Volume 5 (2014) Issue 1
Determining
the Best Pedagogical Practices for Diverse
Grammatical
Features
Andrew
Schenck and Wonkyung Choi (both Daejeon, South Korea)
Abstract
Past
research has emphasized the universality of grammar acquisition over
key differences, resulting in the development of a number of
one-size-fits-all approaches to grammar instruction. Because such
approaches fail to consider disparities of grammatical features, they
are often ineffective. Just as a doctor needs to diagnose an illness
and prescribe a suitable treatment, the teacher must evaluate a
grammatical feature and choose an appropriate instruction. To better
understand how this may be accomplished, highly disparate grammatical
features (the definite article and plural noun) were taught to adult
second language learners, using three different pedagogical
techniques: Explicit focus on meaning, explicit focus on form, and
implicit focus on form. Results suggest that the effectiveness of
these treatments depends upon characteristics of the grammatical
feature, the type of instruction utilized (implicit or explicit), and
the learner’s language proficiency. According to the results, an
empirical method to guide the content of grammar instruction is
proposed.
Keywords: Explicit
grammar instruction, implicit grammar instruction, semantic
complexity, morphosyntactic complexity, learner proficiency, focus on
form, focus on meaning
1
Introduction
Researchers have developed a number of different theories to explain the seemingly universal nature of grammar development (Cook, Newson & Ning 1988, Freidin 2014, McCarthy 2004, Montague 1970, White 2009). Chomsky (1975, 1981, 1986), for example, posited that there is an innate system of syntactic constraints called principles and parameters. Principles are thought to represent the universal elements found in all languages, while parameters represent “on / off switches” set on a language-specific basis (Mitchell & Myles 2004). While such syntactic theories increase the saliency of the acquisition process, they cannot holistically explain linguistic development. Concerning this issue, Pinker (1994: 105) explained that:
The principles and parameters of phrase structure specify only what kinds of ingredients may go into a phrase in what order. They do not spell out any particular phrase. Left to themselves, they would run amok and produce all kinds of mischief.
As
this statement implies, innate notions of grammar structure alone are
not sufficient to explain the formation of language. In addition to
aspects of syntax and morphology, language learners must be born with
innate notions such as place, agent, and patient, which are used to
further regulate linguistic structures and
imbue them with
meaning (Pinker 1994).
While
innate mechanisms controlling morphosyntactic and semantic aspects of
language construction are well known (Chomsky 2011, Costello & Shirai 2011; Helms-Park 2002, Mitchell & Myles 2004, Pinker
1994), they represent only one facet of language formation. Memory is
yet another essential element of the language construction process.
Research suggests that language use relies considerably on
information received from a lexical store, which resides in long-term
memory (Pinker 1991, Eberhard, Cutting & Bock 2005). The lexical
store may best be characterized as a 'mental dictionary' which holds
syntactic, morphological, and semantic information about words and
phrases. Realization of the importance of this type of long-term
memory has led cognitive theorists such as Chomsky (1995) to suggest
that it is a main component of the acquisition process (Gass &Selinker 2008, Mitchell & Myles 2004).
Short-term
memory, also referred to as 'working memory', is yet another
essential element of language construction. It temporarily stores
input that is to be parsed or, conversely, output that is being
constructed. As
posited by Baddeley (1990, 1999), the working memory is governed by a
central executive which is supported by a visuo-spatial sketchpad
and a phonological loop. The visuo-spatial sketchpad stores semantic
information while the phonological loop holds acoustic information
that must constantly be refreshed through articulation (either saying
or thinking the words) (Cook 2008).
2
Theoretical Framework
In
an attempt to integrate and explain essential components of the
language acquisition process, researchers have designed some holistic
theoretical frameworks (Hurford
1989, Levelt
1989, 1996, and 2001). Levelt (1989), for example, developed a
cognitive model explaining the formation of linguistic output (Figure
1). According to this model, a semantic representation of an
utterance is first developed by a cognitive conceptualizer. The
resulting semantic conceptualization is then programmed into language
by a grammatical encoder. Finally, the encoder works to combine and
modify elements of the lexicon so that a grammatically correct
representation of the semantic concept may be constructed (Bock 1986,
Pienemann 2005).
Figure
1: Model of an innate language construction device
As
Levelt’s (1989) linguistic device includes both semantic and
morphosyntactic elements of language construction, it provides a more
holistic view of the Second Language Acquisition (SLA) process. While
the conceptualizer implies the presence of semantic constraints such
as those hypothesized by Pinker (1994), the encoder implies
morphosyntactic constraints such as those hypothesized by Chomsky
(1975, 1981 and 1986). According to the model, elements of a word,
phrase, or sentence in working memory come from the lexicon and are
organized by the innate construction mechanism. If information cannot
be retrieved lexically, it must be cognitively generated. Such
cognitive production is more difficult than retrieval from a lexical
store, since the innate language construction device must be more
extensively utilized. Exhaustive utilization of the device, in turn,
may put burdens
on the working
memory that serve to limit performance. In addition to gaps within
the lexicon,
a more conscious
evaluation of linguistic output, referred to by Krashen (1981) as a
“monitor”, may increase demands on the innate construction
mechanism, thereby influencing linguistic functioning.
Since
the
inception of English
as a Second Language (ESL) instruction, the role of grammar in the
pedagogical process has been of central concern. Initially, students
learned grammar by translating texts
into and from English. While
this instructional method which is referred to as the
grammar-translation
method
allowed
students to understand written texts, it did not
encourage natural use of
the target language. As a result, learners had difficulty gaining
communicative competence (Celce-Murcia 1991, Huang 2010). To address
this issue, educators soon crafted and utilized a new pedagogical
technique. This method, known as the audio-lingual
method,
prompted
students to learn through a process of habit formation. Learners
listened and repeated sentences that contained grammatical features
and expressions. While this technique encouraged the
spoken
use of the target language, there was little natural communication.
As with the grammar-translation
method, students who learned through the audio-lingual method
continued having problems
with both oral and written forms of natural communication (Thornbury
1999: pp.
21).
Due
to shortcomings with prior instructional approaches that were
primarily grammar-based, a new pedagogical approach was developed
which downplayed the importance of grammar and emphasized learning
through natural communicative tasks. Unfortunately, educators soon
realized that learners who acquired a language via this approach
could communicate, but had significant problems utilizing correct
grammar to express themselves (Lightbown 1998, Lightbown 2000). This
shortcoming resulted in a resurgence of interest in grammar-based
instruction (De Jong 2005, Lightbown, Halter, White & Horst
2002). While modern research now suggests that some kind of grammar
emphasis is needed, there is still considerable controversy over how
this emphasis should be pedagogically facilitated (Han 2012, Norris & Ortega 2000, Spada & Tomita 2010).
In
essence, the failure to concretely determine an effective type of
language instruction reflects a broader issue that has plagued both
SLA research and ESL instruction. Just as models of language
acquisition have overstressed the 'universal' nature of morphosyntax,
methods of instruction have overemphasized a one-size-fits-all
paradigm for teaching grammar. In reality, different forms of
instruction are needed to accommodate the highly disparate
characteristics of syntactic and morphological features. Educators
cannot use the same type of instruction for
each grammatical feature and
expect an equally fruitful outcome. To determine what type of
instruction is needed, diverse characteristics of grammatical
features must be considered. As DeKeyser (2005) points out,
grammatical features may differ significantly in both morphosyntactic
and semantic complexity. This assertion is illustrated through
examination of grammatical features associated with the noun phrase.
Regular plural nouns, which use regular forms such as -s,
-ies, or –es
(e.g. cars,
libraries,
grasses),
are relatively simple morphologically as well as semantically (they
only contain the meaning 'plural'). Lexical plural nouns, in
contrast, which require a complete change of the singular noun (e.g.
children,
men,
teeth),
are morphologically more complex, since
they
have a larger
number of variants. The article has very few morphological variants
(a,
an,
the).
Unlike
the plural
nouns, however, it is semantically very complex, and may be used to
signify semantic concepts such as: unique objects in our world (e.g.
the
sun),
groups in society (e.g. the
homeless),
parts of a list (e.g. the
first thing is…),
superlatives (e.g. the
biggest, the best),
and things already mentioned in a story (e.g. A
man talked to the woman. The man said,
“Hi”)
(Celce-Murcia,
Larsen-Freeman & Williams 1983: pp.279).
As
grammatical features differ significantly, educators must learn to
facilitate acquisition by differentiating instruction. The
development of such an approach requires not only a comprehensive
understanding of the factors that influence acquisition, but a firm
knowledge of the pedagogical techniques that effectively utilize
these factors. While many of the causes influencing grammar
development have been established within a universal framework of
acquisition, the contemporary corpus of research has not clearly
established how manipulating these factors can enhance the learning
of different grammatical features (Spada & Tomita 2010). To
resolve this issue, Han (2012) has suggested that the following
elements of instruction be concomitantly considered:
- Variable characteristics of each grammatical feature (e.g. semantic and morphosyntactic)
- Type of instruction (e.g. explicit vs. implicit)
- The developmental level of the learner
The
simultaneous consideration of the above factors will help educators
determine what type of instruction should be used for each
grammatical feature, how instruction should be used, and at what time
it should be used. This type of comprehensive investigation is needed
before curricula and instruction can be engineered to bring about a
desired result. Thus, the current study was designed to holistically
examine three influences on grammar acquisition:
- semantically and morphosyntactically diverse characteristics of grammatical features,
- different types of instruction, and
- learners’ proficiency levels.
2
Method
2.1
Participants
For
this quasi-experimental study, three university
English-as-a-Foreign-Language classes were selected at a small
university in South Korea (N
= 47). Learners were distributed within the following proportions:
Group 1 (n
= 15), Group 2 (n
= 15), and Group 3 (n
= 17). While most of the students were South Korean, learners from
other countries were also represented: Chinese (n=
6), Mongolian (n
= 1), and Tajikistani (n
= 1). All learners ranged in age from 18 to 21 years.
2.2
Procedure
As
pointed out by Asselin (2002), grammar may represent both “the
conscious knowledge of language structures” and “the unconscious
knowledge of language that allows people to produce and comprehend
language” (Asselin 2002: 52). Consequently, both conscious and
unconscious forms of knowledge about grammatical features were
evaluated within this study. To examine the effects of grammar
instruction on the acquisition process, three grammatical features
associated with the noun phrase were chosen: the regular plural (e.g.
cars,
libraries,
grasses,
etc.), the lexical plural (e.g. children,
men,
etc.) and the definite article the.
These features were selected because their morphosyntactic and
semantic characteristics are highly disparate.
Before
treatments were given to the three groups, participants each took a
pretest. The pretest contained two fill-in-the-blank worksheets each
having 10-14 questions - the worksheets were obtained from
learnenglishfeelgood.com
-
to test conscious knowledge of correct article and plural usage.
Learners were given as much time as they needed to complete the
worksheets. As pointed out by Ellis (2005),
untimed exercises
such as these are better measures of conscious or explicit knowledge
than timed exercises. Following the measure of conscious knowledge,
learners were given two images and three minutes to write a paragraph
for each. This task was timed to obtain a better measure of natural
ability or implicit knowledge of each grammatical feature. Timing
the activity minimized the degree that students could consciously
correct errors, ensuring
the validity of the measure (Ellis 2005). To score each test section,
correct answers were divided by the total number of answers (or the
total number of required contexts for the natural writing task). For
the natural writing task, ratings of grammatical errors (missing or
incorrect forms of the target feature) for one of the groups were
compared to those of an additional native English-speaking rater to
assess reliability. The resulting Cronbach’s a
value of 85.4% suggested that the method of assessment was a reliable
measure of grammatical accuracy. After all
the pretests had been scored,
the non-parametric Kruskal-Wallis
statistical formula was used to confirm that the distribution of
pretest scores did not significantly differ across groups. This
suggested that there was no significant difference between the groups
at the beginning of the study.
Following
the administration of the pretest, three treatments were randomly
assigned to three different groups. Each group received an initial
lecture using a PowerPoint with examples from the topics covered in
Appendix A. The PowerPoint and explanation, however, varied depending
upon the treatment:
- Group 1 received explicit treatment of semantic concepts (e.g., images and graphic organizers with explanation);
- Group 2 received implicit treatment of morphosyntactic features (e.g., sentence examples);
- Group 3 received explicit treatment of morphosyntactic features (e.g., sentence examples with explanation).
Following
the treatment, all learners received a handout so that they could
review and practice the concepts learned during the treatment (Table
1):
Group
1
|
Group
2
|
Group
3
|
Explicit
Focus on Meaning:
PowerPoint mini-lecture with images and semantic maps.
|
Implicit
Focus on Form:
PowerPoint mini-lecture with sentence examples.
|
Explicit
Focus on Form:
PowerPoint mini-lecture with explanation of grammatical forms
along with the same sentence examples presented to Group 2.
|
Handout:
The main topics of the semantic map are provided but the
information in the bubbles is blanked out. Learners must fill in
the bubbles and add their own bubbles. They must then orally make
sentence examples.
|
Handout:
Learners see the sentence examples presented in the mini-lecture.
They practice speaking the sentence examples. They are then
encouraged to make additional sentence examples by speaking.
|
Handout:
Learners see the sentence examples presented in the mini-lecture.
They circle all the grammatical features emphasized in the
lecture. They practice speaking the sentence examples and make
additional sentence examples.
|
Table 1: Summary of treatments for each group
Group
1 received an explicit focus on meaning. The teacher intentionally
drew attention to semantic aspects of the target features through
describing images and graphic organizers included
within a PowerPoint.
Following
the PowerPoint, learners
were given a handout with the images and bubbles only. Students had
to fill in the bubbles according to the images. They were then asked
to expand on the bubbles and make sentence examples, using the target
features.
The
second group received an implicit focus on form. Using
a PowerPoint, learners
were each shown sentence examples which contained the same target
features as those presented to Group 1. Sentences such as the
following were read to students (notice that the target features
parallel those presented to the semantic treatment group):
- I went to the airport last night to pick up my best friend. He just came home.
- Could you stop by the supermarket on your way home from work? I need some milk.
- I stopped at the store and bought a new shirt.
- I went to the movie theater to see a new movie, but it was closed.
- He gave me an apple.
- He gave me some apples.
Following
the PowerPoint, learners were given a handout containing the same
sentence examples. They were then asked to practice reading the
sentences out loud to each other and practice speaking, using
vocabulary and expressions from the sentences. No explicit
explanation or mention of the target grammar was included within this
treatment.
The
Group 3 was shown the same sentences in PowerPoint and handout as
Group 2. Group 3, however, received an
explicit grammar explanation. The
teacher explained aspects of syntax and morphology (sentence position
of the target feature and form) for each feature and then read the
same sentence examples as Group 2. Any semantic explanations were
avoided. Learners were then given the same sentences in a handout as
the implicit group. Group 3, however, had to circle and identify the
grammatical features in the handout. After correctly identifying the
target features, learners were asked to read the sentences out load
and to practice making their own sentences.
Following
the treatment, each group was directly given a posttest. The form of
the posttest was identical to that of the pretest, although different
worksheets and images were used (The worksheets were obtained from
learningenglishfeelgood.com).
Because of the high degree of variability in plural forms (due
primarily to lexical characteristics), five different plural forms
covered in the treatment were also tested within the posttest
assessment of conscious plural knowledge use (lives,
sheep,
mice,
houses,
and libraries).
It was thought that such coverage would more accurately assess the
influence of the treatment, since participants could not be expected
to use lexical forms they had never seen. After the posttest,
descriptive statistics revealing the degree of improvement were
calculated and graphically charted. Next, the significance of
statistical differences between pretest and posttest for the target
grammatical features were determined using a paired samples T-test.
Finally, improvement was compared to participants’ English
proficiency scores, which were obtained from a TOEFL-style test of
listening, speaking, reading, writing, and vocabulary skills. To
obtain a basic
estimate of
the influences of language proficiency, learner improvement scores
were further subdivided into two groups (high and low proficiency)
through using median values. While the proficiency assessment was a
useful tool within this study to examine basic relationships between
English proficiency and the effects of different types of grammar
instruction, future studies will need more precise measures and more
detailed differentiation of proficiency levels.
3
Results
and Discussion
3.1
Analysis
of Group Treatment
Results
of the group treatments revealed different rates of conscious
learning for each target feature (Appendix B). Figure 2, which shows
the degree of improvement from pretest to posttest,
reveals that
Group 3 (the Explicit Focus-on-Form Group) had the highest gains for
all three features. The regular and lexical plural features increased
by 25% and 21%, respectively, on the posttest, while the definite
article rose by 9% only. Group 2, which received implicit grammar
instruction, revealed more moderate gains in conscious knowledge of
the regular plural, the lexical plural, and the definite article;
posttest scores increased by 20%, 17%, and 6%, respectively. Group 1,
which received the semantic treatment, yielded a 23% gain on the
lexical plural. This value was higher than that of Group 2, but lower
than that of Group 3. Mean differences for the regular plural and
definite articles
were
lowest within the semantic treatment group (15% and -2%,
respectively).
The
negative value
for the improvement
of the definite article suggests that the semantic treatment hindered
the participants’ abilities to identify the correct article.
Figure 2: Difference in the mean between pretest and posttest of conscious
knowledge
With
the exception of article use for Group 1, the participants' knowledge
of the usage of all grammatical features increased after each form of
treatment. The Wilcoxon Signed Ranks Test indicated that an
improvement for aspects of the plural was significant for each group
(Table 2):
Definite
Article
|
Regular
Plural
|
Lexical
Plural
|
|
Group
1 Z
|
-.142b
|
-2.090b
|
-2.411b
|
Sig.
(2-tailed)
|
.887
|
.037
|
.016
|
Group
2 Z
|
-1.449b
|
-2.400b
|
-2.322b
|
Sig.
(2-tailed)
|
.147
|
.016
|
.020
|
Group
3 Z
|
-2.325b
|
-2.849b
|
-3.221b
|
Sig.
(2-tailed)
|
.020
|
.004
|
0
|
a. Wilcoxon Signed Ranks Test . b. Based on negative ranks. d. The sum of negative ranks equals the sum of positive ranks.
Table
2: Significance of improvement of conscious knowledge of each
grammatical feature
While
the regular and lexical plural revealed significant gains for each
group, Group 3 had the most significant results, yielding respective
p
values of .004
and .001. Groups 1 and 2 did not show any significant gains for the
definite article. Group 3 (the Explicit Focus-on-Form Treatment
Group), however, did reveal significant gains.
Results
concerning the use of grammar on the natural language writing task
differed from those of conscious grammar knowledge (Appendix C,
Figure 3). Group 2, for example, which did not receive any explicit
grammar treatment, seemed to be negatively impacted by the treatment.
None of the grammatical features were used more proficiently on the
posttest. Group 3, which received treatment focusing on the form of
grammatical features in a sentence, revealed
significant improvement in use of articles and lexical plurals,
increasing by 19%
and 32%, respectively. The explicit semantic treatment of Group 1
appeared to bring about slight gains in use of the article (3%) and
regular plural features (10%):
Figure 3: Mean difference between the pretest and posttest of natural
language use
While
Group 1 revealed gains for both the regular plural and definite
article, neither of these gains was significant (Table 3). Only
improvements in use of the regular plural and definite article for
Group 3 were significant, revealing p
values of .012 and .042. Overall, increases in grammatical accuracy
for both groups with explicit instruction suggest that deliberate
focus on either semantic or morphosyntactic concepts encourages
learners to cognitively monitor and correct errors that occur in
natural language:
Definite
Article
|
Regular
Plural
|
Lexical
Plural
|
|
Group
1 Z
|
-.595b
|
-.271b
|
.000c
|
Sig.
(2-tailed)
|
.552
|
.786
|
1.000
|
Group
2 Z
|
-.981c
|
.000d
|
-1.342c
|
Sig.
(2-tailed)
|
.327
|
1.000
|
.180
|
Group
3 Z
|
-2.511b
|
-2.032c
|
-1.786b
|
Sig.
(2-tailed)
|
.012
|
.042
|
.074
|
a.
Wilcoxon
Signed Ranks Test. b. Based on negative ranks.
c.
Based on positive ranks. d. The sum of negative ranks equals the
sum of positive ranks.
Table
3: Significance of improvement of natural ability of each grammatical
feature
3.2
Analysis of Group Treatment and Language Proficiency
While
our analysis
of group achievement revealed important information about the
effectiveness of treatments, the
further
subdivision of improvement according to proficiency level yielded
even deeper insights:
Figure 4: Pretest / posttest mean difference of conscious knowledge based on
proficiency level.
Figure 4 suggests that the semantic treatment provided in Group 1 may be
ineffective only for learners of lower proficiency. Learners in the
higher proficiency level of Group 1 revealed gains that were slightly
higher than those of the high proficiency level of the Explicit
Focus-on-Form Treatment Group. The difference in achievement within
the high and low proficiency levels of the semantic group suggests
that learners must have a certain level of proficiency before they
can benefit from the explicit semantic treatment used within the
study. This view is supported by qualitative analysis. Students with
higher proficiency levels tend to correctly identify and expand upon
concepts included within the handout. Utilizing conceptions of
“unique things in our world” or “unique things in our society”,
for example, high proficiency learners were able to correctly
identify concepts such as the
earth,
the
poor,
the
sun,
and the
hungry,
and were able to generate new concepts such as the
stars,
the
land,
the
sea,
the
UN,
and the
weather.
Learners of lower proficiency, however, had difficulty completing the
handout and tended to just copy or mimic the expressions used within
the treatment. If such semantic treatment is provided to lower
proficiency groups, semantic complexity might need to be simplified
by utilizing just one, highly simplistic semantic concept. The
definite article, for example, could be explained as being simply
“unique”. Learners
can then begin with
a simple semantic concept and expand upon it as semantic information
in the lexicon grows.
In
contrast to the semantic group,
the morphosyntactic treatment group revealed more significant gains
in the low proficiency category for the definite article. This
phenomenon
may be explained by the largely bottom-up technique used within the
treatment. Unlike the top-down semantic explanations associated with
the definite article in Group 1, lower proficiency students of Group
3 received simple explanations about articles and their adjacent
nouns (e.g. “the airport, the movies”). Thus, learners of lower
proficiency could easily focus
their attention
on a singular concept (the adjacent noun) and could get the correct
answer
by copying and mimicking, rather than by advanced
semantic analysis.
The
high-proficiency
group
3 gained more on the lexical plural than their
low-proficiency counterparts. This phenomenon, as with the semantic
treatment of the
definite article in Group 1, may be a reflection of the complexity of
the grammatical feature. Lexical features have a large number of
variants. Thus, higher-proficiency-level learners may benefit more
from the infusion of lexical variants than their lower-proficiency
counterparts. As for the regular plural, achievement was
fairly consistent for each group. Lower proficiency groups achieved
more, while the higher proficiency groups achieved less.
This consistency
suggests that each type of treatment used within the study had
similar benefits for this feature. This phenomenon may be
a reflection of
the morphosyntactic and semantic simplicity of the regular plural.
Learners may be able to understand the feature without lengthy,
explicit explanations or extensive grammar exercises, making
pedagogical intervention less necessary.
The
comparison of proficiency levels and natural use of grammar also
yielded valuable insight (Figure 5). As with measures of conscious
knowledge for Group 1, the semantic treatment for the definite
article appeared to adversely influence the lower proficiency
learners, while it positively influenced high proficiency learners.
Low proficiency learners showed a mean gain of -2.6%, while the high
proficiency group had a mean gain of 7.5%. Also similar to measures
of conscious knowledge, the morphosyntactic-treatment Group 3
revealed larger gains in natural use of the definite article at the
lower proficiency level. Learners at the lower proficiency level, for
example, showed a 27% increase in article use in writing, which
was double that of
the high proficiency level (11%). In Group 1, the low proficiency
group revealed a large gain in the correct use of the regular plural
(22%). As for the lexical plural, the high proficiency groups 1 and 3
both showed mean increases (14% and 43%, respectively). Only the
low-proficiency learners of Group 3 had a mean gain for the plural
lexical feature (21%). While learners in the explicit-treatment
groups 1 and 3 both showed improvements in natural language ability,
the implicit-treatment group 2 showed no improvement. Implicit
learners actually had more problems with accuracy on the posttest
writing task.
Figure 5: Pretest / posttest mean difference of natural language use based
on proficiency level.
The
mean gains or losses for different grammatical features on the
writing task suggest that the effectiveness of treatments on natural
language accuracy will vary according to characteristics of the
grammatical feature, the type of grammar instruction (implicit vs.
explicit), and the learner’s proficiency level. Overall, explicit
rather than implicit instruction appears to help learners to better
monitor and correct natural language when complexity of the explicit
instruction is commensurate with the leaner’s proficiency level.
Low proficiency learners appear to benefit from explicit treatments
that include more simplistic semantic or morphosyntactic concepts.
The regular plural, for example, may have improved for the lower
proficiency learners of semantic-treatment group 1 because the
feature has a singular semantic concept ('plural') which can easily
be understood and consciously focused upon while writing. Unlike
their lower-level counterparts,
advanced proficiency learners may benefit more from explicit
instruction with more semantic (e.g. the definite article) or
morphosyntactic complexity (e.g. features with several lexical
variants such as the lexical plural or the
past irregular verb). This is suggested by improvements on more
complex features in the higher proficiency levels of Group 1 and
Group 3 (Figure 5). Higher proficiency levels may have a highly
developed lexicon and proficiency which allows them to concentrate on
the semantic and morphosyntactic complexity of some grammatical
features.
Through
the analysis
of natural language performance, it becomes clear that if more highly
complex features are to be taught to low proficiency learners, they
should be simplified. By teaching the concept of unique
or only
1,
for example, learners can gain a vague notion of the definite
article’s more complex semantic meaning. As proficiency increases,
explicit semantic instruction covering more precise contexts of the
definite article may be utilized. Like semantic aspects of grammar,
highly complex morphosyntactic elements (e.g. highly lexical
grammatical features) should be explicitly covered at higher levels
of proficiency. If learners are proficient in English, they have a
more highly developed lexicon that can be used to interpret complex
concepts explicitly covered. The more highly developed lexicon
reduces the load of semantically and morphosyntactically complex
explicit instruction on working memory, ensuring that learners are
able to monitor and improve their natural language performance.
Within future research, more detailed measures of the interaction
between instructional complexity and learner proficiency will be
needed to improve pedagogical techniques and curricular designs that
emphasize grammar.
4
Conclusion
Results
obtained from measures of conscious knowledge suggest that both
explicit and implicit forms of instruction help the learner to
understand grammatical features. As learners have a lot of time to
complete measures of conscious knowledge, they can carefully consider
various elements of sample sentences they receive, including
characteristics associated with target features. Like students who
receive explicit instruction, learners who are engaged with implicit
tasks have the time necessary to construe meaning from the multitude
of semantic and morphosyntactic cues they view within instructional
input.
Unlike
measures of conscious knowledge, those of natural language ability
suggest that only explicit forms of instruction are effective.
Because learners who receive implicit instruction do not have a lot
of time to process the morphosyntactic and semantic elements they
encounter, they may fail to cognitively monitor and correct their
errors. Learners who receive explicit instruction, in contrast, have
a narrower scope to contemplate, which decreases burdens on internal
cognitive monitors designed to correct speech errors. Thus, learners
who have received explicit instruction
appear able
to correct errors more quickly and easily in natural language tasks
such as writing.
In
essence, the influences of implicit and explicit treatments on
natural language are a manifestation of cognitive processes which are
associated with an innate language construction device. The
ineffectiveness of implicit instruction suggests that cognitive
monitors of natural language errors are being distracted by other
contextual factors within the input, while mixed results of explicit
instruction seem to imply that cognitive monitors can only work when
the semantic and / or morphosyntactic information being emphasized is
commensurate with the learner’s proficiency level. Learners of low
proficiency who have a rudimentary understanding of basic semantic
concepts need images or graphic organizers that are semantically
simple. Therefore, explicit semantic instruction of the regular
plural, which has little semantic complexity, is appropriate at this
stage. Explicit semantic instruction of the definite article is only
appropriate at this level if semantic explanations are oversimplified
(e.g. emphasis of a singular concept
such as the definite article’s meaning of 'uniqueness'). After
learners have stored basic
semantic concepts within their lexicon, they can benefit from more
holistic explanations of meaning. As with explicit explanations of
semantic characteristics, explicit forms of morphosyntactic
instruction should be used to emphasize simple, more regular
grammatical features at first. Morphosyntactically or lexically
variable grammatical features may then be explicitly emphasized as
the proficiency of a learner grows.
Results
of the study suggest that explicit grammar instruction must be
tailored to both characteristics of the grammatical feature and
proficiency of the individual learner. The following chart
illustrates how this may be done (Table 4):
Proficiency
Level
|
Semantic
Concepts for Explicit Presentation
(Top-down
/ Focus on Meaning)
|
Morphosyntactic
Forms for Explicit Presentation
(Bottom-up
/ Focus on Form)
|
Difficulty
Level of Explicit Instruction
(Load
on Working Memory)
|
Low
|
plural
|
-s
|
1+
1 = 2
|
unique
|
the
|
1+
1 = 2
|
|
plural
|
-s,
-es
|
1
+ 2 = 3
|
|
Medium
|
plural
|
-s,
-es, -ies
|
1+
3 = 4
|
unique
things in our situation / unique things in our neighborhood /
unique things in our city / parts of a list
|
the
|
4
+ 1 = 5
|
|
plural
|
teeth,
feet, children, men, women
|
1
+ 5 = 6
|
|
High
|
plural
|
-s,
-es, -ies, teeth, feet, children
|
1+
6 = 7
|
unique
things in our world / unique things in our society / unique
situation / unique in our neighborhood / unique things in our city
/ elements of a list
|
the
|
6
+ 1 = 7
|
|
generic
things / unique things in our world / unique things in our society
/ unique situation / unique in our neighborhood / unique things in
our city
|
a,
an, the
|
6
+ 3 = 9
|
Table
4: Determining the difficulty level and appropriateness of explicit
grammar instruction
Table
4 reveals how the content of explicit grammar lessons may be designed
and evaluated for difficulty. As revealed in the table, semantic and
morphosyntactic concepts can be carefully tallied and added to get a
score for overall difficulty level. Because this technique considers
both grammatical characteristics of a target feature and cognitive
levels of proficiency, the explicit instructional concepts and
techniques are tailored to the learner’s level of interlanguage
development. While useful, Table 4 remains only a rudimentary guide
to the structure of explicit instruction. More study of the
influences of explicit instruction on learners of different
proficiency levels is needed to increase the efficacy of such
curricular guidelines.
In
the past, approaches to language instruction applied only one general
principle to enact change. As suggested by
the data within this
study, however, instruction requires a multifaceted approach which
has been designed according to characteristics of a grammatical
feature and the cognitive level of a learner’s development. The
grammar-translation, audiolingual, and communicative methods were all
ultimately doomed to fail because they were not synergistically
combined to accommodate learner needs. In the future, the strengths
of each approach should be integrated to increase their
effectiveness. When a new integrated curricular framework for grammar
is designed, the grammar-translation approach may be used to
emphasize form and meaning; the audiolingual approach may be utilized
to emphasize form and phonological characteristics; and the
communicative approach may be used to emphasize semantic concepts,
such as those which are linked to sociolinguistic contexts. Using
this type of framework, instruction of highly variable lexical
features such as the plural noun or the irregular past can be
enhanced through
a more extensive
use of the grammar-translation or audiolingual methods. Instruction
of semantically complex features such as the definite article can be
improved through more communicative techniques that integrate images
and videos from various sociolinguistic contexts. In essence, each
pedagogical approach has a role in the instructional process.
Researchers and educators must therefore develop a multifaceted
framework that harnesses the influences of grammatical acquisition by
utilizing the strengths of each teaching strategy. This will help to
increase the efficacy of second or foreign language instruction.
Appendixes
Appendix
I
Concepts
Covered with Definite Article (8 Concepts)
- Unique things in our world: the earth, the moon, the sun, the sky
- Unique things in our society: the homeless, the hungry, the poor
- Unique things in our neighborhood or city: the airport, the supermarket, the store, the movie theater
- Unique things in our situation: the cup of coffee on the counter, the bathroom in our dormitory, the sofa, the stove, the swimming pool outside my house
- Unique things in my life: the best book I have read, the greatest day of my life, the worst day of my life
- Unique parts of a list: the first thing is, the second thing is, the third thing is
- Unique parts of a process: the beginning, the middle, the end;
- Unique elements in a story: the man, the woman
Concepts
Covered with Plural (14 Concepts: 7 Lexical Plurals / 7 Regular
Plurals)
- Apples
- Women
- Sheep
- Mice
- Libraries
- Furniture
- Feet
- Fish
- Toys
- Children
- Pens
- Lives
- Parents
- Houses
Appendix
II
- Group NumberDefinite Article PretestDefinite Article PosttestRegular Plural PretestRegular Plural PosttestLexical Plural PretestLexical Plural Posttest1Mean.7857.7800.7857.9222.7024.9333N151514151415Std. Deviation.13766.16562.19258.13897.25469.148402Mean.7667.8267.6833.8778.5333.7000N151515151515Std. Deviation.16525.11629.27495.13313.35187.356073Mean.7437.8294.7059.9118.6275.8824N171717171717Std. Deviation.18739.13117.25365.10404.20858.19995TotalMean.7644.8128.7228.9043.6196.8404N474746474647Std. Deviation.16310.13771.24283.12407.27814.26285
Table
II / 1: Pretest / posttest scores for tests of conscious knowledge
Appendix
III
- Group NumberDefinite Article PretestDefinite Article PosttestRegular Plural PretestRegular Plural PosttestLexical Plural PretestLexical Plural Posttest1Mean.7348.7697.8556.9107.9091.8333N151512141112Std. Deviation.20115.27139.30462.19738.21556.325672Mean.6286.5794.8909.8056.8929.7037N1414119149Std. Deviation.33304.34367.30151.34861.28947.454743Mean.6460.8375.8867.7424.5577.8846N171615111313Std. Deviation.23717.17396.27997.38811.50160.29957TotalMean.6696.7346.8781.8284.7829.8186N464538343834Std. Deviation.25842.28388.28646.30973.38824.35145
Table
III / 1: Pretest/posttest scores for tests of natural language
ability
References
Asselin,
M. (2002). Teaching grammar. Teacher
Librarian, 29(5),
52-53.
Baddeley,
A.D. (1990). Human
memory: Theory and practice.
Needham Heights, MA: Allyn & Bacon.
Baddeley,
A.D. (1999). Essentials
of human memory.
East Sussex: Psychology Press Ltd.
Bock,
J. K. (1986). Meaning, sound, and syntax: Lexical priming in sentence
production. Journal
of Experimental Psychology: Learning, Memory, and Cognition, 12(4),
575.
Celce-Murcia,
M. (ed.) (1991). Teaching
English as a second or foreign language (2nd
ed.).
Boston,
MA: Heinle & Heinle Publishers.
Celce-Murcia,
M., Larsen-Freeman, D. & Williams, H. A. (1983). The
grammar book: An ESL/EFL teacher's course.
Rowley, MA: Newbury House.
Cook,
V. (2008). Second
language learning and language teaching.
(4th ed.). Cary, NC: Oxford University Press.
Chomsky,
N. (1975). The
logical structure of linguistic theory.
New York, NY: Plenum.
Chomsky,
N. (1981). Lectures
on government and binding.
Dordrecht, Netherlands: Foris.
Chomsky,
N. (1986). Knowledge
of language: Its nature, origin, and use.
New York, NY: Praeger.
Chomsky,
N. (1995). The
minimalist program.
Cambridge, MA: MIT Press.
Chomsky,
N. (2011). Language and other cognitive systems. What is Special
About Language?. Language
Learning and Development, 7(4),
263-278.
Cook,
V. J., Newson, M., & Ning, C. (1988). Chomsky's
universal grammar: an introduction.
Oxford, England: Blackwell.
Costello,
W., & Shirai, Y. (2011). The aspect hypothesis, defective tense,
and obligatory contexts: Comments on Haznedar, 2007. Second
Language Research, 27(4),
467-480.
De
Jong, N. (2005). Can second language grammar be learned through
listening?: An experimental study. Studies
in Second Language Acquisition, 27(2),
205-234.
DeKeyser,
R. (2005). What makes learning second language grammar difficult? A
review of issues. Language
Learning, 55(1),
1-25.
Eberhard,
K., Cutting, J., & Bock, K. (2005). Making syntax of sense:
Number agreement in sentence production. Psychological
Review, 112(3),
531-559.
Ellis,
R. (2005). Measuring implicit and explicit knowledge of a second
language: A psychometric study. Studies
in Second Language Acquisition, 27(2),
141-172.
Freidin,
R. (2014). Recursion in generative grammar. In Language
and Recursion (pp.
139-147). New York, NY: Springer.
Gass,
S.M., & Selinker, L. (2008). Second
language acquisition: An introductory course
(3rd
ed.). New York, NY: Routledge.
Han,
Z.H. (2012) Second
language Acquisition.
(Ed. J. Banks) Encyclopedia of diversity in education. doi:
10.4135/9781452218533.
Helms-Park,
R. (2002). The need to draw second language learners’ attention to
the semantic boundaries of syntactically relevant verb classes.
Canadian
Modern Language Review, 58(4),
576-598.
Huang,
J. (2010). Grammar instruction for adult English language learners: A
task-based learning framework. MPAEA
Journal of Adult Education, 39(1),
29-37.
Hurford,
J. R. (1989). Biological evolution of the Saussurean sign as a
component of the language acquisition device. Lingua, 77(2),
187-222.
Krashen,
S.D. (1981). Second
language acquisition and second language learning.
Oxford: Pergamon.
Levelt,
W.J.M. (1989). Speaking:
From intention to articulation.
Cambridge, MA: MIT Press.
Levelt,
W. J. (1996). A theory of lexical access in speech production. In
Proceedings
of the 16th conference on Computational linguistics-Volume 1 (pp.
3-3). Association for Computational Linguistics.
Levelt,
W. J. (2001). Spoken word production: A theory of lexical access.
Proceedings
of the National Academy of Sciences, 98(23),
13464-13471.
Lightbown,
P. (1998). The importance of timing in focus on form. In C. Doughty & J. Williams (Eds.), Focus
on Form in Classroom Second Language Acquisition
(pp. 177-196). New York, NY: Cambridge University Press.
Lightbown,
P.M. (2000). Anniversary article: Classroom SLA research and second
language teaching. Applied
Linguistics, 21(4),
431-462.
Lightbown,
P.M., Halter, R.H., White, J.L., & Horst, M. (2002).
Comprehension-based learning: The limits of 'do it yourself'. The
Canadian Modern Language Review, 58(3),
427-464.
McCarthy,
J.J. (2004). Optimality
theory in phonology.
Malden, MA: Blackwell Publishing.
Mitchell,
R., & Myles, F. (2004). Second
language learning theories
(2nd
ed.). London, UK: Hodder Arnold.
Montague,
R. (1970). Universal grammar. Theoria, 36(3),
373-398.
Norris,
J.M. & Ortega, L. (2000). Effectiveness of L2 instruction: A
research synthesis and quantitative meta-analysis. Language
Learning, 50(3),
417-528.
Pienemann,
M. (2005). Cross-linguistic
aspects of processability theory.
Amsterdam, Netherlands: John Benjamins Publishing Company.
Pinker,
S. (1991). Rules of language. Science,
253(5019),
530.
Pinker,
S. (1994). The
language instinct: How the mind creates language.
New York, NY: HarperCollins Publishers Inc.
Spada,
N., & Tomita, Y. (2010). Interactions between type of instruction
and type of language feature: A meta-analysis. Language
Learning, 60(2),
263-308. doi: 10.1111/j.1467-9922.2010.00562.x
Thornbury,
S. (1999). How
to teach grammar.
Essex, England: Pearson Education Limited.
White,
L. (2009). Second
language acquisition and universal grammar.
New York, NY: Cambridge University Press.
Authors:
Andrew
Schenck
Assistant Professor
Pai Chai University
Ju Si-Gyeong College
Liberal Arts Education Center (LAEC)
Pai Chai University
Ju Si-Gyeong College
Liberal Arts Education Center (LAEC)
Daejeon,
South Korea
E-Mail: Schenck@hotmail.com
E-Mail: Schenck@hotmail.com
Wonkyung
Choi
Associate
Professor
Pai Chai University
Ju Si-Gyeong College
Liberal Arts Education Center (LAEC)
Ju Si-Gyeong College
Liberal Arts Education Center (LAEC)
Daejeon, South Korea
E-Mail:
wkchoi@pcu.ac.kr