CLIL

 (&  
Validiv)  
Content  

Language   Motivation  

Cognition  

Figure 2: Schematic overview research intention

Before discussing which underlying cognitive processes are possibly influenced by

CLIL, it is important to clarify what CLIL is and why it could have an influence on
cognitive development. This will be done in the following section.

1.2 Defining CLIL

Although the EU heavily promotes CLIL, there seems to be no such thing as “the”
CLIL format. CLIL appears in many different forms and under many different names
(EMILE, TTO, AICLE, to name but a few). This flexibility has many advantages, as
it means that CLIL can be moulded in such a way to fit in nearly all contexts. One of
the disadvantages of this flexibility is that most definitions for CLIL are vague:

“CLIL refers to situations where subjects, or parts of subjects, are taught through a
foreign language with dual-focused aims, namely the learning of content and the
simultaneous learning of a foreign language” (Marsh, 1994, p.27)

“Content and Language integrated learning is a dual-focused educational approach

in which an additional language is used for the learning and teaching of both content
and language” (Coyle et al., 2010, p.1)

“The term Content-and-Language-Integrated-Learning (CLIL) refers to educational

settings where a language other than the students’ mother tongue is used as a medium
of instruction” (Dalton-Puffer, 2007, p.1)

As a consequence of these vague definitions and the lack of a single CLIL-format,

there exist many different forms of CLIL with different intensities. For example in
Wallonia, the French-speaking part of Belgium, up to 75% of the curriculum can be
taught in the target language. In Flanders, the Dutch-speaking part of the same
country, only 20% of the curriculum can be used for CLIL (Onderwijsdecreet XXIII,

28
2013). An important remark to be made is that foreign language classes are not
included in this percentage for Flanders, yet in Wallonia they are (Décret
d’immersion, 2007). Explicit instruction plays an important role in second language
learning processes (see deGraaf & Housen, 2009), so for scientific purposes it is
important to know if these classes are part of the curriculum or not, as they will
influence proficiency in the target language. Presence of foreign language classes will
influence CLIL classes and vice versa, so it can be argued that these foreign language
classes actually are part of CLIL.

Exposure to the target language is only one of the factors that characterise different
CLIL-types. Another important element is whether or not the content used for CLIL
has already been covered in the MSLE or not. For example in the case of the
STIMOB (“Stimulerend Meertalig Onderwijs”, the CLIL-project in Brussels, see
below), topics used for CLIL have already been covered in the regular class before
they are repeated and elaborated in the CLIL class. Repetition has many advantages in
the learning process and comparing CLIL forms that use repetition to CLIL forms that
offer new content are likely to find different effects, especially when it comes to
content acquisition.

According to Bruton (2013) there are three types of CLIL: (i) the foreign language is
learned separately, in order to learn the content through the foreign language; (ii) the
foreign language is learned through the content, which has already been learnt in the
L1; (iii) content and foreign language learning go hand in hand. Although these seem
valid distinctions (CLIL in Wallonia is an example of the first type, STIMOB of the
second case and CLIL in the Netherlands of the third), they are very vague. For
example, it is not clear in the first version if foreign language classes are still offered
once the learners start with CLIL and how CLIL forms where (part of) the learners
speak the target language as mother tongue are to be categorised. If all these
programmes are called CLIL and if no clear distinction between these different
versions is made, this will create problems on both the scientific and the policy levels.

On the scientific level, the problems are straightforward. Comparing results from
studies with learners who have up to 75% of their curriculum used for CLIL and who
have specific language classes in the target language to studies with learners who
have much less exposure to the target language and no explicit language course in the
target language is not scientifically valid. It seems that the term CLIL can be used for
the specific content course(s) that is (are) taught in an additional educational language
but also for the entire programme that includes foreign language classes. A suggestion
to make different manners of CLIL implementation more transparent is by using a
distinction between the CLIL-trajectory and the CLIL-class. The latter is the specific
CLIL-class in which a content course is taught through an additional educational
language, whereas the former is the entire package in which not only the CLIL class
but also the traditional foreign language class and in some cases also the content
course in the MSLE is incorporated. Bruton’s above-mentioned distinction of

29
different types of CLIL are then distinctions of different types of CLIL-trajectories
and form a good starting point, yet it has to be elaborated to ensure all sorts of
trajectories are included. By differentiation between a CLIL class and trajectory, it
will be possible to distinguish actual classroom practices from differences in
implementation and it will also emphasise the fact that CLIL is not just teaching a
course in another language, it is part of a bigger trajectory that also influences content
and language courses. The questions can be asked however if it is really necessary to
have so many different forms of CLIL and where the line has to be drawn between
what is CLIL and what is not.

To get a clear picture on what the effects of CLIL are, it first needs to be clarified
what CLIL is and how it has to be implemented. An important question is to what
extent CLIL is different from other approaches like content-based instruction (CBI),
immersion and task-based learning. Lasagabaster & Sierra (2009b) stated that there
were some clear differences between CLIL and immersion, yet Somers and Surmont
(2012) refuted his arguments by showing there were some similarities and differences
between the two approaches. Cenoz (2015) came to a similar conclusion when
comparing CLIL and CBI, as he found that there were no essential differences
between the two approaches and argued that research findings of both approaches had
to be combined. All language-learning programmes in which content and language
learning is combined initially had their own vision, intentions and implementation.
Through new insights in the language (learning) processes and changing
demographics the way these approaches were executed slowly changed and it seems
that the borders between these approaches have become blurred. Therefore the time is
right not only to redefine CLIL but also to investigate if there still is a distinction with
other similar programmes. Especially what happens in the classroom is important, as
contextual elements always have to be taken into account and uniformity is
unreachable. For example: European schools are privately funded and have therefore
much more funding than state schools, yet when it comes to the classroom level, it
would be very interesting to see to what extent the same pedagogies are used as in
CLIL. If elements like the school budget influence the labels of pedagogies and
approaches are called, then a distinction also has to be made between CLIL for private
schools and CLIL for public schools. It is not unrealistic to presume that private
schools and European schools have learners with a higher socio-economic status,
which is an element that always has to be taken into account when looking at learner
outcomes. A more straightforward way would be to make a distinction based on
classroom practices and implementation into the curriculum and not based on the
funding source of the school (for example European school, private school, public
school).

The vague CLIL-definitions commonly used in research also have an effect on

societal and political discussions about CLIL. Peter De Roover, for example, a
member of the Flemish Nationalist Party (NVA), stated that a CLIL class is basically
a language class taught by a motivated teacher who includes a specific content area

30
(Volt, 1/09/2011). He also stated that he was an expert in CLIL as he teaches
economics in Dutch to pupils who do not have Dutch as their mother tongue (Volt,
1/09/2011). The current definitions cannot counter these statements, although it can
be said with absolute certainty that CLIL is not intended to be a submersive approach,
(i.e. using a different educational language without any support for the learners, which
is the opposite of an immersive approach) which is exactly what De Roover is using
the term for. As long as definitions are not clear on what is meant by CLIL they may
be easily misinterpreted by policy makers, which might lead to “misimplementations”
of CLIL.

1.2.1 What are the issues with the current definitions?

One of the first issues in the definitions is that the target language of CLIL is
considered to be a “foreign language” (Marsh, 1994), an “additional language” (Coyle
et al., 2010) or a “language other than the mother tongue” (Dalton-Puffer, 2007). It is
unclear to whom this target language has to be foreign, additional or different from
the mother tongue: is it for the school, the teachers or the learners? If it has to be
foreign, additional or other than the mother tongue for the learners, then it
immediately raises a number of new questions, three of which are mentioned below.

First, does this imply that learners who are now in traditional education, but who do
not speak the mainstream educational language as mother tongue are actually already
enrolled in a type of CLIL? In Dutch-speaking schools in Brussels, for instance, 60.93
% of all the learners do not have Dutch as (one of) their mother tongues3. Does this
mean that they already attend some form of CLIL? Most likely not, as will be
explained below CLIL is more than just teaching through another language, yet the
current definition is inconclusive about this.

Second, does this imply that pupils who speak the CLIL target language as mother
tongue (as for example in the Basque country in Spain, or in Brussels, Belgium)
cannot enrol in CLIL programmes? Again this is not the case, because there is no
reason why these learners could not benefit from CLIL. On the contrary, they might
even benefit the most from CLIL, as they will then also receive instruction in their
mother tongue, which has been proven to increase academic achievement (Ramírez,
1992; Alanís, 2000).

Third, what happens with the inflow of speakers of other languages who do not speak
either the traditional educational language or the CLIL target language as mother
tongue, as is often the case with migrant children? Should this group be allowed to
follow CLIL? Here no specific research has been done, so no definite statements can
be made. Yet there seems no reason to exclude this group, as the pedagogies of CLIL

3
see https://www.vlaamsparlement.be/parlementaire-documenten/schriftelijke-
vragen/942851 (last visited 27 May 2015)

31
(see later) stimulate the natural learning process, where it does not matter which
mother tongue the learner speaks.

All these questions show that the current terminology of “foreign language” and
“additional language” are again proof of a very restrictive monolingual perspective.
The new definition should take into account that there is a large heterogeneity in the
linguistic background of learners all over Europe and that terms like foreign language
and mother tongues are not always applicable to the school population. Therefore,
descriptions like “mainstream language of education” (MSLE), “additional language
of education” or “target language” are recommended to avoid confusion and ensure
clarity.

When it comes to which language should be the target language in CLIL it seems that
in most cases, schools or governments have chosen for English. However, there are
also many examples of other languages being used, as for example in Finland both
Swedish and Finnish is used (and in certain regions even an indigenous language) and
in Belgium the majority of the schools in Wallonia opt for Dutch as target language.
Often CLIL is seen as a form of English as a Foreign Language (EFL) instruction, but
this is incorrect and restricting CLIL to something that it is not. It is logical that
schools often prefer a high status language as target language, yet theoretically it is
perfectly possible to also use for example a migrant language as target language.
CLIL is an approach and therefore not bound to one specific language (or language
status). In the new definition it should then also be clear that the target language is not
necessarily English.

A second issue is related to the dual-focus of CLIL. It is clear that one of the aims of
CLIL is that pupils learn subject matter through a language other than the traditional
educational language. The dual-focus of CLIL entails that a teacher has both content
and language goals for each class and therefore theoretically functions both as a
content and language teacher. The question has to be raised to what extent the focus is
actually on language learning in the CLIL class. Coyle (2006) states that in a CLIL
class the teacher works on a continuum on which language and content are located at
each end and that depending on the activity the focus is more on either language of
content. Every teacher uses language to convey content, so every teacher has to take
into account the language aspect and can therefore be seen as a language teacher, no
matter if it is a CLIL class or not. Especially when they have to teach learners with
multilingual backgrounds and/or when (some of) the pupils do not speak the language
of education as their mother tongue. The language-content continuum is unique to
CLIL but a characteristic of a good teacher in general. When the CLIL-trajectory also
includes foreign language classes, it seems redundant that much explicit instruction
regarding language will be given in the CLIL class. This of course does not mean that
corrective feedback cannot explicitly focus on for example a certain grammar rule,
but is only when the situation requires it and cannot be the goal of the class. It is

32
therefore better to state that a CLIL trajectory is dual-focussed, as in the CLIL class
language is subordinate to content.

Again this emphasises the importance of the foreign language class. One of Bruton’s
(2013) types of CLIL is the one where there is no formal foreign language class. This
seems inadvisable as during the specific foreign language class the teacher can
explicitly go into certain aspects of the language, which cannot be highlighted
(enough) during the CLIL-class. The inclusion of implicit and explicit instruction is a
very effective way of learning (see earlier) and a combination of CLIL and explicit
language instruction will maximise and optimise learner outcomes.

Stating that language is subordinate to content in a CLIL class also has its
implications for evaluation. In a CLIL class, it is not recommended to penalise
language mistakes, as the aim is to learn to use the language. This lowers the
threshold for language usage and therefore also the stress; Dalton–Puffer (2007)
points out that in CLIL anxiety levels are indeed lower. Schools still have to measure
progress and as CLIL aims to increase content and language proficiency, the question
can be asked how language should then be evaluated. The problem in the evaluation
question is that it focuses on the CLIL class whereas a better view on the progress of
the learner would be if the focus would be on the entire CLIL trajectory. A good
collaboration between the foreign language teacher and the CLIL teacher is advisable
here when it comes to evaluating language proficiency, just as a good collaboration
between the content teacher and the CLIL teacher is advisable when focusing on
content. A related question to this is to what extent learners need to be graded and if it
would not be better to use for example a personal portfolio to follow their progress.

Thirdly the question has to be asked what is meant with “subjects or part of subjects”.
In a foreign language class content is also offered and one can ask if this is then can
also be classified as CLIL. For example when the French Revolution is discussed in a
French as a foreign language class, the subject is taught through the foreign language
and the aims are both content and language acquisition. Following the current CLIL
definitions, it becomes hard to argue that this is not CLIL. But this is not what Europe
had in mind when favouring CLIL as a teaching approach. Considering CLIL is “the
European answer to immersion” (Dalton-Puffer, 2007), it can be stated that the goal
of CLIL is that a content course like geography or science is taught through another
language than the normal language of education. Using the term “subject” is not
specific enough and it would be better to state subjects of which the goals can be
reached through any language. In for example mathematics the aims are to learn how
to do multiplications or fractions, which are competences than can be taught through
any language. In French not all goals can be reached if the French language is not
used.

Bruton (2013) states that the range of CLIL subjects seem to be very arbitrary. This is
true, as in many cases schools select the courses based on what the capacities of their

33
teachers are (Gierlinger, 2007). It is unclear however why this would be a problem.
There has been no research study, to my knowledge, on which course should be used
for CLIL classes. Lo (2014) did find out that in humanities courses learners talked
significantly more and longer than in science classes, giving an indication that the
choice of the course could have an influence on oral proficiency. Although there is no
reason to assume that certain courses are less fit for CLIL than others future research
should look into this. It could be plausible that for example scientific courses like
mathematics and sciences are good courses to start with, as they have the language of
science as an extra support for the uptake of the target language and that in later
stages social sciences like history could be used as here more aspects of language are
used and needed. Insight in which course to use when could help policy makers to
write guidelines for CLIL implementation.

Surprisingly, none of these definitions mentions the multilingual aspect. One of the
reasons Europe has become an advocate of this approach is that it would increase
multilingualism, so the absence of this word “multilingual” in the definitions is
remarkable. By including this word, it would become impossible to consider pupils
who do not speak the educational language of the school as taking part in a “100%
immersion programme” and therefore stating that these pupils already follow CLIL
(Dhaene & Daniels, 2015). The inclusion of the term multilingual in the new
definition would refute the consideration that total immersion programmes can be
considered CLIL, as it would then not be multilingual. One can even question if total
immersion programmes can still be considered immersion and why they are not
submersion, but this is a different discussion and not the aim of this dissertation. It
also has to be pointed out that CLIL aims for additive bilingualism, meaning that the
acquisition of a new language is not at the expense of the previously acquired
language(s) (Van de Craen et al., 2011)

A final point of critique on the current definitions of CLIL is that they do not mention
its most important asset, namely its pedagogies (Coyle, 2007; deGraaf et al., 2007).
There are three main aspects of CLIL pedagogies. The first one is the meaningful
environment in which previous knowledge is activated. The second one is that content
is learned through interaction and that the learner plays an active role in the
“discovery” of the content. The final aspect is the one where language support is
offered through scaffolding.

To create a meaningful environment that builds on previous knowledge,

translanguaging is used. Translanguaging is a pedagogic approach in which the entire
linguistic knowledge of the learner is seen as one single resource (García, 2009). This
moves away from the traditional “one language per classroom” principle, where usage
of other languages is not allowed. In a CLIL classroom usage of mother tongues and
other known languages is allowed and in some cases even recommended. This will
allow learners to bridge gaps in their knowledge and also overcome terminology
issues. The idea behind this is that through translanguaging the learner can build on

34
previously acquired knowledge and increase their insights in both language and
content. As mentioned in the introduction on cognitive development, learning means
building on previous experiences and knowledge. This is exactly what
translanguaging does, namely using previous (language) knowledge to create new
knowledge and insights. Usage of translanguaging in the classroom therefore
stimulates the natural learning process by keeping the anxiety levels of the pupils as
low as possible.

Activating methods are didactics that force learners to participate in the creation of
knowledge instead of just listening to what the teacher has to say (Dufresne et al.,
1996). By giving the learners an active role in their learning process -instead of just
letting them process the given input- they not only have to listen to the target
language but they are also forced to use it4. Of course it is important that the teacher
supports them in this, and this is best done through scaffolding and translanguaging.
Scaffolding is a teaching method that requires the teachers to support the learner in
bridging the gap between what is already known and mastered and what is yet
unknown and not yet mastered. This gap is what Vygotsky (1978) called the “zone of
proximal development” (see further). There are three different types of scaffolding
namely verbal scaffolding, content scaffolding and learning process scaffolding
(Echevarrìa et al., 2010). The first one means that the (CLIL) teacher adapts his
language to the level of the learner in order to ensure that communication can take
place. Content scaffolding means that the teacher is constantly using techniques (such
as discussions) that assist and support the learner in their understanding of and
engagement with the content (Massler et al., 2011). Learning process scaffolding are
techniques (such as teaching to each other) used by teacher to support learners
working processes but also their learning processes (for an elaborate discussion on
how this is translated itself to the classroom, see Massler et al., 2011). Research has
shown that CLIL classes provide more opportunities for learners to use discourse
pragmatic strategies as they often use the foreign language for more diverse functions
and for more complex meaning negotiations than their peers in language lessons
(Nikula, 2005; 2008)

Combining these didactics creates a meaningful learning environment that is not

teacher-centred (or as Freire (1974) calls it “the banking model”) but pupil-centred,
where learning is achieved through activating methods, scaffolding and
translanguaging. Important to note is that these pedagogies not only influence the
acquisition of language, but also the uptake of information. By forcing the learners to
actively take part in the learning process, higher order thinking processes are
stimulated as well as an increased insight in conceptual and procedural knowledge. In

4
For some examples of these didactics, please visit
http://www.steunpuntgok.be/secundair_onderwijs/materiaal/lesmateriaal/bronnenboek
/interactieve_werkvormen.aspx

35
Anderson and Krathwohl’s revision of Bloom’s taxonomy (2001, see later) of
thinking processes it becomes clear how thinking processes and knowledge
construction are related and how the interactive methods of CLIL stimulate both
content and language acquisition.

Table 3: Bloom’s taxonomy of thinking processes in Anderson and Krathwohl’s revision

(Anderson & Krathwolh 2001, p. 67-68)

To conclude, CLIL is an interactive teaching approach that creates a meaningful

environment in which the leaner actively has to participate in the creation of
knowledge on both content and language. Through didactical techniques like
scaffolding and translanguaging, both content and language learning are supported

36
which maximises the learning effect. The current definitions do not represent these
elements and lack clarity on aspects concerning language usage, the dual-focus of
CLIL and the choice of subject. A new definition is therefore highly recommended

1.1.2 The new definition

A new definition for CLIL has to fulfil a number of requirements that previous
definitions were unable to account for: (i) it has to ensure that it is applicable to
different learners in various contexts, (ii) it has to ensure that additive bilingualism is
envisaged and (iii) it has to include the word multilingual. Secondly, it also has to
take into account some specific CLIL characteristics such as that (i) all languages can
be used for CLIL; (ii) it is implemented in the curriculum in combination with a
foreign language class; (iii) it is considered a content course; (iv) it is an interactive
teaching approach that activates previous knowledge, both with respect to content and
language.

The suggestion for a new definition is as follows:

CLIL is an interactive teaching approach in which an additional language of

education is used for subjects whose goals are not related to the use of one specific
language only. It aims to stimulate the development of additive multilingualism in
learners without increasing the number of hours in the curriculum or without
replacing the explicit (foreign) language classes.

This definition takes into account the seven requirements, but may need some
explanation when it comes to the description of the course. CLIL is intended for
content courses like history or mathematics, yet the previous definitions gave the
impression that when language teachers taught cultural information in their target
language they were also using CLIL. To avoid this kind of confusion, it is clearly
stated that the aims of the course used for CLIL are not language specific. In a history
class the aim is for example that the learners know all about the world wars or the
Roman Empire and it does not really matter if they know this in French, German or
Spanish. However, the aims of a French class can only all be met when the course is
taught in French. The elaborate description is therefore to ensure that these really are
content courses that are used for CLIL and not language courses, because the explicit
language course still has to be present (as discussed above).

Although this new definition of CLIL has the potential to create more clarity into
what the approach entails, leading to advantages on both academic and policy making
level, it will not lead to one type of CLIL. Intensity of implementation will still vary
according to the region, meaning that in certain regions a “light” version of CLIL will
be implemented (as in for example Flanders), whereas in other areas an “intensive”
version will be implemented (as in for example Wallonia) and other regions will opt

37
for a more “balanced” version (as in for example Spain). This therefore still has to be
taken into account when performing research and it cannot be stressed enough that
when comparing CLIL results, not only the CLIL class has to be investigated and
discussed but also the CLIL trajectory has to be taken into account.

This is of course undeniable, but context specific elements do not imply that there
cannot be a clear definition and pedagogy on a specific approach. The context may be
specific, yet learning processes and how they are best stimulated are universal. The
possible value and impact of the new definition is therefore high.

1.3 Speakers of other languages and education

The increase of speakers of other languages in educational settings has been going on
for a number of years. This has challenged the educational systems of many western
nations, as there often is an achievement gap between the native population and some
immigrant groups (Crul & Schneider, 2009). Flanders is one of the most extreme
cases as results indicate that the socio-economic inequality and the achievement gap
between immigrants and native speakers in Flanders is one of the highest among
Western countries (OECD, 2006). Educational policy makers and teachers link this
result with linguistic deficiencies, even though hardly any scientific evidence can be
found for this assumption (Blommaert & Van Avermaet, 2008; Agirdag, Van
Avermaet & Van Houtte, 2013). There seem to be two main reasons for this
deficiency.

1.3.1 Terminology that stigmatises minority groups: “bad bilingualism”

Although European policies are very much in favour of bilingualism (or
multilingualism), it seems that this attitude is not always reflected in educational
settings. Often there is a discrepancy between so-called “good bilingualism” and “bad
bilingualism”. The difference between the two types is related to the languages
known: if these languages are considered “important” or “influential” then the
bilingualism is considered to be of the good type. However, if one of the languages
known is a migrant language then the bilingualism is often considered to be bad and
detrimental for academic development (Agirdag, 2010, Agirdag et al., 2013).
Bourdieu (1977) states that somebody’s linguistic capital, meaning knowledge of
certain languages, is only worth the value that the social context of that person
attributes to those languages. Agirdag et al (2014) underline this statement by
showing how in Flanders a distinction is being made between good and bad
bilingualism by calling speakers of more than one high status language “bilinguals”
whereas speakers who speak one migrant language are called “anderstalig” (i.e.
linguistically different).

38