EA J V N
Revisiting the role of metalanguage in L2 teaching and
Nanyang Technological University, Singapore
One topic that is sidelined in the professional literature on second language
(L2) teaching and learning is the use of metalanguage in the classroom. In
the past three decades, explicit and formal instruction in L2 grammar has
fallen from its centrality in traditional pedagogical approaches and been
relegated to a peripheral position in many classrooms, due to the joint
inﬂuences of some popular theoretical claims, ﬁndings from early empirical
studies about the disassociation between learners’ explicit knowledge of L2
target structures and their ability to use these structures, and communicative
language teaching, which, in its application, sometimes sets great store
by the development of communicative competence and ﬂuency rather
than grammatical competence. Because of its time-honored association
with formal grammar instruction, metalanguage has been downplayed
or even rejected as a legitimate component of pedagogical practices in
many L2 classrooms. is paper discusses recent empirical research on
the relationships between L2 proﬁciency, metalinguistic awareness, and
metalinguistic knowledge. It also considers several potential advantages
that understanding and using metalanguage oﬀer to L2 teaching and
learning. Based on the discussion, it argues that metalanguage deserves a
place in L2 classrooms.
e use of metalanguage – namely, technical or semi-technical terminology employed
to analyse or describe language (Crystal, 1997; James & Garrett, 1992) – is a
pedagogical topic that is rarely discussed today in the professional literature on L2
teaching and learning (Borg, 1999). is general lack of attention has resulted, in
large measure, from the unfortunate, entrenched linkage of metalanguage with explicit
and formal instruction in L2 grammar usage (Berry, 1997; Eisenstein, 1987; Francis,
1994). Over the decades, instruction in L2 grammar has fallen from its centrality in
traditional pedagogical approaches (e.g. the Grammar-Translation Method) and been
relegated to a less important or insigniﬁcant position in many classrooms (Elder &
Manwaring, 2004). Several sources of inﬂuence have contributed to this sidelining
of formal grammar instruction.
EA J V N
One major source of inﬂuence was the widely publicised position championed by
some early second language theories (see Krashen, 1981, 1985; Paradis, 1994; Zobl,
1992) that explicit knowledge learned as a result of formal grammar instruction would
not contribute to language acquisition or underlie spontaneous language use. Such a
position provided a theoretical basis for the rejection of explicit and formal teaching
of L2 grammar (Carter, 1995). A second important source of inﬂuence consisted in
some empirical studies (e.g. Alderson, Clapham, & Steel, 1997; Grigg, 1986; Seliger,
1979; Steel & Alderson, 1994) suggesting that L2 learners’ metalinguistic knowledge
was not related to their L2 proﬁciency or actual use of the target language. ese studies
provided some empirical evidence that lent support to a questioning of the usefulness
of explicit, formal grammar instruction. A third major source of inﬂuence was the rise
and spread of communicative language teaching (CLT) (Hu, 2002a; Savignon, 2005).
While there are diﬀerent versions of CLT, they all set great store on the development
of communicative competence (i.e. the ability to use the target language to engage in
meaningful and eﬀective communication) rather than just grammatical competence
(Richards & Rodgers, 2001). Although not all CLT practitioners (especially those
adopting a weak version of CLT) are opposed to explicit and systematic teaching of
L2 grammar in lessons (see Batstone, 1994a, 1994b; Scheﬄer & Cinciała, in press),
many CLT-oriented classrooms (in particular those implementing a strong version of
CLT) ‘downplay the importance of explicit grammar instruction’ (Elder & Manwaring,
2004, p. 145; Carter, 2003). us, the CLT movement has provided a pedagogical
impetus that has contributed to a growing distrust in and an increasing marginalisation
of formal grammar instruction in many L2 classrooms.
Because of its long association with explicit and formal grammar instruction,
metalanguage in particular has also been marginalised or even rejected as a legitimate
component of pedagogical practices in many L2 classrooms (Berry, 2009). Alderson
(1997), for example, questions ‘the assumption that teachers need to have metalanguage’
(p.2) and declares ‘I have long suspected that this is why teachers use metalanguage
in class, to emphasise their position of knowledge and authority, to reinforce their
power’ (p.16). Garrett (1986) claims that the use of metalanguage constitutes a
major problem with formal grammar instruction because ‘it cannot of itself invoke
understanding of the processing which leads to the production of a structure’ (p.141).
In a similar vein, Mohammed (1996) asserts that grammar instruction based on
linguistic terms and concepts can hardly achieve the goal of adding to or modifying
the rules discovered by learners themselves ‘through the natural process of hypotheses
formation and testing’ because ‘such terms and concepts constitute an additional
learning burden and remain as a separate body of knowledge that has nothing to do
with the way people actually process language’ (p.283). ‘Another problem in using
EA J V N
[linguistic] terms’, Mohammed claims, ‘is that the learner may focus on these terms
and learn them by heart either because he thinks that these terms are what the teacher
or textbook writer wants him to know or because he believes that learning a language
is a matter of learning such terms’ (p. 287).
Metalanguage, metalinguistic knowledge, and L2 Proﬁciency
Criticisms of the use of metalanguage in the classroom, such as those mentioned above,
are misguided because they fail to recognise the part that metalanguage can play in
facilitating the development of metalinguistic knowledge, namely analysed, often
verbalisable, knowledge about the L2 (Hu, 2002b). In addition, in contrast to the early
empirical studies cited above, several recent studies have found substantial positive
correlations between knowledge of metalanguage and L2 proﬁciency. In a study
involving 372 ﬁrst-year undergraduate students in Hong Kong, Berry (1997) found
that the participants diﬀered widely in their knowledge of 50 items of metalanguage
and that this knowledge was signiﬁcantly correlated with their A-level grades for
English, suggesting ‘a connection between knowledge of [linguistic] terminology and
proﬁciency in English’ (p.140) as measured by that instrument. e strong correlation
between metalinguistic knowledge and L2 proﬁciency was replicated in another study
by Berry (2009) which involved ﬁrst-year English majors just starting their university
study in Poland, Austria, and Hong Kong. For the Hong Kong sub-sample (i.e.
the only group for whom English proﬁciency scores were available), students who
scored higher in their A-level English examination also scored signiﬁcantly higher
on the test of metalinguistic knowledge. Similar results were also obtained by Elder
and Manwaring (2004) in their study of the relationship between L2 grammatical
knowledge and assessment results in a sample of undergraduate students of Chinese
at an Australian university. Consistently strong positive correlations were found
between the second-year students’ knowledge of grammatical terms and their semester
achievement scores for speaking, listening, reading and writing in Chinese
It is true that the number of empirical studies reporting such positive relationships
is still quite small. However, there are a reasonably large number of empirical studies
that have found strong correlations between learners’ metalinguistic knowledge of
various L2 structures and their L2 proﬁciency. In the Elder and Manwaring study,
robust and strong correlations were found between various measures of metalinguistic
knowledge of Chinese and end-of-semester Chinese achievement scores. Similarly,
Han and Ellis (1998) obtained substantial correlations between 48 adult English-
L2 learners’ scores on a measure of metalinguistic knowledge and their scores for
two international English proﬁciency tests, TOEFL and Secondary Level English
Proﬁciency Test. In a study involving university, advanced-level, French L2 learners,
Renou (2001) also found consistently signiﬁcant correlations between metalanguistic
EA J V N
awareness and L2 proﬁciency in a sub-sample of learners coming from a ‘grammar
approach to learning French as a second or foreign language’ (p.253). Notably, such
relationships were missing for the sub-sample coming from a communicative approach
to learning the target language. Even stronger correlations between various measures
of metalinguistic knowledge and L2 proﬁciency tests were obtained in Ellis (2006)
and Roehr (2007). Such empirical ﬁndings suggest that knowledge of metalanguage
may have an indirect inﬂuence on, in the words of Garrett, ‘the processing which
leads to the production of a structure’ through its relationship with metalinguistic
knowledge. Indeed, several recent studies (Elder & Manwaring, 2004; Ellis, 2005; Hu,
1999) have found strong and positive correlations between L2 learners’ metalanguistic
knowledge and their receptive or productive knowledge of metalanguage. For example,
both Hu (1999) and Elder and Manwaring (2004) obtained correlation coeﬃcients
in the high range of .70 - .72. Such ﬁndings support Ellis’s (2004) observation that
‘it is possible that an increase in the depth of explicit knowledge will occur hand in
hand with the acquisition of more metalanguage, if only because access to linguistic
labels may help sharpen understanding of linguistic constructs’ (p.240).
To the extent that explicit discussion of and deliberate reﬂection on linguistic patterns
and properties are helpful for developing an essential knowledge of the underlying
regularities and relationships in the target linguistic system (Renou, 2001; Swain, 2005;
Swain & Lapkin, 1995; Storch, 2008), a judicious use of metalanguage can get the
job done in a more eﬃcient manner, because linguistic terms are essentially succinct
ways of categorising patterns and relationships found in a language. To understand
and learn these terms can be a useful step to understanding and learning the patterns
and relationships that they label. As pointed out by Ellis (2004), metalanguage ‘may
assist learners in developing explicit knowledge that has greater precision and accuracy’
(p. 261). A case can be made that a judicious use of metalanguage can beneﬁt L2
learners who are cognitively mature and developmentally ready.
Potential advantages of using metalanguage
Metalanguage oﬀers several advantages that can be exploited in the L2 classroom
(for arguments in favor of metalanguage see Berry, 2005; Borg, 1999). e ﬁve
most important ones will be discussed brieﬂy below. To begin with, as the ﬁndings
from the research reviewed in the preceding section suggest, knowledge and use of
metalanguage have the potential to facilitate the development of an L2 learner’s
metalinguistic awareness – that is, ‘an enhanced consciousness of and sensitivity to
the forms and functions of language’ (Carter, 2003, p.64); which can in turn enhance
language development (Berry, 2005; Swain, 2005). By its very nature, metalanguage
entails a reﬂexive focus on language (Berry, 2005). Reﬂecting upon language is an
important means whereby language learners develop their understandings of how
EA J V N
forms and functions are mapped in the target language (Swain & Lapkin, 1995; Storch,
2008) and how language ‘interacts with culture and ideology’ (Carter, 1995, p.1).
Furthermore, as argued by Ellis (2004), access to metalanguage can sharpen a learner’s
understanding of the structure of the target language and lead to verbalisable, analysed
knowledge, which according to Bialystok (1990) and Schmidt (1990) constitutes the
highest level of consciousness of language.
Second, in societies where analytical study and metalanguage feature strongly in L1
literacy instruction, the use of metalanguage in the L2 classroom is a useful way to
tap the wealth of metalinguistic awareness that learners have developed in the process
of acquiring L1 literacy. In mainland China for example, a typical Chinese child is
brought into formal contact with metalanguage through mother-tongue instruction at
primary school. By the time the child reaches secondary school, an essential knowledge
of metalanguage is already in place. Consequently, the secondary school teacher of
English is in a good position to exploit the child’s metalinguistic knowledge to assist
him or her in making sense of the L2 system. Basic linguistic terms (e.g., noun,
verb, adjective, subject, and object) which label concepts and relationships common
to Chinese and English can be used to help the child see how strings of sounds or
words are organised and patterned in the L2. is point is clearly borne out by Elder
and Manwaring’s (2004) study, which found that on an L2 metalanguage assessment
task learners of Chinese performed best on grammatical features which also existed
in their L1.
A third advantage of using metalanguage in grammar instruction can be appreciated
only in view of the disadvantages and diﬃculties that arise in classrooms where
teachers and learners refrain from using such terms. An example suﬃces to illustrate
this point. Even in a CLT-oriented classroom, it is sometimes desirable to have an
explicit discussion of the structural and functional features of complex structures
(Carter, 1995; Rutherford, 1987). Such a discussion can be used either to raise learners’
consciousness about the target structures or to provide an opportunity for them to
conﬁrm or modify the rules they have internalised as a result of their own hypothesis
formation and testing. However, it is diﬃcult to see how an explicit discussion of
complex structures, such as unreal conditionals and relative clauses, can be conducted
without recourse to metalanguage. As Ellis (2006) notes, unreal conditionals ‘cannot be
easily explained without reference to … terms like ‘main clause,’ ‘subordinate clause,’
‘past perfect,’ ‘modal verb,’ ‘past participle’’ (p.457). at is, a concise and economic
vocabulary about language (i.e. metalanguage) is often needed to talk about language
explicitly or, in the words of Crystal (1985, p.22), ‘to take the pieces of language apart
and put them back together again’.
Still another potential advantage of using metalanguage in teaching grammar concerns
EA J V N
the explanatory precision with which linguistic generalisation can be made and the
eﬃcient delimitation of the contexts to which the generalisation applies. Metalanguage
that is appropriately used can preempt both under- and over-generalisation of the
rules in question. Take, for example, the use of the simple present in temporal and
conditional clauses for future reference. If a teacher wishes to draw the learners’
attention to this particular use of the simple present but refrains from using any
metalanguage, he or she will have to provide them, in one way or another, with a
whole range of exemplars covering clauses introduced by subordinators such as after,
as, before, once, till, until, when(ever), as soon as, if, unless, as long as, etc. to prevent
potential under-generalisation of the regularity. Even if the teacher succeeds in making
the learners know that the simple present can be used in these clauses, there is no easy
way for him or her to preempt over-generalisation. It is probable that some learners
may be induced to believe that the use can be generalised to other types of adverbial
clause. at is, there is always a danger that a wrong generalisation is inferred from
a set of examples which illustrate where a rule applies but not where it does not
apply. ese problems can be greatly alleviated by appealing to linguistic terms such
as temporal clause and conditional clause, provided that the learners know what these
terms refer to.
Finally, metalanguage can be used proﬁtably by teachers to help their learners link
up newly encountered structures with knowledge of the target system that has
already been acquired. In other words, metalanguage and concepts of L2 structural
properties already learned can be exploited as points of reference or anchoring sites
for the assimilation of new knowledge. is is most obvious when teachers compare
and/or contrast new grammatical features (e.g. given, assuming, and presuming) with
previously learned features (e.g. if and unless), and make clear the connections between
them through linguistic terms (e.g. subordinating conjunctions and conditional clauses)
that learners have learned and can understand. One may argue that some ingenuity
can enable a teacher to get the job done without reference to metalanguage. Such
ingenuity surely could be enlisted for more worthwhile pedagogical practices in the
L2 classroom than merely to circumvent the use of metalanguage.
e discussion above points to a need to reconsider the role of metalanguage in
L2 teaching and learning in the light of current research on the inter-relationships
between metalanguistic knowledge, metalinguistic awareness, and L2 proﬁciency.
e empirically identiﬁed positive correlations among the three, as well as the many
potential advantages that use of metalanguage can oﬀer, lead to the conclusion that
metalanguage deserves a place in L2 teaching. A case can be made that, unless it is
EA J V N
taught purely for its own sake, metalanguage can be a valuable means of facilitating
L2 learning rather than ‘an additional learning burden’ that ‘remain[s] a separate
body of knowledge’ (Mohammed, 1996, p.283). is reappraisal of the role of
metalanguage, however, does not resolve all pedagogical issues concerning the use
of metalanguage. While a return to the boring and sterile pedagogical practices of
the traditional approaches is out of the question, many issues (e.g. when and how to
use metalanguage) remain to be explored. In particular, there is a need for empirical
research, along the direction taken by such studies as Basturkmen, Loewen, and
Ellis (2002) and Storch (2008), that seeks to explore how metalinguistic awareness
and metalanguage can be most fruitfully integrated into a meaning-focused,
communication-oriented L2 pedagogy.
Alderson, J. C. (1997). Models of language? Whose? What for? What use? In A. Ryan
& A. Wray (Eds.), Evolving models of language (pp. 1-22). Clevedon, UK: British
Association for Applied Linguistics & Multilingual Matters.
Alderson, J. C., Clapham, C., & Steel, D. (1997). Metalinguistic knowledge, language
aptitude and language proﬁciency. Language Teaching Research, 1, 93-121.
Basturkmen, H., Loewen, S., & Ellis, R. (2002). Metalanguage in focus on form in the
communicative classroom. Language Awareness, 11, 1-13.
Batstone, R. (1994a). Grammar. Oxford, UK: Oxford University Press.
Batstone, R. (1994b). Product and process: Grammar in the second language
classroom. In M. Bygate, A. Tonkyn, & E. Williams (Eds.), Grammar and the language
teacher (pp. 224-236). Hemel Hempstead, UK: Prentice Hall.
Berry, R. (1997). Teachers’ awareness of learners’ knowledge: e case of metalinguistic
terminology. Language Awareness, 6, 136-146.
Berry, R. (2005). Making the most of metalanguage. Language Awareness, 14, 3-20.
Berry, R. (2009). EFL majors’ knowledge of metalinguistic terminology: A comparative
study. Language Awareness, 18, 113-128.
Bialystok, E. (1990). Communication strategies: A psychological analysis of second language
use. Oxford, UK: Basil Blackwell.
Borg, S. (1999). e use of grammatical terminology in the second language classroom:
A qualitative study of teachers’ practices and cognitions. Applied Linguistics, 20, 95-126.
EA J V N
Carter, R. (1995). How aware should language aware teachers and learners be? In D.
Nunan, R. Berry, & V. Berry (Eds.), Language awareness in language education (pp.
1-15). Hong Kong: University of Hong Kong.
Carter, R. (2003). Language awareness. ELT Journal, 57, 64-65.
Crystal, D. (1985). e past, present and future of English parsing. English Today,
Crystal, D. (1997). e Cambridge encyclopedia of language (2nd ed.). Cambridge, UK:
Cambridge University Press.
Eisenstein, M. (1987). Grammatical explanations in ESL: Teach the student, not the
method. In M. Long & J. Richards (Eds.), Methodology in TESOL (pp. 282-297).
New York: Newbury House.
Elder, C., & Manwaring, D. (2004). e relationship between metalinguistic
knowledge and learning outcomes among undergraduate students of Chinese.
Language Awareness, 1, 145–162.
Ellis, R. (2004). e deﬁnition and measurement of L2 explicit knowledge. Language
Learning, 54, 227-275.
Ellis, R. (2005). Measuring implicit and explicit knowledge of a second language: A
psychometric study. Studies in Second Language Acquisition, 27, 141-172.
Ellis, R. (2006). Modelling learning diﬃculty and second language proﬁciency: e
diﬀerential contributions of implicit and explicit knowledge. Applied Linguistics, 27,
Francis, G. (1994). Grammar teaching in schools: What should teachers be aware of?
Language Awareness, 3, 221-236.
Garrett, N. (1986). e problem with grammar: What kind can the language learner
use? e Modern Language Journal, 70, 133-148.
Grigg, T. (1986). The effects of task, time and rule knowledge on grammar
performance for three English structures. University of Hawaii Working Papers in
ESL, 5(1), 37-60.
Han, Y. J., & Ellis, R. ( 1998). Implicit knowledge, explicit knowledge and general
language proﬁciency. Language Teaching Research, 2, 1–23.
Hu, G. W. (1999). Explicit metalinguistic knowledge at work: e case of spontaneous
written production by formal adult Chinese learners of English. Unpublished doctoral
dissertation, Nanyang Technological University, Republic of Singapore.
EA J V N
Hu, G. W. (2002a). Potential cultural resistance to pedagogical imports: e case
of communicative language teaching in China. Language, Culture and Curriculum,
Hu, G. W. (2002b). Psychological constraints on the utility of metalinguistic
knowledge in second language production. Studies in Second Language Acquisition,
James, C., & Garrett, P. (1992). e scope of language awareness. In C. James & P.
Garrett (Eds.), Language awareness in the classroom (pp. 3–20). London: Longman.
Krashen, S. (1981). Second language acquisition and second language learning. Oxford,
Krashen, S. (1985). The input hypothesis: Issues and implications. London:
Mohammed, A. M. (1996). Informal pedagogical grammar. International Review of
Applied Linguistics in Language Teaching, 34, 283-291.
Paradis, M. (1994). Neurolinguistic aspects of implicit and explicit memory:
Implications for bilingualism and SLA. In N. C. Ellis (Ed.), Implicit and explicit
learning of languages (pp. 393-419). London: Academic Press.
Renou, J. (2001). An examination of the relationship between metalinguistic awareness
and second-language proﬁciency of adult learners of French. Language Awareness, 10,
Richards, J. C., & Rodgers, T. S. (2001). Approaches and methods in language teaching
(2nd ed.). Cambridge, UK: Cambridge University Press.
Roehr, K. (2007). Metalinguistic knowledge and language ability in university-level
L2 learners. Applied Linguistics, 29, 173-190.
Rutherford, W. (1987). Second language grammar: Learning and teaching. London:
Savignon, S. J. (2005). Communicative language teaching: Strategies and goals. In
E. Hinkel (Ed.), Handbook of research in second language teaching and learning (pp.
635-651). Mahwah, NJ: Lawrence Erlbaum.
Scheﬄer, P., & Cinciała, M. (in press). Explicit grammar rules and L2 acquisition.
ELT Journal. doi:10.1093/elt/ccq019
Schmidt, R. (1990). e role of consciousness in second language learning. Applied
Linguistics, 11, 129-158.
EA J V N
Seliger, H. W. (1979). On the nature and function of language rules in language
teaching. TESOL Quarterly, 13, 359-369.
Steel, D., & Alderson, J. C. (1994). Metalinguistic knowledge, language aptitude
and language proﬁciency. In D. Graddol & S. omas (Eds.), Language in a changing
Europe (pp. 92-103). Clevedon, UK: Multilingual Matters.
Storch, N. (2008). Metatalk in a pair work activity: Level of engagement and
implications for language development. Language Awareness, 17, 95-114.
Swain, M. (2005). e output hypothesis: eory and research. In E. Hinkel (Ed.),
Handbook of research in second language teaching and learning (pp. 471-483). Mahwah,
NJ: Lawrence Erlbaum.
Swain, M., & Lapkin, S. (1995). Problems in output and the cognitive processes
they generate: A step towards second language learning. Applied Linguistics, 16, 371-
Zobl, H. (1992). Sources of linguistic knowledge and uniformity of nonnative
performance. Studies in Second Language Acquisition, 14, 387-402.
Guangwei Hu is an Associate Professor of Applied Linguistics at the
National Institute of Education, Nanyang Technological University,
Singapore, where he teaches graduate courses in language assessment,
psycholinguistics, and research methodology in applied linguistics.
He has published extensively on language education, language policy,
metalinguistic knowledge, and teaching of L2 writing.