Adverb Use and Language Proficiency
Adverb Use and Language Proficiency
Adverb Use and Language Proficiency
Our research examines the use of general adverbs by learners across grades 5,
6, 9 and 10 in the International Corpus of Crosslinguistic Interlanguage (ICCI)
by looking at whether this use increases with age. For our research we use data
from the Polish, Spanish and Chinese components in the ICCI, in particular,
those from the “food” and “money” topics. Our results show that general ad-
verbs are more widely used as age increases. Statistically significant differences
were found between grade 6 and 10 learners across all three L1 groups in terms
of the frequency of use of general adverbs, which suggests that 10-graders inte-
grate adverbs in their discourse in ways that differ from those in previous years.
This study, together with Pérez-Paredes & Díez-Bedmar’s (2012), suggests that
learners below grade 9 are more unlikely to use adverbs.
1. Introduction
This paper focuses specifically on the use of general adverbs (GAs) in learner
language. For this study, GAs are taken to be the broad group of words associ-
ated with the POS tag RR in the CLAWS 7 tag set, i.e. any adverb except for those
included in the following categories: adverbs after nominal head (else, galore),
adverbs that introduce appositional constructions (namely), degree adverbs (very,
so and too were tagged as RR in our research), wh- degree adverbs (how), wh-ever
degree adverbs (however), comparative degree adverbs (more, less), superlative
degree adverbs (most, least), locative adverbs (alongside, forward), prepositional
adverb particles (about, in), wh- general adverbs (where, when, why, how), wh-
ever general adverbs (wherever, whenever), comparative general adverbs (better,
longer), superlative general adverbs (best, longest) and, finally, quasi-nominal ad-
verbs of time (now, tomorrow).
Our motivation is to explore Schmitt’s (2010) claims, among others, that
there is a link between increased communicative competence and use of adverbs.
Specifically, we are interested in testing this claim in the context of very young
learners of English in different educational and cultural settings, as studies that
deal with pre-university students’ use of adverbs are not abundant (Pérez-Paredes
& Díez-Bedmar 2012). This motivation is additionally based on previous research
that has found that lexical competence in learners of English as a Foreign Lan-
guage (EFL) is rather low in Spanish learners (Pérez Basanta 1996, 2005; Lawley
2010; Pavón & Rubio 2010; Sánchez-Hernández & Pérez-Paredes 2005). As the
presence of adverbs in discourse is positively correlated with a more sophisticated
use of the language, we therefore set out to document the presence of GAs in
learner language in three ICCI (Tono 2012) sub-components of different essay
mean length: low (Spanish ICCI), average (Polish ICCI) and high (Chinese ICCI).
In this corpus, the Spanish learners were the group that output the lowest mean
of words per essay in all grades: 42.4 in grade 9 and 57.5 in grade 10. The Polish
wrote 68.3 and 98.2, respectively, while the Chinese were the ones that wrote the
longest essays: 135.5 and 145.6, respectively.
Two research questions arise at this point:
i. Are general adverbs significantly more widely used by learners as age in-
creases? If so, can we identify cut-off points where learners significantly use
general adverbs more frequently?
ii. Is there any difference between speakers of different L1s?
In the following sections we will review research in the area of learner language
and adverb use as well as the morphological foundations of adverbs in Chinese,
Polish and Spanish. Section 2.1 presents an overview of studies analysing adverb
use by learners of different ages, L1s and proficiency levels and highlights the im-
portance of mastering adverb use to achieve native-like competence. This outline
of previous research concludes that, in general, learners lack register awareness,
which is linked the overuse or underuse of certain types of adverbs. Section 2.2
profiles the formation, functions and meaning of adverbs in Chinese, Polish and
Spanish, the three languages under examination. This morphological, syntactic
and semantic outline will serve to illustrate the similarities and differences of ad-
verbs in the three languages.
180 Pascual Pérez-Paredes and María Sánchez-Tornel
1978, 1985, 2008), since the latter favours the use of adverbs derived from adjec-
tives of Romance origin and shows lower rates of short adverbs of native origin.
The situation is not dissimilar to that of other word classes analysed in this study,
which shows that basic characteristics of academic written register are missing
in EFL students’ production. This lack of register awareness, the authors argue
(Granger & Rayson 1998: 7), may be due to the influence of the communicative
approach to language teaching, which often deprives students of opportunities to
be exposed to solid expository or argumentative writing.
In a recent description of linking adverbial (LA) use by Chinese students of
EFL, Liu (2013) compares the incidence of these cohesive devices in written and
oral production contrasted to native-speaker production. The data compiled in
the Louvain Corpus of Native English Essays (LOCNESS), consisting of British’s
students A Levels’ essays and British and American university students’ essays,
were contrasted with the Chinese Learners’ English Corpus, which contains es-
says written by Chinese learners of English ranging from senior high school stu-
dents to juniors and seniors of English majors. The study considers five syntactic
forms of linking devices: single adverbs (therefore), adverb phrases (even so),
prepositional phrases (in addition), finite clauses (that is to say) and non-finite
clauses (to conclude), but we will only refer here to single adverbs. With regard to
written production, the comparison of the CLEC with the control corpus of NS
language, the LOCNESS, yields the following results: four linking adverbials are
overused, of which two are single adverbs (so, then); one adverb (finally) presents
no difference in use by NSs and NNSs; ten linking adverbials, all of them single
adverbs, are underused (anyway, also, well, indeed, though, actually, still, yet, even,
therefore).
As for spoken language, the results obtained after contrasting frequency data
from the College Learners’ Spoken English Corpus (COLSEC) and the London-
Lund Corpus (LLC) suggest that Chinese EFL learners overuse 13 linking ad-
verbials, being most of them single word adverbs (so, also, well, therefore, then,
even, yet, still, finally, indeed, anyway). The opposite situation is found in the
case of the adverbs though and actually, which tend to be underused by NNSs.
When contrasting the occurrence of LAs in spoken and written production of
native-speakers and non-native speakers, Liu (2013) finds that Chinese speakers
and English speakers behave differently. Thus, linking adverbials are significantly
more frequent in spoken production in the case of NNSs whereas NSs tend to use
more linking adverbials when they write. The researcher offers reasons to explain
these findings, which have to do with mother tongue transfer, the influence of
previous language instruction, poor register awareness, lack of understanding of
semantic traits of LAs and pragmatic considerations.
184 Pascual Pérez-Paredes and María Sánchez-Tornel
which fewer errors are encountered; 1 out of 7 times in the former and 0.5 out of 7
times in the latter. Overall, the findings in this study support the hypothesis of the
incremental nature of vocabulary acquisition posited by Schmitt (2010).
In general, the studies reviewed above highlight the importance of the use
of adverbs to achieve native-like performance in a second or foreign language.
A common finding is the difference in adverb choice by native and non-native
speakers, which contributes to hinder ‘nativelikeness’. In many cases, despite
showing an optimal degree of grammatical accuracy, non-native speakers include
in their written or oral production adverbs that would not be chosen by native
speakers in the same context. This is linked to previous research that points in the
direction that there is an extended lack of register awareness among EFL learners
of various L1 backgrounds, which is evidenced by the high occurrence of adverbs
typical of spoken communication in written production in the above-mentioned
studies. Gilquin & Paquot (2008) offer a series of possible factors that influence
the “spoken-like nature of learner writing” (Gilquin & Paquot 2008: 43): the in-
fluence of speech, transfer from the learner’s mother tongue, teaching-induced
factors, and developmental factors.
The three languages under scrutiny in this study present similarities in the forma-
tion, function and meaning of adverbs. The morphological processes whereby
adverbs are formed in Mandarin and Cantonese Chinese, Polish and Spanish are
similar in various respects, since the three languages admit two types of adverbs:
those formed by adding derivational morphemes to already existing adjectives
and those that are not derived from other words, i.e. simple, uninflected words
that cannot be divided into smaller units. Regarding their function, adverbs in
Chinese, Polish and Spanish can function either as modifiers (modifying another
adverb, an adjective or other elements, as in He was obviously wrong) or as adver-
bials (functioning as a clause element on its own, as in Obviously, she had to move
out to a new house after the fire). In the same way, adverbs in these three languages
denote very similar semantic categories, including time, place, frequency, manner
or degree, among many others. A brief semantic, syntactic and morphological
characterisation of Chinese, Polish and Spanish adverbs is presented next.
The description of Chinese adverbs covers briefly the characteristics of this
word class in the two “dialects” spoken by the contributors to the ICCI corpus:
Mandarin, spoken by the participants from Singapore and Taiwan, and Canton-
ese, spoken by contributors from Hong Kong. From a morphological perspec-
tive, being an isolating language, Mandarin is characterized by the limited use
186 Pascual Pérez-Paredes and María Sánchez-Tornel
of derivational suffixes to form adverbs from other words (Li & Thomson 1981).
Suffixation is restricted to manner adverbs derived from adjectives by adding the
suffix -de with or without reduplication of the adjective, as in kuài (“quick”) > kuài-
kuài-de (“quickly”) and zhènding (“calm”) > zhènding-de (“calmly”). In the case of
Cantonese, most of the adverbs that have an equivalent in Mandarin are mono-
syllabic or disyllabic, whereas trisyllabic and polysyllabic adverbs are Cantonese-
specific (Wong 2002). Apart from derived adverbs, there exists a large group of
adverbs which are not derived from other words, such as jìng (“unexpectedly”) or
yòu (“again”) (Po-Ching & Rimmington 2004).
From a syntactic perspective, adverbs in Cantonese and Mandarin serve as
modifier, when they operate at phrase level, and as adverbial, when they operate
at clause level. The most salient difference between Chinese adverbs and adverbs
in other languages is the fact that, in some cases, they can act as modifiers of noun
phrases, and not only as modifiers of adjective and verb phrases, as is often the
case in other languages. A different classification of Mandarin adverbs accord-
ing to their position is presented in Li & Thomson (1981: 320), who distinguish
between movable adverbs, which modify the entire sentence (time and attitude
adverbs); non-movable adverbs, which occur only after the topic or subject (man-
ner and non-manner adverbs); and post-verbal adverbials, which are placed after
the verb to indicate frequency or duration.
The semantic classification of adverbs in Cantonese and Mandarin bears a
strong resemblance to that of English and other languages. Thus, the notions most
often denoted by this word class include: time, frequency, scope, degree, mood or
tone, spirit or manner, affirmation and negation (Bingyong & Felley 1990). Other
classifications (Li & Meizhen 2008) include categories such as time, degree, scope,
repetition, negation, estimation, tone, and inquiry.
The characterisation of adverbs in Polish does not differ significantly from
that of other languages. Morphologically, Polish adverbs can be divided into pri-
mary adverbs (non-adjectival or not derived from other words, e.g. teraz (“now”)
and adjectival adverbs formed by affixation (Swan 2002). Suffixation is a very pro-
ductive mechanism for the formation of adverbs in Polish, being the morphemes
-o and -‘e (equivalent to -ly in English) the most widely used, as in ładnie (“pret-
tily”), derived from the adjective ładny. From a syntactic perspective, adverbs
function either as adverbials or as modifiers of verbs, adjectives or other adverbs.
As in other languages, adverbs in Polish convey meanings related to time, place,
degree, manner, frequency or modification, as illustrated by the classification of-
fered by Bielec (1998).
A very similar picture is found in the case of Spanish regarding the process
of formation of adverbs, as well as their syntactic function and their meaning.
The Real Academia Española (RAE), the official body that regulates the use the
Adverb use and language proficiency in young learners’ writing 187
Spanish language in Spain and Latin America, distinguishes two types of adverbs
depending on their morphological configuration. The Spanish equivalent for
the English suffix -ly is -mente, which is very frequently added to adjectives to
form new adverbs, as in rápido (“quick”) > rápidamente (“quickly”). There are also
simple adverbs not derived from other words such as siempre (“always”) or lejos
(“far”). According to their syntactic role, Spanish adverbs are divided into ar-
gumentative (obligatory in most cases), attributive (those functioning as subject
predicative or object predicative) and adjuncts (modifiers that can be omitted).
The latter category can be further subdivided into nuclear or central adjuncts,
which occur within the long verb phrase, and peripheral adjuncts, occurring
outside it. The same body of experts offers a semantic classification of adverbs
into degree, place, time, manner, affirmation, negation, doubt or aspect (Real
Academia Española 2009). According to their syntactic role, Spanish adverbs are
divided into argumentative (obligatory in most cases), attributive (those func-
tioning as subject predicative or object predicative) and adjuncts (modifiers that
can be omitted). The latter category can be further subdivided into nuclear or
central adjuncts, which occur within the long verb phrase, and peripheral ad-
juncts, occurring outside it.
3. Method
In the following paragraphs we will describe the data and the subjects that took
part in our research as well as the rationale behind our paper.
Our research explores three sub-sets of the ICCI (Tono 2012), specifically
the Chinese, the Polish and the Spanish ICCI components. Over 7,000 students
in grades 3 to 12 produced the ICCI texts in 2009 and 2010, and the entire cor-
pus is made up of 844,400 running words. As pointed by Tono (2012: 28), given
the peculiarities of primary education and the emphasis on oral communica-
tion in the early school years in general, more “data were collected from the
sixth year onward”. The texts were originally handwritten within twenty min-
utes in class, and were then scanned and transcribed to machine-readable texts
(Hong 2012). The thematic areas covered were argumentative and descriptive
essays over a dozen of topics. The writing tasks were assigned as in-class timed
essays and students were not allowed to prepare in advance or use dictionaries
or other resources.
For the purpose of this research, we decided to study the non-elicited uses of
general adverbs in two of the writing tasks in the ICCI. In the “food” topic learn-
ers had to answer the following question: Which is your favourite food? Why? In
the “money” topic, learners had to write about the following: Imagine you win the
188 Pascual Pérez-Paredes and María Sánchez-Tornel
lottery. What do you choose to do with the money? In the entire ICCI, over 2,000
essays were collected for the “food” and “money” topics.
The Chinese ICCI component was contributed by 951 students that produced
95,058 tokens; the Polish ICCI component by 751 students that produced 65,556
tokens; finally, the Spanish ICCI component was contributed by 654 students that
produced 49,174 tokens. The Chinese wrote 99.9 words per essay, the Polish stu-
dents 87.2 words per essay, whereas the Spanish students wrote a mean of 75.1
words. The “food” and the “money” argumentative topics were the most popular
among the contributors. Overall, 334,914 tokens were contributed by the students
that chose the “food” essay, whereas 288,600 tokens were contributed by those
writing on the “money” topic. In total, 20,859 tokens were labelled as general ad-
verbs (the RR tag in the CLAWS 7 tag list) in the whole ICCI. Our interest within
the boundaries of the present paper is confined then to the occurrence of general
adverbs excluding the rest of adverbs except for so and too that were tagged as RR.
The most frequent general adverbs in the entire ICCI are, in descending order of
frequency, usually (1,815 occurrences), so (1,733 occurrences), also (1,382 occur-
rences), always (1,243 occurrences), too (1,044 occurrences), really (807 occur-
rences), only (656 occurrences), often (594 occurrences), just (557 occurrences)
and maybe (450 occurrences). There are 359 hapax legomena, that is, words (or
adverbs, in this case) that only occur once in the entire corpus, while there are
78 dis legomena, that is, words that occur twice in the corpus. The amount of the
former can be explained as it was decided to retain the original spelling of every
single word written by the students that completed the writing tasks in the corpus.
Thus, *persanally occurs once, and so does *actully. On the contrary, other gen-
eral adverbs that were correctly spelt, and only used once, include significantly,
strictly or effectively.
For the purposes of this study, we will examine the GAs produced by 616
learners from grade 5 to grade 10 and therefore analyse the total of GAs used by
every single learner in the three sub-corpora. This is the ‘RR All’ variable in our
study. We will similarly look at the totals of a set of the most frequently used GAs
in every sub-corpus per grade and topic. After a first analysis of the corpora, this
particular set of GAs was composed of nine adverbs: also, always, finally, just,
never, often, once, really and too. This is the ‘RR 9’ variable in our study. For the
sake of comparability, we will focus our attention on grades 5 and 6, on the one
hand, and grades 9 and 10, on the other. The reason to exclude grades 7 and 8
from this inter-group contrast lies in the fact that in the Spanish sub-component,
no essays were written for the “food” topic in these groups. Every single GA was
evaluated by the researchers in order to make sure that tagging was correct. Thus,
to provide an example, we discarded instances of once as GA when the writers,
especially in the “money” topic, meant a charity bearing the same name in Spain.
Adverb use and language proficiency in young learners’ writing 189
4. Results
In the following paragraphs, we will offer an account of the GAs found in every
sub-corpus per topic and grade. Instances of these GAs are provided in italics and,
whenever they are misspelt by the learners, we will keep the original written form
preceded by an *.
The normalized frequencies in Figure 1 show a tendency for learners to use
more GAs across grades.
30
25
CH
20
PL
15
10 SP
5
0
G5 G6 G9 G10
Figure 1. Normalized frequencies of GAs (RR All) per 1,000 words
While the frequency increase is not linear in the case of the Spanish informants,
Polish and Chinese young writers use more GAs as they progress through Sec-
ondary Education.1 Let us examine these differences across topics and grades in
all three data sets.
In grade 6, we found 30 GAs in the “food” topic: too occurred 7 times used by 4
different writers, 3 occurrences of usually and often written by different subjects
each; 2 occurrences of a bit, also and so and one occurrence of, among others,
once, commonly, easily, *solorely, and almost. In grade 9, 95 occurrences of GAs
were found in the “food” topic: so was found 20 times, also 16, usually 10, always
and only 7, too 6, (long) ago and often 4, really, much, maybe and never 3, sweetly,
finally and still 2, while some other GAs were found only once. In the “money”
topic, 190 occurrences of GAs were found: so was found 22 times, maybe 17, also
15, really 10, at last and still 9, finally 8, always, never, often and too 6, only, at first,
so on, perhaps and already 5.
In grade 10, 87 occurrences of GAs were found in the “food” topic: also was
found 15 times, only 6, always 5, finally 4 and, among others, suddenly 1. In the
“money” topic also was found 10 times, always 8, only 7, thirdly 2 and, among
others, quickly 1.
Table 1 shows the normalized frequencies per 1,000 words.
Table 1. Normalized frequency of GAs per 1,000 words across grades and essay topics
in the Chinese ICCI sub-component
“food” topic “money” topic Combined
grade 6 (11-year olds) 8.74 − 8.74
grade 9 (14-year olds) 23.08 20.93 22
grade 10 (15-year olds) 30.06 23.69 26.88
There was a statistically significant difference between grade 6 and grades 9 and
10 as determined by one-way ANOVA (F = 38.9, p = .000) and the Games-Howell
post-hoc test for both the total number of GAs (RR All) and the most frequently
used GAs in the ICCI (RR 9) (F = 13.1, p = .000) in the “food” topic. No statisti-
cally significant difference was found between grades 9 and 10 for RR All and RR
9 in the “money” topic.
In grade 5, 4 occurrences of a GAs were found for the “food” topic, never ap-
peared twice, while good and usually occurred once; in the “money” topic, just
occurred once. In grade 6, we found 29 GAs in the “food” topic: too occurred 10
times used by 8 different writers; often 7 times and always 5 times. In the “money”
topic, we found 10 GAs, too occurred 3 times, while the rest of adverbs, including
still or really, occurred once. In grade 9, 26 occurrences of GAs were found in the
“food” topic: always was found 6 times, usually 5, often 3, while also, maybe and
really occurred twice. 5 occurrences of GAs were found in the “money” essays:
also, finally, just, still and really all occurred once. In grade 10, 26 occurrences of
GAs were found in the “food” topic: often was found 10 times, always 6, *realy 5,
ever, usually and also 4 times each. In the “money” topic, we only found 95 occur-
rences of GAs: really was found 12 times used three times by three writers each,
also was found 10 times, maybe 9 times, so and too 6 times, always, all, and never
5 times, while probably and just were recorded 4 times.
If we normalize these frequencies per 1,000 words, we find the following:
Adverb use and language proficiency in young learners’ writing 191
Table 2. Normalized frequency of GAs per 1,000 words across grades and essay topics
in the Polish ICCI sub-component
“food” topic “money” topic Combined
grade 5 (10-year olds) 1.78 16.95 9.36
grade 6 (11-year olds) 13.02 5.44 9.23
grade 9 (14-year olds) 14.38 8.40 11.39
grade 10 (15-year olds) 19.87 27.38 23.62
There was a statistically significant difference between grade 5 and Grades 9 and
10 groups and between grade 6 and grade 10 groups as determined by one-way
ANOVA (F = 15.2, p = .000) and the Games-Howell post-hoc test for the total
number of GAs (RR All) in the “food” topic. In a similar way, there was a sta-
tistically significant difference between grade 5 and grades 6, 9 and 10 groups as
determined by one-way ANOVA (F = 11.1, p = .000) and the Games-Howell post-
hoc test for the most frequently used GAs in the ICCI (RR 9) in the “food” topic.
There was a statistically significant difference between grade 5 and 10, grade
6 and 10 and grade 9 and 10 groups as determined by one-way ANOVA (F = 11.6,
p = .000) and the Games-Howell post-hoc test for the total number of GAs (RR
All) in the “money” topic. In a similar way, there was a statistically significant
difference between grade 5 and 10 and grade 6 and 10 groups as determined by
one-way ANOVA (F = 9.6, p = .000) and the Games-Howell post-hoc test for the
most frequently used GAs in the ICCI (RR 9) in the “money” topic.
In grade 5, only one occurrence of a GA was found for the “food” topic, just, and
another for the “money” topic: also. In grade 6, we find 4 GAs: also occurred 4
times used by 4 different writers in the “food” topic. Two occurrences of too were
found in the “money” topic, both written by the same learner. In grade 9, 7 occur-
rences of GAs were found: too was found 4 times used by 4 different kids, while in
general, never and twice were found once each. 34 occurrences of GAs were found
in the “money” essays: finally occurred 7 times used by 7 different students, too
6, all of them contributed by different kids, always 4, one writer used it twice, all,
only, also and well, used twice, while so, last, never, much, usually, of course, forev-
er, even and maybe occurred once. In grade 10, 39 occurrences of GAs were found
in the “food” topic: once was found 10 times, usually 6, too 6, twice 4, always 2,
seldom 2, whereas normally, typically, yet, ever, fast, finally, well, hardly and slowly
all occurred once. However, in the “money” topic, we only found 7 occurrences of
GAs in grade 10: also was found 4 times, and well, much and too were found once.
192 Pascual Pérez-Paredes and María Sánchez-Tornel
Table 3. Normalized frequency of GAs per 1,000 words across grades and essay topics in
the Spanish sub-component
“food” topic “money” topic Combined
grade 5 (10-year olds) 6.06 7.09 6.57
grade 6 (11-year olds) 10.61 11.11 10.86
grade 9 (14-year olds) 10.37 6.51 8.43
grade 10 (15-year olds) 25.98 6.17 16.07
There was a statistically significant difference between grade 10 and grades 6 and
9 groups as determined by one-way ANOVA (F = 8.02, p = .001) and the Games-
Howell post-hoc test for the total number of GAs (RR All) in the “food” topic. In
a similar way, there was a statistically significant difference between grade 10 and
grades 6 and 9 groups as determined by one-way ANOVA (F = 6.8, p = .002) and
the Games-Howell post-hoc test for the most frequently used GAs in the ICCI
(RR 9) in the “food” topic.
There was no statistically significant difference between the groups as deter-
mined by one-way ANOVA (F = 0.6, p = .85) for the total number of GAs (RR All)
or for the most frequently used GAs in the ICCI (RR 9) (F = 0.2, p = .81) in the
“money” topic.
There was a statistically significant difference between Spanish 9-graders and their
Polish and Chinese peers as determined by one-way ANOVA (F = 87.1, p = .000)
and the Games-Howell post-hoc test for the total number of GAs (RR All). Sim-
ilarly, there was a statistically significant difference between Spanish 9-graders
and their Polish and Chinese peers as determined by one-way ANOVA (F = 20.6,
p = .000) and the Games-Howell post-hoc test for the most frequently used GAs
in the ICCI (RR 9).
There was a statistically significant difference between Spanish 10-graders
and their Polish and Chinese peers as determined by one-way ANOVA (F = 18.1,
p = .000) and the Games-Howell post-hoc test for the total number of GAs (RR
All). In a similar way, there was a statistically significant difference between
Spanish10-graders and their Polish and Chinese peers as determined by one-way
ANOVA (F = 7.5, p = .001) and the Games-Howell post-hoc test for the most fre-
quently used GAs in the ICCI (RR 9).
Adverb use and language proficiency in young learners’ writing 193
5. Discussion
Our research examines the non-elicited use of learners’ general adverbs across
some of the ages and grades represented in the ICCI, one of the few resources
available to understand learner language as used in the early stages of school-
based formal learning. In particular, our interest lies in documenting whether
GAs are more widely used by learners as their age increases. Our primary moti-
vation is then to explore Schmitt’s (2010) claims that there is a link between in-
creased communicative competence and the use of adverbs in the context of very
young learners of English in different educational and cultural settings, an under-
explored area in learner language research. To do this, we have used data from the
Chinese, the Polish and the Spanish sub-components in the ICCI, in particular,
those from the “food” and “money” topics. Let us discuss our research questions.
The first research question aimed to discover whether general adverbs are signifi-
cantly more widely used as the age of learners increases and, if so, whether it is
possible to identify cut-off points where learners significantly use general adverbs
more frequently. In general terms we can state that GAs are more widely used as
age increases. This is clearly seen in the case of Chinese learners, where the nor-
malized frequencies increase every year. Statistically significant differences were
systematically found, between grades 6 and 10 across the two topics and national
groups in terms of the total number of GAs (RR All) and the most frequently
used GAs across the sub-corpora (RR 9) used by the learners except for the Span-
ish learners that completed the “money” topic. In the case of Polish and Spanish
learners, we can see how there is an upward progression when one looks at the
frequencies of GAs in grades 5 and 10. However, the path is not incremental as in
the case of the Chinese learners. The Polish in the “money” topic group in grade
6 offer a very low frequency of use (5.4 GAs/1,000 words). Possibly the small
number of informants in grade 5 may distort the whole picture, especially when
we contrast the “food” topic normalized mean in grade 5 (1.7 GAs/1,000 words)
against that in grade 6 for the same topic (13.02 GAs/1,000 words).
In the case of the Spanish informants the situation is different. While it is true
that the number of essays is small, it cannot be denied that the number of essays
in grade 9, 22 in the “food” topic and 95 in the “money” topic, should be enough
to offer evidence on our research topic. Notwithstanding this, the normalized
mean of GAs decreases from grade 6 to that in grade 9 in both topics. Most of
the adverbs in this grade were finally, too and always. Our results suggest that, for
194 Pascual Pérez-Paredes and María Sánchez-Tornel
this group of learners at least, the money topic elicited a lower frequency of ad-
verb use. This is corroborated by the absence of statistically significant differences
among the grades in the Spanish corpus for the “money” topic. In the “food”
topic, however, Spanish 10-graders showed a much higher normalized frequency
mean of GAs: 25.98/1,000 words, similar to that of Chinese learners (30.06) and
much higher than that of Polish learners in the same topic (19.87).
In the Polish sub-component significant differences were found between
grades 5 and 6 with grade 10, and between grade 5 and 9 in the “food” topic for
RR All, and between grade 5 and grades 6, 9 and 10 for RR 9 in the same topic. As
for the “money” topic, Polish writers showed significantly different performance
in RR All between grades 5, 6 and 9 and grade 10. If we look at RR 9, these sta-
tistically significant differences are only found between grades 5 and 6 and grade
10. However, these results should been analysed cautiously as other researchers
(Johansson 2008) have reported no differences between 5 and 10-graders in terms
of L1 lexical density and, in particular, lexical variety.
Our results suggest that GAs may behave in a different way as other lexical
words and, consequently, at least in terms of frequency of use, there is evidence
that non-elicited use of GAs increases along with age. The problematic nature
of adverb usage reported in the literature (Hsue-Hueh Shih 2000, Gilquin 2007,
Gilquin et al. 2007) should thus be confined to areas other than lack of use. In
a similar way, our results show that between 5/6 grades and 10 grade there is a
development gap where young writers incorporate GAs into their discourse. Fur-
ther analyses should look at how this incorporation manifests along the 6-year
gap comprised in our study. For example, we have found in our data indication
that this GA use is not only more frequent but also more sophisticated. In grade
6, uses of too, the most frequent GA, are often linked to expressing likes in con-
nection with food, as in Examples (1) and (2):
(1) rice is my favourite food too[icci_chn0026]
(2) Because I like seed, too, eating bread is [icci_chn0048]
In grade 10, however, the Chinese learners have become more sophisticated and
offer a wider range of adverbs to express frequency, as in Example (3):
(3) I eat vegetables nearly every day [icci_chn0931]
Our results suggest that Chinese learners, apart from showing higher frequencies
of GA use, present an earlier cut-off point, in grade 9, in terms of the frequency
of use of GAs. In the case of their Polish and Spanish fellow students it is not until
grade 10 that significant differences with earlier stages of formal training appear.
In other words, it takes three years for a Chinese student to significantly increase
the frequency of use of GAs in both RR All and RR 9, while it takes four years for
the Polish and Spanish to do the same. The exception is for Polish writers in the
“food” topic and RR 9 where there is a significant difference between grade 5 and
6. As we lack evidence from Chinese 5-graders, we cannot approach this issue
in earlier stages. Polish learners do show significant differences between grade
5 and 6, although the low frequency of GAs in grade 5 in the “food” topic (1.78
GAs/1,000 words) may play a major role here. The same can be said about the
written production of Spanish learners, where the analysis yields a misleading
mean of GA use of 6.06. This is due to the fact that only one occurrence of GA was
found in this cross-section of the ICCI, which consists only of a few essays. Based
on this evidence, we can conclude that in our data set Polish and Spanish 5-graders
did not incorporate GAs as part of their linguistic repertoire in communicative
situations that did not explicitly elicit the use of this part of speech. Unfortunately,
the unavailability of essays for this grade in the Chinese sub-component makes it
impossible to establish a comparison with these learners.
Our data corroborate the findings of Jarvis et al. (2003) and Lu (2012) as GA
use is statistically more frequent in grade 10 across all nationalities and the two
topics. Comparisons with Hinkel (2002) or Chen (2006) are difficult to establish,
as their informants are adult learners of English in university settings. Our re-
search presents then new insight into the acquisition path of adverbs in EFL than
can be further examined by researchers in the SLA field.
The second research question considered whether there are differences between
speakers of different L1. Irrespective of their mother-tongue, 5-graders do not
show evidence of GA use in their writing, possibly because the young learner is
not equipped yet to display the range of subtleties that are associated with ad-
verbs. McNally & Kennedy (2008: 2) state that notions “such as speaker-oriented
adverbial modification arguably require a semantic or discourse model which
makes some sort of reference to the speech act or dialogue move being made”; and
later “manner adverbs have been classically treated as verb phrase modifiers and
assigned a corresponding semantics (such as a property of events denotation),
while so-called “speaker-oriented” adverbs like fortunately have been analyzed as
196 Pascual Pérez-Paredes and María Sánchez-Tornel
Our results are in line with research by Pérez-Paredes & Díez-Bedmar (2012)
that found 8 and 9 grades as cut-off points in terms of the use of intensifying
adverbs, and Grant & Ginther (2000) that show that, as proficiency increases, so
does adjective and adverb use of L2 writers even under timed, test conditions.
In our data set, grade 10 sets a dividing line for most of the L1 groups and the
two topics. The same could be said about grade 9, although this is not the case
for both variables (RR All and RR 9) analysed in the case of the Polish corpus.
Our findings apparently clash, however, with those of Levitzky-Aviad (2012) that
reports lack of significant progress during school in the use of low-frequency
words. Subsequently, further research should examine the status of adverbs in
Adverb use and language proficiency in young learners’ writing 197
References
Aguado-Jiménez, P., Pérez-Paredes, P. & Sánchez, P. 2012. “Exploring the use of multidimen-
sional analysis of learner language to promote register awareness”. System, 40 (1), 90–103.
DOI: 10.1016/j.system.2012.01.008
Aitchison, J. 1987. Words in the Mind: An Introduction to the Mental Lexicon. Oxford: Basil
Blackwell.
Altenberg, B. & Tapper, M. 1998. “The use of adverbial connectors in advanced Swedish learn-
ers’ written English”. In S. Granger (Ed.), Learner English on Computer. London: Addison
Wesley Longman, 80–93.
Biber, D. 1988. Variation Across Speech and Writing. Cambridge: Cambridge University Press.
DOI: 10.1017/CBO9780511621024
Bielec, D. 1998. Polish: An Essential Grammar. London: Routledge.
Bingyong, Y. & Felley, M. 1990. Chinese Romanization: Pronunciation and Orthography. Beijing:
Sinolingua.
Chen, C. W. 2006. “The use of conjunctive adverbials in the academic papers of advanced Tai-
wanese EFL learners”. International Journal of Corpus Linguistics, 11 (1), 113–130. DOI:
10.1075/ijcl.11.1.05che
Crystal, D. 1995. The Cambridge Encyclopaedia of the English Language. Cambridge: Cambridge
University Press.
de Haan, P. 1999. “English writing by Dutch speaking students”. In H. Hasselgard & S. Oksefjell
(Eds.), Out of Corpora. Amsterdam: Rodopi, 203–212.
Gilquin, G. 2007. “To err is not all. What corpus and elicitation can reveal about the use of col-
locations by learners”. Zeitschrift für Anglistik und Amerikanistik, 55 (3), 273–291.
Gilquin, G., Granger, S. & Paquot, M. 2007. “Learner corpora: The missing link in EAP ped-
agogy”. Journal of English for Academic Purposes, 6 (4), 319–335. DOI: 10.1016/j.jeap.
2007.09.007
Gilquin, G. & Paquot, M. 2008. “Too chatty: Learner academic writing and register variation”.
English Text Construction, 1 (1), 41–61. DOI: 10.1075/etc.1.1.05gil
198 Pascual Pérez-Paredes and María Sánchez-Tornel
Granger, S. 1998. “Prefabricated patterns in advanced EFL writing: Collocations and formulae”.
In A. P. Cowie (Ed.), Phraseology: Theory, Analysis and Applications. Oxford University
Press: Oxford, 145–160.
Granger, S. & Rayson, P. 1998. “Automatic Profiling of Learner Texts”. In S. Granger (Ed.),
Learner English on Computer. London: Addison Wesley Longman, 119–131.
Grant, L. & Ginther, A. 2000. “Using computer-tagged linguistic features to describe L2 writ-
ing differences”. Journal of Second Language Writing, 9 (2), 123–145. DOI: 10.1016/S1060-
3743(00)00019-9
Hinkel, E. 2002. Second Language Writers’ Text: Linguistic and Rhetorical Features. Mahwah,
NJ: Erlbaum.
Hinkel, E. 2003. “Adverbial markers and tone in L1 and L2 students’ writing”. Journal of Prag-
matics, 35 (7), 1049–1068. DOI: 10.1016/S0378-2166(02)00133-9
Hong, H. 2012. “Compilation and exploration of ICCI corpus for learner language research”. In
Y. Tono, Y. Kawaguchi & M. Minegishi (Eds.), Developmental and Crosslinguistic Perspec-
tives in Learner Corpus Research. Amsterdam: John Benjamins, 47–62.
Hsue-Hueh Shih, R. 2000. “Compiling Taiwanese Learner Corpus of English”. Computational
Linguistics and Chinese Language Processing, (5) 2, 87–100.
Jarvis, S., Grant, L., Bikowskia, D. & Ferris, D. 2003. “Exploring multiple profiles of highly
rated learner compositions”. Journal of Second Language Writing, 12 (7), 377–403. DOI:
10.1016/j.jslw.2003.09.001
Johansson, S. 1978. Some Aspects of the Vocabulary of Learned and Scientific English. Goteborg:
Acta Universitatis Gothoburgensis.
Johansson, S. 1985. “Word frequency and text type: Some observations based on the LOB corpus
of British texts”. Computers and the Humanities, 19 (1), 23–36. DOI: 10.1007/BF02259615
Johansson, V. 2008. “Lexical diversity and lexical density in speech and writing: A developmen-
tal perspective”. Lund Working Papers in Linguistics, 61–79.
Lawley, J. 2010. “Conspicuous by their absence: The infrequency of very frequent words in
some English as a Foreign Language textbooks”. In C. Abello-Contesse, R. Chacón-Beltrán
& M. Torreblanca López (Eds.), Insights into Non-native Vocabulary Teaching and Learn-
ing. Bristol: Multilingual Matters, 145–156.
Leśniewska, J. & Witalisz, E. 2012. “Native vs. non-native English: Data driven lexical analysis”.
Studia Linguistica Universitatis Iagellonicae Cracoviensis, 129, 127–137.
Levitzky-Aviad, T. 2012. “Lexical richness and variation in the writing of school-age EFL learn-
ers at different learning stages and different educational systems”. In Y. Tono, Y. Kawaguchi
& M. Minegishi (Eds.), Developmental and Crosslinguistic Perspectives in Learner Corpus
Research. Amsterdam: John Benjamins, 159–168.
Li, C. N. & Thomson, S. A. 1981. Mandarin Chinese: A Functional Reference Grammar. Berkeley/
Los Angeles: California University Press.
Li, D. & Meizhen, C. A. 2008. A Practical Chinese Grammar for Beginners. Beijin: Beijin Language
and Culture University Press.
Liu, G. 2013. “On the use of linking adverbials by Chinese college English learners”. Journal of
Language Teaching and Research, 4 (1), 149–155.
Liu, Y. & Zhang, H. 2012. “Use and misuse of cohesive devices in the writings of EFL Chinese
learners: A corpus-based study”. In Y. Tono, Y. Kawaguchi & M. Minegishi (Eds.), Devel-
opmental and Crosslinguistic Perspectives in Learner Corpus Research. Amsterdam: John
Benjamins, 169–186.
Adverb use and language proficiency in young learners’ writing 199
Lorenz, G. 1999. Adjective Intensification – Learners vs. Native Speakers: A Corpus Study of Ar-
gumentative Writing. Amsterdam: Rodopi.
Lu, X. 2012. “The relationship of lexical richness to the quality of ESL learners’ oral narratives”.
The Modern Language Journal, 96 (2), 190–208. DOI: 10.1111/j.1540-4781.2011.01232_1.x
McNally, L. & Kennedy, C. 2008. Adjectives and Adverbs: Syntax, Semantics, and Discourse.
Oxford University Press, Oxford.
Osborne, J. 2008. “Adverb placement in post-intermediate learner English: A contrastive study
of learner corpora”. In G. Gilquin, Sz. Papp & M. B. Díez-Bedmar (Eds.), Linking up Con-
trastive and Learner Corpus Research. Amsterdam: Rodopi, 127–146.
Paquot, M. 2012. “EFL writing in the disciplines: Insights from the VESPA learner corpus”. Pa-
per presented at the 5th WU Symposium organized by the Department of Foreign Language
Business Communication, 29th November, 2012. Vienna.
Pavón, R. & Rubio, F. 2010. “Teachers’ concerns and uncertainties about the introduction of
CLIL programmes”. Porta Linguarum, 14, 45–58.
Pérez Basanta, C. 1996. “La integración de los contenidos léxicos en los métodos comunica-
tivos: Una cuestión pendiente”. In L. Durán & P. Bertrán (Eds.), Segundas Jornadas Sobre
Estudio y Enseñanza del Léxico. Granada: Método, 239–310.
Pérez Basanta, C. 2005. “Assessing the receptive vocabulary of Spanish of English Philology:
An empirical investigation”. In E. Martínez-Dueñas (Ed.), Towards an Understanding of
the English Language, Past, Present and Future: Studies in Honour of Fernando Serrano.
Granada: Universidad de Granada, 545–564.
Pérez-Paredes, P. 2010. “The death of the adverb revisited: Attested uses in native and non-
native comparable corpora of spoken English”. In M. Moreno Jaén, F. Serrano Valverde &
M. Calzada Pérez (Eds.), Exploring New Paths in Language Pedagogy: Lexis and Corpus-
based Language Teaching. London: Equinox Publishing, 157–172.
Pérez-Paredes, P. & Díez-Bedmar, M. B. 2012. “The use of intensifying adverbs in learner writ-
ing”. In Y. Tono, Y. Kawaguchi & M. Minegishi (Eds.), Developmental and Crosslinguistic
Perspectives in Learner Corpus Research. Amsterdam: John Benjamins, 105–123.
Pérez-Paredes, P., Sánchez Hernández, P. & Aguado, P. 2011. “The use of adverbial hedges
in EAP students’ oral performance: A cross-language analysis”. In V. Bhatia, P. Sánchez
Hernández & P. Pérez-Paredes (Eds.), Researching Specialized Languages. Amsterdam:
John Benjamins, 95–114. DOI: 10.1075/scl.47
Philip, G. 2007. “Decomposition and delexicalisation in learners’ collocational (mis)behav-
iour”. In M. Davies, P. Rayson, S. Hunston & P. Danielsson (Eds.), Proceedings of Corpus
Linguistics 2007. Birmingham: Birmingham University Press. Available at: http://ucrel.
lancs.ac.uk/publications/cl2007/paper/170_Paper.pdf (accessed May 2013).
Philip, G. 2008 “Adverb use in EFL student writing: From learner dictionary to text produc-
tion”. In E. Bernal & J. De Cesaris (Eds.), Proceedings of the XIII Euralex International Con-
gress. Barcelona: Universitat Pompeu Fabra. Institut Universitari de Lingüística Aplicada,
1301–1310.
Po-Ching, Y. & Rimmington, D. 2004. Chinese: A Comprehensive Grammar. London/New York:
Routledge.
Quirk, R., Greenbaum, S., Leech, G. & Svartvik, J. 1972. A Grammar of Contemporary English.
London: Longman.
Rankin, T. 2010. “Advanced learner corpus data and grammar teaching: Adverb placement”. In
M. C. Campoy-Cubillo, B. Bellés-Fortuño & L. Gea-Valor (Eds.), Corpus Based Approaches
to English Language Teaching. London: Continuum, 205–215.
200 Pascual Pérez-Paredes and María Sánchez-Tornel
Authors’ addresses
Pascual Pérez-Paredes
Departamento de Filología Inglesa
Regional Campus of International Excellence “Campus Mare Nostrum”
Universidad de Murcia
30071, Murcia
Spain
pascualf@um.es
María Sánchez-Tornel
Departamento de Filología Inglesa
Regional Campus of International Excellence “Campus Mare Nostrum”
Universidad de Murcia
30071, Murcia
Spain
mstornel@um.es
Copyright of International Journal of Corpus Linguistics is the property of John Benjamins
Publishing Co. and its content may not be copied or emailed to multiple sites or posted to a
listserv without the copyright holder's express written permission. However, users may print,
download, or email articles for individual use.