigates the effect
of second-language-English on first-language-Spanish writing. The motivation
for this focus and the choice of tools from Systemic Functional Linguistics
(SFL) for genre and clause analysis are discussed. Reflecting on the difficulties
involved in contrastive discourse analysis, in particular the challenges of com-
paring texts, it is suggested that contrastive work benefits from a more differ-
entiating analytical method and a more dynamic conception of language. The
implications of an influence from English are also considered, with the theses
of hybridity and of homogeneity contributing to indicate a role for language
awareness work in schools.

Keywords
Content and Language Integrated Learning (CLIL), systemic functional linguistics,
homogenization, hybridization, contrastive rhetoric, genre
1
The Chinese University of Hong Kong, Hong Kong

Corresponding Author:
Corinne Maxwell-Reid, Department of Curriculum and Instruction, The Faculty of Education,
The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong
Email: cmaxwellreid@cuhk.edu.hk
418 Written Communication 28(4)

Discourse analysis that compares texts across languages is most widely asso-
ciated with contrastive rhetoric (Connor, 1996). Contrastive rhetoric (CR) was
originally carried out to help students from other language backgrounds
adapt their written English for study purposes in English-speaking con-
texts. The Anglo-centric assumptions underlying such intentions have
been pointed out (Canagarajah, 2002b; Kubota, 1997), and CR has developed
beyond such a deficit approach, but it is still closely associated with the teach-
ing of English (Connor, Nagelhout, & Rozycki, 2008). However, the compari-
son of texts across languages is also needed for purposes other than teaching
English. One such purpose is to investigate the influence of second-language
English on first languages, a generally less studied direction of interest (Arcay
Hands & Coss, 2004; Jarvis & Pavlenko, 2007), but increasingly relevant, as
concern for cultural diversity accompanies the increased role for English
through globalization (Bernrdez, 2008; Siguan, 2005). This article considers
the challenges of carrying out such contrastive work in the context of a Con-
tent and Language Integrated Learning (CLIL) program in Spain. CLIL is a
form of bilingual education currently promoted for foreign language learning
in Europe and particularly associated with the learning of English (Coyle,
Hood, & Marsh, 2010). The effect of bilingual education on first-language
proficiency has been widely studied (Genesee, 2004); however, its effect on
first-language discourse choices and cultural norms is only now starting to be
addressed. Maxwell-Reid (2010) reports an initial study into the influence of
English on CLIL students Spanish writing, which will be briefly summarized
here along with the results of a follow-up study to provide a basis for the
discussion of difficulties that emerged from this work. The discussion starts
from specific difficulties with the study data, particularly in terms of the com-
plexities of comparing different texts. It then moves onto more general issues
for contrastive work, considering the effectiveness of analytic tools, and the
implications and response to an influence from English. First, however, the
motivation for this focus on the impact of English on other languages will be
further considered by examining the view of English as threatening cultural
diversity.

Culture, Language, and Schooling,

and the Global Role of English
The languageculture connection underlying this concern at the increase in
English learning and use is most closely associated with the work of
Benjamin Whorf. Historically interpreted in cognitive terms as the influence
of language on thought, Whorf (1956) has received negative and positive
Maxwell-Reid 419

reactions that divide along universalist and relativist lines, respectively, and
though the universalist orientation has been dominant (Cameron, 1999), lin-
guistic relativity has been revisited more favorably (Gumperz & Levinson,
1996). This revitalized relativist view demonstrates a focus, not on cognition,
but on the relationship between language in use and the cultural context of
that use (Risager, 2006).
The term culture is used in many different ways; here the emphasis is on
the knowledge of interacting systems of meaning, including language, which
we use to operate as a member of a community (Kachru, 1995). The degree to
which other meaning systems such as clothing or food depend on language,
and thus, the strength of the languageculture connection, does not receive
general agreement. Risager (2006) warns against essentialism and emphasizes
the limitations to concepts such as an indivisible languaculture, arguing that
language and culture can also be viewed separately in specific instances.
Nevertheless, she points out that language, including a language used as a
lingua franca, is always a bearer of culture (Risager, 2006, p. 134). Hence,
the unease at the spread of English: As it is used increasingly across Europe in
education, academic work, and EU organizations, there is growing concern
that it may lead to a more monocultural community (Bernrdez, 2008; Dendrinos,
2002; Phillipson, 2003; Siguan, 2005). The role of schooling in this process is
particularly important, as in school children simultaneously learn language and
learn through language, making it a key site for the development of their lan-
guaculture (Halliday, 1978; Risager, 2006).
However, attitudes toward the influence of English are mixed, and a num-
ber of frameworks have been developed to explore the different positions,
most notably the contrasting views of homogenization and hybridization
(Alcn Soler, 2007; Canagarajah, 2002a; Pennycook, 2007; Singh & Doherty,
2004). Homogenization presents the increased use of English as a negative
development, with the standardization of globalization bringing unfortunate
consequences for linguistic and cultural diversity (Phillipson, 2003), whereas
hybridization focuses on the diversity in use of language that globalization
may foster through processes of synthesis or fusion (Alcn Soler, 2007;
Canagarajah, 2002a). These distinctions will be returned to in relation to
understandings of the influence of English on students first-language dis-
course and also as regards contrastive work more generally.

Spanish and English Discourse Compared

Contrastive discourse work recognizes that the influence of one language on
another is most appropriately investigated at the level of discourse, as it is
420 Written Communication 28(4)

with text that we achieve our social purposes (Martin & Rose, 2008). Users
of Spanish and English are aware of discourse differences between the two
languages, with Spanish text described as more complex, more elaborate, and
less linear or explicit (St. John, 1987; Snchez Escobar, 1996). CR investi-
gates such subjective understandings by linking them to specific linguistic
features with, for example, complexity and elaboration measured in sentence
or clause length and also subordination, and linearity or digression through
thematic progression or use of markers of semantic relations. Spanish dis-
course has repeatedly been found to use longer sentences and more subordina-
tion, be more digressive and use less metatext to orient readers (Mur Dueas,
2007; Neff, Dafouz, Dez, Prieto, & Chaudron, 2004; Reppen & Grabe, 1993;
Snchez Escobar, 1996). Such discourse choices can be strongly defended,
with the use of English norms in a translation from Spanish even resulting in
accusations of cultural imperialism (Beeby Lonsdale, 1996). Another sugges-
tion is that Spanish depends less on established norms in text organization
(Newman, Trenchs-Parera, & Pujol, 2003; Snchez Escobar, 1996). These
insights provide potential forms of analysis for the comparison of the CLIL
and non-CLIL student texts. However, the contrastive work mentioned here
analyzes texts from a range of contexts, with difference between studies
including geographical location, age of writers, purposes for writing, and
writerreader relationships. Such variables make cross-study comparison
difficult and have resulted in accusations of overgeneralizing (Kachru, 1995;
Leki, 1991). Other criticisms of CR include that it disregards inequalities of
power between discourses and has a normative, Angi-centric orientation
(Canagarajah, 2002b). The reconception of CR as intercultural rhetoric
(Connor, 2004) has helped to address some of these issues, but adaptations
are still needed to clarify findings and increase the explanatory power of the
analysis. Such adaptations made to the analytic tools for the study considered
here will be discussed next; the relative effectiveness of approaches, along
with other criticisms of CR, will be reconsidered after a brief summary of
that study.

Tools for Text Analysis

Tools for the comparison of the CLIL and non-CLIL students writing will
be considered first at the lexicogrammatical and then the discourse level. As
introduced above, studies report Spanish as tending to use longer sentences
and clauses and also more subordination than English discourse. However,
each of these areas requires greater specification for the comparison of the
CLIL and non-CLIL students writing. The case of subordination illustrates
Maxwell-Reid 421

the uncertainty surrounding some CR work and also introduces issues needed
for the analysis of sentences and clauses more generally.
Subordination is particularly problematic as a category for analysis because
it includes two types of complexity: complexity within the clause and com-
plexity between clauses. This distinction is reflected in the two concepts of
embedding and hypotaxis as used in Systemic Functional Linguistics (SFL).
Embedded clauses are clauses that have been downranked to function within
another grammatical unit, thus creating complexity within the clause, whereas
hypotaxis describes relationships between clauses, specifically the unequal
relationship between a dependent clause and an independent clause (Halliday
& Matthiessen, 2004). One sentence from a CLIL students text illustrates the
difference:

(1) 4a Tampoco me gustan los padres [[que no se preocupan

nada de sus hijos]],
4b ya que, aunque le quieran,
4c le tienen que llamar
4d para ver
4e si esta bien.
4a Neither do I like the parents [[that dont worry at all
about their children]],
4b since, although they love them,
4c they have to call them
4d to see
4e if they are okay.
[[ ]] indicates embedded clause.

The first line, 4a (the first clause of the fourth sentence in the text), includes
a clause downranked to act not at clause level but within the nominal group
los padres (the parents). (Traditional grammar would term this a defining
relative clause, and it can be contrasted with a nondefining relative clause,
which does function at clause level.) The following clauses, 4b to 4e, repre-
sent four hypotactic clauses, that is, ranking clauses (clauses functioning at
clause level) that are dependent on the main or independent clause, 4a. This
distinction between hypotaxis and embedding is important when considering
language at text level, as hypotactic clauses have relations with other clauses
and so contribute directly to discourse structure, which embedded clauses
cannot. In much CR work, however, these different clauses, and thus rela-
tions within text, are not distinguished.
422 Written Communication 28(4)

The above extract also illustrates more general issues for clause analysis.
Relationships between clauses, and the identification of clause boundaries, are
determined by two systems within SFL: taxis and logico-semantic relations.
Taxis describes the degree of interdependency between clauses, with hypo-
taxis as introduced above describing an unequal relation of dependency, and
parataxis indicating clauses of equal status. (That is, each clause in a paratactic
relationship is independent and could stand alone.) The second system in the
relationship between clauses is logico-semantic relations, which also informs
the identification of clauses and thus is needed for analysis. Logico-semantic
relations divide into the two general relationships of expansion and projection
(Halliday & Matthiessen, 2004; Lavid, Ars, & Zamorano-Mansilla, 2010).
Expansion can be illustrated from (1) above, with, for example, 4b qualifying
4c. Projection is a different type of relationship, where the speech or thought
of one clause is projected through another clause, as exemplified in 4d and 4e
in (1) above (see here meaning find out rather physical perception).
The two systems of taxis and logico-semantic relations can thus inform the
analysis of the students sentences into clauses, and also distinguish clause
complexes and simplexes (sentences of one single clause). In this way the
relative complexity of the CLIL and non-CLIL texts can be investigated in
similar, but more specific, terms to previous CR comparisons of English and
Spanish. However, the sentence is an orthographical rather than a grammatical
unit, and relying on it alone as the unit of analysis could hide or exaggerate
other differences. An alternative measure is the t-unit, developed by Hunt
(1965) to study writing development, and consisting of an independent clause
and any clauses dependent on it.
At the discourse level, the intuitions around English linearity as opposed
to Spanish digression are less easily investigated, with thematic progression
found to be more complicated than previously thought (McCabe, 2004).
Genre, with its work on staging of texts depending on purpose, provides a
clearer starting point for analysis. Factual genres used in schooling include
reports, explanations, and expositions/argument. Reports are typically used to
describe physical phenomena, for example, classifying different organisms in
one ecosystem, and explanations operate with sequences of cause and effect
(Martin & Rose, 2008). Argumentative writing is particularly associated with
schooling and also more likely to differ between cultures (Hatim & Mason,
1990). Martin and Rose describe three expository or argumentative genres in
terms of purpose and organizational basis. Exposition promotes one viewpoint
and follows a general organization of introduction with thesis, arguments in
support, and conclusion with thesis restated. Discussion is a two-sided
(or multisided) argument that considers alternative positions relating to an
issue with the purpose of indicating the preferred response. Third, a challenge
Maxwell-Reid 423

starts with an opposing position, responds to that position using rebuttals and
counterarguments, and finishes with the preferred viewpoint. The signaling
of these frameworks can also be used to investigate reader orientation.
The tools described here then allow for more specific analysis at lexico-
grammatical and discourse levels and were used to compare CLIL and non-
CLIL student texts written in response to two prompts. However, difficulties
were still encountered with the analysis, and the results raised further issues
for contrastive discourse work. A brief summary of the analysis and results
will be presented here, and then the challenges for contrastive discourse that
arose from the study will be discussed. Greater attention will be paid to the
texts of Prompt 2, as results from Prompt 1 have already been discussed in
Maxwell-Reid (2010).

Summary of Textual Analysis

The follow-up study that gave rise to this article then asks whether the find-
ings from a second prompt suggest a similar influence from English discourse
norms on the CLIL students written Spanish as was found with the results of
Prompt 1. That is, whether the CLIL texts, in comparison with the non-CLIL
texts, show:

1. Shorter grammatical and orthographical units (words per sentence,

t-unit and clause)
2. Less subordination (fewer embedded clauses and ranking clauses
per sentence)
3. More simplexes
4. Greater use of text frameworks
5. More signaling of text frameworks

The prompts used were as follows (English versions).

Prompt 1: School Uniform

At the moment, only students at private schools wear a uniform.
However, some politicians also want public schools to have a uniform
for their students. Do you think it is a good idea for public schools to
have school uniform?
Write a page for your school magazine on this question, explaining
your opinion on the topic. Include examples to help make your expla-
nations clear.
424 Written Communication 28(4)

Prompt 2: Parental Control

Teenagers often complain that their parents are always telling them
what to do. Parents say that they know what is best for their children,
and that teenagers are too young to make their own decisions. Do you
think parents control their teenage children too much?
Write a page for your school magazine on this question, explaining
your opinion on the topic. Include examples to help make your explana-
tions clear.

Context of Study and Data Collection

The texts were collected from a secondary school2 in Madrid taking part in
the Spanish MEC/British Council Bilingual Project that extends from Primary
1 to Secondary 4. In the secondary schools the students study with the regular
Spanish-medium students for two thirds of the curriculum and have separate
classes for English and subjects taught through English, in this case social
sciences and IT. The texts were written by 14-year-old Secondary 3 students,
CLIL and non-CLIL, during their normal Spanish classes, with in each case
half the students writing on school uniforms (Prompt 1) and half on parental
control (Prompt 2). (Each student later wrote in English on the topic they had
not addressed in Spanish; these English texts are not reported in this article.)
Thus, 12 CLIL students and 12 non-CLIL students wrote on each prompt.
The students were allowed 5 minutes to discuss the topic and then another
40 minutes to write the text individually.

Summarized Results
As can be seen in Table 1, Prompt 2 produced much smaller differences
between the CLIL and non-CLIL texts than Prompt 1, and even some reversed
findings. Thus, for both prompts, CLIL texts have shorter t-units and fewer
embedded clauses per sentence, but relative text and sentence lengths do not
correspond between the two prompts and nor do ranking clauses per sentence.
(Example 1 above illustrates both ranking and embedded clauses.)
Table 2 shows that for both prompts the CLIL students used a higher pro-
portion of clause simplexes, and correspondingly fewer clause complexes,
than the non-CLIL students. Again the difference was greater for Prompt 1
than Prompt 2, and there were also differences in the length of clause com-
plexes between the two prompts. (Examples 1 above and 5 below show a
clause complex of five clauses; Example 4 below shows a simplex and a two-
clause complex.)
Maxwell-Reid 425

Table 1. Units Measured

Prompt 1 Prompt 2

CLIL Non-CLIL CLIL Non-CLIL

students students students students
Words per text 129 183 178 170
Words per sentence 23 28 29 28
Words per t-unit 16 17 15 17
Words per ranking clause 8.3 9 8.1 8.1
Ranking clauses per sentence 2.7 3.1 3.6 3.5
Embedded clauses per sentence 1.0 1.4 1.2 1.3

Table 2. Clause Complexing: Total Counts and Percentages

Clause simplexes Clause complexes

% of % of
Totals sentences Totals sentences
Prompt 1
CLIL students 22 32 46 68
Non-CLIL students 7 9 71 91
Prompt 2
CLIL students 20 27 53 73
Non-CLIL students 12 16 61 84

Note: CLIL = Content and Language Integrated Learning.

As Table 3 shows, the results for text structure are very different for the
two prompts, with Prompt 1 producing more homogeneous CLIL texts that
differ more clearly from the non-CLIL texts. (Prompt 1 organizational frame-
works do not add up to 12 as some texts did not clearly use any of these
frameworks.)

Issues Raised by the Two Sets of Results

Thus, Prompt 2 rendered results that do not fully correspond to the findings
from Prompt 1 and do not consistently align the CLIL texts with English
rather than Spanish discourse norms. Further research is then needed to
clarify the influence of English on CLIL students first-language written
discourse; such research involves a number of challenges, some of which
will now be considered.
426 Written Communication 28(4)

Table 3. Text Structure Totals

Prompt 1 Prompt 2

CLIL Non-CLIL CLIL Non-CLIL

students students students students
Organized as classification 0 0 3 0
Organized as exposition (one-sided 0 1 2 3
argument)
Organized as discussion (2+ sided 7 2 7 8
argument)
Organized as challenge 0 (2 0 1
(counterargument & rebuttal) partially)
Organizational framework signposted 6 1 4 2

Note: CLIL = Content and Language Integrated Learning.

Methodological and Conceptual Challenges

in Contrastive Discourse Analysis
First, methodological challenges arising directly from this study will be dis-
cussed, particularly the comparison of responses to different prompts. These
practical issues will then lead to the consideration of two more conceptual
areas of challenge: tools for contrastive discourse analysis, and understand-
ings of the influence of English.

Comparing Texts
The differences in the text structure results for the two prompts (Table 3)
demonstrate the problems of comparing even quite similar text types and
also indicate the need for a multidimensional theory of language to explore
differences. The complexities of comparing texts will be illustrated with
three issues connected to the very central role of genre: the students differing
interpretations of Prompt 2, the difficulty of using genre categories, and the
interdependency between genre and specific linguistic choices. The role of a
theory of language will be addressed in a later section.
It became clear that Prompt 2 could be interpreted in two ways, with some
students focusing on whether parents did control their children too much, and
others on whether they should control children so much. Texts also varied in
who they addressed, with some directed at fellow students and others for a
more general readership. This varied interpretation had an impact on the find-
ings and was perhaps due to insufficient contextualization: The prompts asked
Maxwell-Reid 427

students to write for the school magazine (which does exist), but the purpose
of the text could have been clarified and the intended readership emphasized.
Categorizing the texts was also frequently difficult, with texts showing ele-
ments of more than one genre. Categorization can lead to artificial boundaries,
but comparison between two sets of data is difficult without it, and thus, the
more borderline texts were decided according to the overarching framework
used. For example, three CLIL texts used classification as an organizing prin-
ciple for Prompt 2; the following extract illustrates the use of separate para-
graphs to describe and then comment on particular parenting styles in one of
the texts:

(2) Algunos padres son muy estrictos y estn siempre encima de sus
hijos, preguntandoles que es lo que es lo que hacen cuando salen con
quienes se relaciona. . . Este tipo de padres yo creo son un poco estrictos
porque se preocupan mucho por sus hijos y tal vez sus hijos se sienten
agoviados.
[Some parents are very strict and always on top of their children,
asking them what they do when they go out who they mix with . . . I
think this type of parent is a bit strict because they worry a lot about
their children and perhaps their children feel smothered.]

One non-CLIL text seemed similar to Example 2, but rather than using the
classification as the ordering principle within which each parenting style is
evaluated, the non-CLIL text starts with the overall judgment, that parents
almost always overcontrol their children, and then introduces limitations to
that judgment using two categories of situation. The contrasting organizations
thus correspond to the overarching purpose of each text: classification in the
case of CLIL Text Example 2, and argument for non-CLIL Text Example 3.

(3) Yo pienso que los padres en casi todas las ocasiones, controlan
demasiado a sus hijos. A m no me parece bien, porque muchos de esos
hijos tienen suficiente conciencia para controlarse ellos solos. Pero en
otros casos si me parece bien, porque a los hijos les da todo igual y hay
que controlarles para que se den cuenta de la realidad.
[I think that parents on almost all occasions control their children too
much. I dont think this is good, because a lot of these children have
enough awareness to control themselves on their own. But in other cases
I do think its good, because the children could not care less and they
have to be controlled so that they wake up to reality.]
428 Written Communication 28(4)

A further indicator of the importance of genre was the contrast in use of

complexes between the two studies. For both prompts, CLIL texts used more
simplexes and fewer complexes than non-CLIL texts, but there were differ-
ences in the use of complexes between the two prompts, with both CLIL and
non-CLIL texts using longer clause complexes for Prompt 2 texts (parental
control) than for Prompt 1 texts (school uniform). As has been found else-
where (Schleppegrell & Colombi, 1997), complexing choices correspond to
stages in the text, and different text types will thus have different requirements
for the use of simplexes and complexes. For the two prompts, both CLIL and
non-CLIL texts use simplexes to state the opinion or thesis, in contrast with
explanations and reasons, which are generally expressed in complexes. The
more explicit signaling of the genre structure thus tends to correlate with
greater use of simplexes. For example, a Prompt 2 CLIL text starts with a
simplex to express a main point, following it with a two-clause clause com-
plex to illustrate the point:

(4) 1 En mi opinin, hay veces [[que se preocupan demasi-

ado]].
2a En cuanto llegas 15 minutos tarde
2b ya estn llamando a tu telfono mvil, al vecino, al
primo, a tu amigo
[In my opinion there are times when they worry too much.
As soon as you are 15 minutes late
they are already calling your mobile phone, the neighbor, the
cousin, your friend . . . ]
(Double square brackets [[ . . . . ]] indicate embedding)

There is a similar tendency in the texts for Prompt 1, with simplexes used
not only to express opinions but also to mark shifts, for example, from disad-
vantages to advantages of school uniform. However, the relatively longer
complexes of Prompt 2 may be linked to the different requirements of the two
topics. Many of the Prompt 2 arguments were highly nuanced and used more
extended explanation than Prompt 1 texts. An example from a non-CLIL text
of a two-clause opinion sentence followed by a five-clause development of
the point illustrates:

(5) 1a Creo
1b que en muchas ocasiones los padres si que controlan
demasiado a los hijos.
2a A veces porque piensan
Maxwell-Reid 429

2b que no somos lo suficientemente responsables [[para asumir

o hacer algunas cosas]]
2c o por querer protegernos,
2d pero en algunos casos esa actitud puede producir [[que
el adolescente quiera hacer con mas ganas aquello [[que
le prohben]]]]
2e  para demostrar [[que si es capaz y responsable [[para
hacerlo]].
[I believe
that on many occasions parents do control their children too much.
Sometimes because they think
That we are not responsible enough to take on or do some things
or because they want to protect us,
but in some cases this attitude can make the adolescent want to do
what they have forbidden them even more
in order to demonstrate that they are capable and responsible enough
to do it.]

These three examples of the central role of genre thus underline the dangers
of considering features of text in isolation and the possible pitfalls of compar-
ing across even similar text types. Differences in specific text type entail dif-
ferences in the use of lexicogrammatical resources, as with the complexes
above, even when many aspects of text such as the readerwriter relationship
remain comparable. Furthermore, failure to fully clarify the purpose of writing
can result in varying interpretations of the task, leading to differences in task
realization, and reduced comparability within responses to one prompt. Even
without these two problems, categorization of texts into genres is in itself fre-
quently awkward as it places discrete boundaries on what is more properly a
continuum. These contrastive issues will now be further considered in the
more general discussion of tools for contrastive discourse work.

The Practice and Tools of Contrastive Discourse Analysis

The contrastive work on text structure described above involves a number of
complexities. However, the lexicogrammatical analysis using approaches from
previous studies in Spanish-English contrastive rhetoric (CR) is also problem-
atic. Practical difficulties with using analytic tools from CR were illustrated
with the case of subordination as described above. This category was clarified
through division into ranking and embedded clauses, and inconsistencies in
some results may be related to the use of different prompts as discussed above
430 Written Communication 28(4)

for clause complexing. Nevertheless, the lexicogrammatical analysis still

shows limitations, as complexity involves more than length and combina-
tions of grammatical or orthographical units. In addition to the practical
limitations of contrastive tools, and as was suggested earlier in this article,
CR suffers from more conceptual problems.
One issue that combines the practical and the conceptual challenge to CR
is its tendency toward a static conception of language. A contrastive study
such as this one is in danger of viewing languages as discrete, fixed entities by
comparing Spanish with English. This type of essentialist position ignores
the political motivations to language boundaries and the complex flow of
influences between instances of language use (Bernrdez, 2008; Dendrinos,
2002; Pennycook, 2007; Risager, 2006). The CR community has shown some
awareness of this problem in their move to intercultural rhetoric (IR): IR and
related scholars advocate the study of small cultures and warn against general-
izing too far from data (Connor, 2008). However, while reducing the risk of
essentialism, the move from CR to IR illustrates some of the drawbacks to the
opposing, relativist, position. IR, with its emphasis on small cultures and the
process of communication between cultures (Connor, 2008), may be only able
to comment on specific cases of intercultural communication, without any
possibility of generalizing beyond them. Indeed, in a sense CR/IR had little
choice but to retreat from generalizations, as they had no underlying theory or
system to generalize from or to, and thus, were frequently reduced to specula-
tion or stereotypes (Leki, 1991; Li, 2008). An increased role for SFL in con-
trastive work could help address these difficulties, as its multidimensional
view of language as a network of systems allows for both greater specification
and also a degree of generalization. In particular, the cline of instantiation
positions a text as a particular instance, and also as part of the overall potential
of the system (Halliday & Matthiessen, 2004, p. 27), thus enabling the dual
perspective of the specific case and its relation to a more general system, plus
intermediary positions such as text type. The multidimensional view also
increases explanatory power, as features can be discussed in relation to their
function within systems, as was briefly illustrated above with the case of
clause complexing and text types.
For the study discussed here, the choice of analytic tools was driven largely
by a concern to stay close to previous forms of contrastive analysis and so has
not fully taken advantage of the resources of SFL. However, the analysis of
this study was more revealing when it did depart from previous CR, for exam-
ple, in the consideration of clause complexes and simplexes, and in the more
structural approach to text organization. Additional possibilities for analysis
include the investigation of complexity through the use of grammatical
Maxwell-Reid 431

metaphor. The concept of grammatical metaphor describes the process whereby

meanings associated primarily with one type of linguistic resource are
expressed in another; for example, a verbal process is realized by a nominal-
ization of the verb, thereby enabling, among other things, greater lexical den-
sity (Halliday & Matthiessen, 2004).

Implications of a CLIL Influence

on First-Language Discourse
Further evidence of an influence from English on CLIL students written
Spanish would lead to other issues that require consideration. One such issue,
outside the scope of the present discussion, is to investigate the source of the
influence from English. For the study that led to this discussion, such investiga-
tion involved comparing the CLIL and non-CLIL students experiences of text,
primarily in terms of their English lessons, but also considering other CLIL
subjects, Spanish lessons, and experiences from outside the regular classroom.
Maxwell-Reid (2010) provides more information on these investigations.
An additional challenge is to understand and respond to an influence from
English. Discussion has been polarized through the concepts of homogeneity
and hybridity as introduced above (Canagarajah, 2002a; Pennycook, 2007;
Singh & Doherty, 2004). The homogeneity position might interpret corre-
spondence between CLIL students Spanish writing and English discourse
norms as a further example of the dominance of English leading to a more
unified, Anglo-centric discourse (Phillipson, 2003) but would have less to
say on the process of change within language/s. A hybridity position, on the
other hand, has processes of change as central, but divergence in how they are
viewed leads to two alternative responses. Hybridization has been used to
argue that English, particularly English as a Lingua Franca (ELF), has no
impact on other languages (Alcn Soler, 2007). An alternative position,
developed principally by Canagarajah (2002b), but also, for example, by
Pennycook (2007), has greater relevance to the present study and to the dis-
cussion of relativism. Canagarajah (2002b) combines the political awareness
of the homogeneity arguments with the understanding of the multiple, fluid
nature of language within the hybridity rhetoric, and thereby, also addresses
some of the problems associated with CR work as mentioned above. Holders
of this more complex position agree that languages are not fixed, discrete
entities to be preserved unchanged, but are also aware that a hierarchy exists
in this cultural flow and that certain forms of English, along with the speakers
of these Englishes, have valuable cultural or linguistic capital (Bernrdez,
2008; Canagarajah, 2002b), possibly to the disadvantage of others. This point
432 Written Communication 28(4)

of view might then suggest the value of language, or languacultural, aware-

ness work with CLIL students (Canagarajah, 2002b).
Language awareness work could be carried out in both language and con-
tent classes with CLIL students and include consideration of published texts
as well as working with students own texts. Students could consider pub-
lished texts in terms of purpose, structure of texts, and related options for the
writer to help them realize that language use involves choices and that differ-
ent choices achieve different effects. The students own texts can also be used
as the focus for investigation, for example, with manipulations such as divid-
ing up a text altered to have no punctuation or rewriting a text as a different
genre. These are not new ideas for language teaching, but the activities are put
to a slightly different purpose: raising students awareness that different
options exist, for combining information into clauses, for organizing texts and
so on, and also, if this turns out to be the case, that certain of these options tend
to be favored by different communities. Equipped with this greater awareness,
students could then decide which discourse norms to put into writing and thus
learn to manage their own language in global context.

Conclusion
This article argues that in our more globalized world contrastive discourse
work needs to be extended to the effect of English on students first languages.
It discusses the challenges for such contrastive discourse work based on the
example of a study investigating the influence of second-language English
on first-language written Spanish in the context of a CLIL program, though
the challenges are not restricted to this context. It is suggested that genre is
of central importance when considering text comparability and that the effec-
tiveness of contrastive work is increased through the use of a more differen-
tiating analytical method and a more dynamic conception of language.

Authors Note
I would like to thank Christina Haas and the anonymous reviewers for their very help-
ful comments and advice during the preparation of this article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interests with respect to the authorship
and/or publication of this article.

Funding
The author(s) received no financial support for the research and/or authorship of this
article.
Maxwell-Reid 433

Notes
1. Results for Prompt 1 of this study have been previously published as Maxwell-
Reid (2010). Content and Language Integrated Learning (CLIL): The Influence of
Studying Through English on Spanish Students First Language Written Discourse.
Text & Talk 30(6), 679-699.
2. I would like to thank the school for their great generosity with their time and assis-
tance during this study.

References
Alcn Soler, E. (2007). Linguistic unity and cultural diversity in Europe: Implications
for research on English language and learning. In E. Alcn Soler & M.P. Safont
Jord (Eds.), Intercultural language use and language learning (pp. 23-39). New
York, NY: Springer.
Arcay Hands, E., & Coss, L. (2004). Academic biliteracy and the mother tongue. In
C. L. Moder & A. Martinovic-Zic (Eds.), Discourse across languages and cul-
tures (pp. 285-299). Philadelphia: John Benjamins.
Beeby Lonsdale, A. (1996). Teaching translation from Spanish to English: Worlds
beyond words. Ottawa, Ontario, Canada: University of Ottawa.
Bernrdez, E. (2008). El lenguaje como cultura: Una crtica del discurso sobre el
lenguaje. [Language as culture: A critique of the discourse on language]. Madrid,
Spain: Alianza.
Cameron, D. (1999). Linguistic relativity: Benjamin Lee Whorf and the return of the
repressed. Critical Quarterly, 41(2), 153-156.
Canagarajah, S. (2002a). Reconstructing local knowledge. Journal of Language,
Identity, and Education, 1(4), 243-259.
Canagarajah, S. (2002b). Multilingual writers and the academic community: Towards
a critical relationship. Journal of English for Academic Purposes, 1, 29-44.
Connor, U. (1996). Contrastive rhetoric: Cross-cultural aspects of second language
writing. Cambridge: Cambridge University Press.
Connor, U. (2004). Intercultural rhetoric research: Beyond texts. Journal of English for
Academic Purposes, 3, 291-304.
Connor, U. (2008). Mapping multidimensional aspects of research: Reaching to intercul-
tural rhetoric. In U. Connor, E. Nagelhout, & W. Rozycki (Eds.), Contrastive rheto-
ric: Reaching to intercultural rhetoric (pp. 299-315). Philadelphia: John Benjamins.
Connor, U., Nagelhout, E., & Rozycki, W. (Eds.). (2008). Contrastive rhetoric:
Reaching to intercultural rhetoric. Philadelphia: John Benjamins.
Coyle, D., Hood, P., & Marsh, D. (2010). CLIL: Content and language integrated
learning. Cambridge, UK: Cambridge University Press.
Dendrinos, B. (2002). The marketisation of English as a global(ising) language: Dis-
courses and counter discourses. In M. Kalantzis, G. Varnava-Skoura, & B. Cope
434 Written Communication 28(4)

(Eds.), Learning for the future: New worlds, new literacies, new learning, new
people (pp. 241-255). Altona, Melbourne, Victoria, Australia: Common Ground.
Genesee, F. (2004). What do we know about bilingual education for majority-language
students? In T. Bhatia & W. Ritchie (Eds.), The handbook of bilingualism
(pp. 547-576). Oxford, UK: Blackwell.
Gumperz, J., & Levinson, S. (Eds.). (1996). Rethinking linguistic relativity. Cambridge,
UK: Cambridge University Press.
Halliday, M. A. K. (1978). Language as social semiotic: The social interpretation of
language and meaning. London: Edward Arnold.
Halliday, M. A. K., & Matthiessen, C. M. I. M. (2004). An introduction to functional
grammar. London: Arnold.
Hatim, B., & Mason, I. (1990). Discourse and the translator. New York: Longman.
Hunt, K. (1965). A synopsis of clause-to-sentence length factors. The English Journal,
54(4), 300-309.
Jarvis, S., & Pavlenko, A. (2007). Crosslinguistic influence in language and cognition.
New York: Routledge.
Kachru, Y. (1995). Cultural meaning and rhetorical styles: Toward a framework for
contrastive rhetoric. In G. Cook & B. Seidlhofer (Eds.), Principle and practice:
Studies in honour of H. G. Widdowson (pp. 171-184). Oxford, UK: Oxford
University Press.
Kramsch, C. (1998). Language and culture. Oxford, UK: Oxford University Press.
Kubota, R. (1997). A reevaluation of the uniqueness of Japanese written discourse:
Implications for contrastive rhetoric. Written Communication, 14(4), 460-480.
Lavid, J., Ars, J., & Zamorano-Mansilla, J. R. (2010). Systemic functional grammar
of Spanish: A contrastive study with English. London: Continuum.
Leki, I. (1991). Twenty-five years of contrastive rhetoric: Text analysis and writing
pedagogies. TESOL Quarterly, 25(1), 123-143.
Li, X. (2008). From contrastive rhetoric to intercultural rhetoric: A search for collec-
tive identity. In U. Connor, E. Nagelhout, & W. Rozycki (Eds.), Contrastive rheto-
ric: Reaching to intercultural rhetoric (pp. 11-24). Philadelphia: John Benjamins.
Martin, J. R., & Rose, D. (2008). Genre relations: Mapping culture. London: Equinox.
Maxwell-Reid, C. (2010). Content and language integrated learning (CLIL): The
influence of studying through English on Spanish students first language written
discourse. Text & Talk, 30(6), 679-699.
McCabe, A. (2004). Thematic progression patterns and text types in history textbooks.
In D. Banks (Ed.), Text and texture: Systemic functional viewpoints on the nature
and structure of text (pp. 215-237). Paris: LHarmattan.
Mur Dueas, P. (2007). Same genre, same discipline; however, there are differences: A
cross-cultural analysis of logical markers in academic writing. ESP Across Cultures,
4, 37-53.
Maxwell-Reid 435

Neff, J., Dafouz, E., Dez, M., Prieto, R., & Chaudron, C. (2004). Contrastive dis-
course analysis: Argumentative text in English and Spanish. In C. L. Moder &
A. Martinovic-Zic (Eds.), Discourse across languages and cultures (pp. 267-283).
Philadelphia: John Benjamins.
Newman, M., Trenchs-Parera, M., & Pujol, M. (2003). Core academic literacy princi-
ples versus culture-specific practices: A multi-case study of academic achievement.
English for Specific Purposes, 22(1), 45-71.
Pennycook, A. (2007). Global Englishes and transcultural flows. New York: Routledge.
Phillipson, R. (2003). English-only Europe? Challenging language policy. New York:
Routledge.
Reppen, R., & Grabe, W. (1993). Spanish transfer effects in the writing of elementary
school students. Lenguas Modernas, 20, 113-128.
Risager, K. (2006). Language and culture: Global flows and local complexity. Clevedon,
UK: Multilingual Matters.
St. John, M. J. (1987). Writing processes of Spanish scientists publishing in English.
English for Specific Purposes, 6, 113-120.
Snchez Escobar, . (1996). La retrica contrastiva del prrafo ingls y espaol y sus
repercusiones en la enseanza del ingls escrito. [The contrastive rhetoric of Eng-
lish and Spanish paragraphs and its impact on the teaching of written English]. In
M. Martnez Vzquez (Ed.), Gramtica contrastiva ingls-espaol [Contrastive
English-Spanish grammar] (pp. 327-350). Huelva, Spain: Universidad de Huelva.
Schleppegrell, M., & Colombi, M. C. (1997). Text organization by bilingual writers:
Clause structure as a reflection of discourse structure. Written Communication,
14(4), 481-503.
Siguan, M. (2005). The future of languages in a globalized world. In X. P. Rodrguez-
Yez, A. M. Lorenzo Surez, & F. Ramallo (Eds.), Bilingualism and education:
From the family to the school (pp. 51-61). Munich, Germany: Lincom Europa.
Singh, P., & Doherty, C. (2004). Global cultural flows and pedagogic dilemmas:
Teaching in the global university contact zone. TESOL Quarterly, 38(1), 9-42.
Whorf, B. L. (1956). Language, thought and reality: Selected writings of Benjamin Lee
Whorf (ed. John Carroll). Cambridge: MIT Press.

Bio
Corinne Maxwell-Reid teaches in the Faculty of Education at The Chinese University
of Hong Kong. Her research interests include written discourse, contrastive dis-
course, classroom language learning, and the use of systemic functional linguistics to
investigate these areas.