Grammar in Academic Writing

HOW GRAMMAR MATTERS IN NNS ACADEMIC WRITING: THE RELATIONSHIP
BETWEEN VERB TENSE AND ASPECT USAGE PATTERNS AND L2 WRITING
PROFICIENCY IN ACADEMIC DISCOURSE

BY
KYUNG EUN MIN

THESIS
Submitted in partial fulfillment of the requirements
for the degree of Master of Arts in Teaching of English as a Second Language
in the Graduate College of the
University of Illinois at Urbana-Champaign, 2013

Urbana, Illinois
Master’s Committee:
Professor Fred G. Davidson, Chair
Associate Professor Randall W. Sadler

ABSTRACT
Grammar teaching in ESL/EFL classes has been controversial as it tackles various
issues such as the necessity, scope, contents, and methods of teaching grammar. This thesis
addresses the significance of teaching grammar in ESL/EFL academic writing courses by
restating the question of “whether or not to teach grammar” to “how and when to teach which
grammar.” This study used 120 de-identified academic essays written by NNS students—40
essays from three levels of students—during the official English Placement Test writing
exam at UIUC. By counting the raw numbers of the English verb tense and aspect
combinations in four main aspectual categories and analyzing the functional uses of each
tense and aspect from actual student writing samples, this study aimed to investigate whether
the usage of verb tense and aspect would have a positive relationship with the students’
language proficiency. Although it was possible to hypothesize that more proficient students
would have better knowledge of verb tense and aspect and use more variety of verb
structures, the results strongly suggested that grammatical knowledge represented by the
NNS’ usage patterns of verb tense and aspect system was not the only factor; rather, the
analysis of students’ writing samples highlighted the role of academic discourse as a
significant variable to the frequency of each verb structure. Moreover, the error analysis of
students’ writing samples implied that merely counting the frequency rate of “correct uses” or
“avoided uses/ non-uses” did not fully reflect their L2 writing proficiency; it seemed to have
close relationship with their knowing of more variety of functions each verb tense and aspect
and their ability to make opportune uses of each function for the appropriate content. Put in
another way, these findings suggested that patterns of using English verb tense and aspect
was relevant to the students’ L2 writing proficiency because higher level students were able
to incorporate their grammatical knowledge to the paper’s purpose, contents, and discourse
register. Thus, this study concludes by proposing that “grammar” is neither an opposing
ii

component of “contents” when teaching L2 academic writing nor a competing factor against
“structure or coherence” when assessing students’ writing proficiency. It would be most ideal
and reasonable to choose grammar content that does not stand alone but have significance in
meaning, function, and purpose as they merge into the entire writing process.

iii

ACKNOWLEDGMENTS
This paper would not have been possible without the support of many people. I would
like to acknowledge my thesis adviser, Dr. Fred Davidson, for his insightful advice and
guidance as well as his support and encouragement throughout the process. I also give thanks
to Dr. Randall Sadler and Sun Joo Chung for their feedback and support. And special thanks
to my family and boyfriend for being my motivation, hope, and love. I praise and glorify
God, my good shepherd, for leading every path of mine.

iv

TABLE OF CONTENTS
CHAPTER 1 INTRODUCTION ................................................................................................. 1
CHAPTER 2 LITERATURE REVIEW ..................................................................................... 6
CHAPTER 3 METHODOLOGY .............................................................................................. 27
CHAPTER 4 RESULTS ............................................................................................................. 34
CHAPTER 5 DISCUSSION AND CONCLUSION ................................................................. 74
REFERENCES ............................................................................................................................ 91

v

CHAPTER 1
INTRODUCTION
“When you grade the essays, focus on the content and look at the structure of the entire
essay. Does it have a thesis statement in the first paragraph? Are they using the right type of
evidence as the supporting details…?”
Above quote is the typical type of advice given during the EPT rater training sessions. In the
beginning of each semester, instructors of English as a Second Language (ESL) courses get busy
with the overflowing number of international students at University of Illinois-Champaign
Urbana. Most international students are required to take English Placement Test (EPT) 1and
enroll in an ESL course according to their English proficiency. In this process of rating the
essays, raters are advised to follow a holistic rubric and to concentrate on the structure and the
content of the essay as quoted above; it is not difficult to notice that much emphasis is put on the
macro-features of academic writing (e.g. organization of the essay, relevancy to the essay
prompt, coherence, unity, etc.) while grammatical aspects receive minimal, if any, attention. In
the same venue, grammar is not explicitly taught in any of the ESL writing courses.
However, from the experiences of teaching ESL students of various levels, I have
noticed that grammatical errors often inhibit unity, coherence, and clarity of the entire content
and keep NNS writers’ writing from being as natural and academic as their NS counterparts’
papers. In fact, a large number of studies have also supported grammar teaching (Norris &
1

“Many newly admitted UIUC international students are required to take the EPT according to their pre-arrival
scores on either of two international ESL tests: the Test of English as a Foreign Language (TOEFL) or the
International English Language Testing System (IELTS). Or, individual departments and/or advisors may require or
encourage students to take the EPT, and students themselves may elect to take the test so that they can enroll in ESL
courses at UIUC. The EPT consists of two parts: a written test and an oral test. The writing test requires students to
produce an academic essay based on the information obtained from a reading passage and a short lecture. In the oral
test, students are given a topic on which to speak for three minutes. If students speak intelligibly, they will be
exempted from further oral testing. Otherwise, students will be required to take another oral exam.” (UIUC EPT
website retrieved from: http://www.linguistics.illinois.edu/students/placement/)
1

Ortega, 2000; Ellis, 2006; Roberts & Cimasko, 2008; Zhou, 2009) based on the analysis of L2
production errors, and many have come to advocate teaching grammar, especially in regard to its
meaning, function, and usage (Celce-Murcia & Larsen-Freeman,1999; VanPatten, Williams, &
Rott, 2004; Ellis, 2006). In this light, I also wanted to analyze NNS writing samples from EPT
database by categorizing the grammatical errors found in three-leveled groups of students—ESL
500, ESL 501, and the group whose students are exempt from taking ESL, either by direct
exemption based on the EPT or by subsequent proficiency in the early weeks of the semester. I
was planning to compare the three groups by each category of grammatical errors in order to see
if there would be a relationship between the overall writing proficiency of students and their
ability to correctly use grammatical knowledge. However, by reviewing the previous literature, I
realized that it could be very difficult to categorize errors without overlapping them. Besides,
some studies have implied that certain grammatical errors may cause more confusion and
interference in communication, which is explained by the “error gravity” research (Johansson,
1978; Khalil, 1985; Sheorey, 1986; Santos, 1987; Tomiyana, 1980, as cited in Roberts &
Cimasko, 2008, p. 137). Moreover, much research has acknowledged the potential risks of
relying only on the errors to analyze L2 learners’ language, for errors may not truly reflect the
target language actually used by NNS, who often resort to avoiding certain grammar structures
instead of making errors (Schachter & Celce-Murcia, 1971; Schachter, 1974; Mattar, 2002;
Mattar, 2003; Hinkel, 2004; Gass & Selinker, 2008). Therefore, this study refines the scope of
research to English verb tense and aspect system, which is one of the most problematic and
erroneous areas in NNS’ writing, but this project develops the research methods to consider not
only salient production errors but also the raw frequency (or frequency rates) of each verb tense
and aspect combination that NNS writers actually employ in their writing.
2

English verb tense and aspect system (including modality only partially) has been
suggested as one of the areas in which NNS make most errors (Amsberg, 1984; Ubol, 1981;
Kim, 1983; Meziani, 1984; Bryant, 1984; Richards, 1985; Hantrakul, 1990; Pongsiriwet, 2001)
and from which much misunderstanding generates (Reid, 2006). Roberts and Cimasko (2008)
have claimed verbal errors to be most well-detected and markedly noticed by the NS professors
in various fields of study. Zhou’s (2009) study has reported verb tenses/verbal forms to be the
aspect of grammar which is the “most frequently mentioned objects of goals by the ESL
learners,” (p.38) which means that students actually desire to learn more about verb tenses in
order to improve their writing skills. Combining what has been claimed of the English verb tense
and aspect system, it seems very reasonable and worthwhile for the current study to focus on this
particular aspect of grammar when identifying the relationship between ESL students’ grammar
knowledge and their language proficiency in academic writing.
The current study intends to go a step further and anticipate possible variables in NNS
writers’ pattern of using verb tenses and aspects. That is, if there is no salient relationship
observed between NNS writers’ usage patterns of verb tense/aspect and their writing proficiency
level, it would be necessary to investigate the influence of academic discourse, which is widely
agreed to play a significant role in disciplining styles and conventions of students’ language uses
in the writing process (Shaughnessy,1976/1977; Bartholomae, 1985 as cited in Zamel, 1987;
Biber et al., 1999; Gee, 2003; Coxhead & Bryd, 2007; Osborn, 2009; Gray, 2011).
When ESL students make grammatical errors or use unpolished expressions in their
academic papers, they are usually the ones blamed for not meeting the high standards of
academic discourse; they may often be accused of having insufficient linguistic knowledge,
lacking practice, and/or even being lazy. However, given that the subjects of the study are
3

graduate students admitted to a high-ranking school in the US, it is reasonable to assume that
they are highly educated intellectuals who are likely to have received much English education.
Their unsatisfactory outcomes in writing could be attributed to the discrepancy between their
knowledge of academic discourse disciplines and how they actually apply their knowledge to the
writing practice. In fact, a majority of the existing studies, references, and guidebooks on
academic writing discourse have mentioned only issues related to a limited range of verb tense
and aspects (e.g. present simple and past simple) and have explained only partial meaning,
function, and usage of these structures. In particular, there has been little, if any, guidance to use
certain tenses and aspects, such as the perfect, the progressive, and the predictive/hypothetical
aspect. Nevertheless, Hinkel (2004) has noticed how native speakers actually use more variety of
verb tenses and aspects to makes subtle meaning differences, and Wallwork (2011) has
mentioned that every tense has different functions in different contexts.
The purpose of this study is, therefore, to analyze NNS writers’ patterns of using verb
tense and aspect in academic discourse and find out the actual cause for such phenomena. If
there is a positive relationship—that is, if advanced students employ more variety of verb
structures while less advanced students avoid using certain types of verb structures—it would
indicate that ESL students’ usage of English verb tense and aspect can be one of the factors to
assess and anticipate their language proficiency, and thereby advocate the inclusion of grammar
instructions in ESL writing courses at the university level. It could also be extended to propose
an analytic test component in addition to current writing assessment for improving the EPT. On
the contrary, if there is no salient relationship, or if certain patterns exist across the language
proficiency levels, the reason may derive from the disciplinary role that academic discourse
plays in L2 writing; in this case, it would be essential that ESL writing courses design syllabus
4

and teach students in two critical areas. First, the fundamentals of English grammar contents
should be taught even if there seems to be no direct, immediate benefit in academic writing.
Simultaneously, the writing teachers should also address the conventions of academic discourse
so that the students may apply general descriptive grammar knowledge to a more unique and
specific context. Therefore, the focus of this study will not only uncover the underlying
assumptions about NNS writers’ verb tense and aspect usage but also propose how the new
findings should practically be implemented in courses so that ESL students may employ English
verb tense and aspect system appropriately in their writing to successfully present their ideas and
meet the standards of academic discourse. The current study also intends to suggest any
necessary modifications for the EPT written exam for further development.

5

CHAPTER 2
LITERATURE REVIEW
Grammar Teaching in ESL/EFL Classes
Grammar teaching in ESL/EFL classes has long been a topic of heated debate, triggering
issues like “should we teach grammar?”, “which grammar should we teach?” and “should
grammar be taught explicitly or implicitly?”(Ellis, 2006, p. 83-84) Although grammar teaching
might be defined in a narrower aspect as a reference to traditional grammar-translation type of
“presenting and explaining grammar (Ur, 1996)” or “presenting or practicing grammar (Hedge,
2000),” it could also be viewed in a broader term (Ellis, 2006, p.84):
“Grammar teaching involves any instructional technique that draws learners’ attention to
some specific grammatical form in such a way that it helps them either to understand it
meta-linguistically and / or process it in comprehension and/or production so that they can
internalize it”
In this broader perspective, much recent research such as Norris and Ortega’s (2000) metaanalysis of 49 studies and Ellis’ (2002a) research (as cited in Ellis, 2006, p. 85) have advocated
grammar teaching for its effectiveness in second language teaching. According to Ellis (2006),
“there is now convincing direct and indirect evidence to support the teaching of grammar
(p.86).” It has also been widely agreed that meaning-focused, “descriptive grammar” (CelceMurcia and Larsen-Freeman, 1999) or functional grammar for communicative uses of language
should be the center of attention in ESL/EFL classes (VanPatten, Williams & Rott, 2004; Ellis,
2006) in contrast to the traditional form-focused grammar.
However, deciding “which grammar to teach” and “how to teach” have remained more
controversial. Some have claimed that only minimal concepts of grammar should be taught while
others have argued for the necessity of teaching the entire grammar. At the one end of the
6

spectrum is Krashen’s (1982) minimalist position (as cited in Ellis, 2006); it claims that grammar
instruction is unnecessary for second language learners and supports only minimal amount of
explicit teaching which will enable L2 learners to access the principles of Universal Grammar
(UG). At the other end stands the comprehensive position (Walter & Swan, 1990; Murphy, 1994;
Quirk, Greenbaum, Leech, & Svartvik, 1972, as cited in Ellis, 2006), which Ellis (2006) explains
as the teaching of all grammar contents available. Not surprisingly, both positions at the ends of
the line have been noted of limitations (Ellis, 2006), and such inconclusive research failed to
provide a clear guidance to the appropriate L2 grammar instruction (Zhou, 2009). Thus, it
became more urgent and significant to select the right grammatical content for L2 learners (Ellis,
2006).
As the means to maximize the effectiveness of grammar instruction, several
methodologies of selecting the grammar content have been considered. Mackey (1976) examined
the frequency of the input learners receive and considered the utility of certain grammar contents
for the learners. Ellis (2006) proposed L1 influence and markedness as the key factors to decide
on the appropriate L2 grammar contents. He then proposed using learners’ production errors as
the solution to disambiguate the concept of markedness and as the key to select which
grammatical contents to teach.
According to Gass and Selinker (2008), error analysis is “a type of linguistic analysis that
focuses on the errors learners make” (p. 102). They cited from Corder (1967) who first
considered learners’ errors “not as something to be eradicated, but rather as something [that can
be] important in and of themselves,” and since then learners’ errors have often been viewed as
“evidence of an underlying rule-governed system” which may indicate their current
interlanguage developmental state (p.102). From this perspective, analyzing L2 learners’
7

production errors has been used as a valuable research method (Dulay & Burt, 1974; El-Sayed,
1982; Ghadessy, 1980; Yang, 1994; Corder, 1981 as cited in Pongsiriwet, 2001) as it may extend
to selecting which grammar to teach, sometimes proposing certain priority and hierarchy among
the essential elements of the core grammar contents. Furthermore, more recent studies have gone
further to suggest considering “error gravity hierarchies” (Johnason, 1978; Khalil, 1985;
Sheorey, 1986; Santos, 1987; Tomiyana, 1980, as cited in Roberts & Cimasko, 2008) to select
the grammar contents whose misuses would disrupt meaning deliverance most significantly.
These methodologies, but not limited to what is mentioned above, have guided language
teachers and curriculum designers to decide on the selection and the priority of the core grammar
contents to teach, and not surprisingly, these selections have been implemented in ESL textbooks
and/or grammar reference books.
Teaching English Verb Tense and Aspect in ESL/EFL Writing Courses
As the result of various attempts to select the “right grammar content to teach in
ESL/EFL classes,” English verb tense and aspect system has long been treated with significance
and priority in many ESL textbooks. In particular, many popular ESL textbooks which are
related to academic writing (Hinkel, 2004) frequently address various structures of English verb
tense and aspect; it has been found that these writing reference books include “at least a very
short section on verb tenses and voice and their uses in formal written prose” (Raimes, 1992,
1999; Swales & Feak, 1994; Beason & Lester, 2000; Axelrod & Cooper, 2001; Hacker, 2003 as
cited in Hinkel, 2004, p.7-8).In addition, it is not difficult to find a grammar reference book that
dedicates at least one chapter solely to English verb system, such as Swan and Smith’s (2011)
Learner English; Celce-Murcia and Larsen-Freeman’s (1999) The Grammar Book, Swan’s
(2005) Practical English Usage, and Azar’s (1999) Understanding and Using English Grammar.
8

However, such “trends” to include the topic of English verb tense and aspect in popular
ESL/EFL textbooks and reference books do not support its importance. As Ellis (2006) points
out, the selection of grammar content remains problematic because the long lifespan of such
selections in popular ESL textbooks may have resulted for the sake of safety. Zhou (2009) also
claimed that research up to date is inconclusive, which implies the need for additional evidence
to verify the significance of teaching English verb tense and aspect system. Especially, the
benefits of teaching English verb tense and aspect to improve NNS’ writing proficiency needs to
be scrutinized in order to make sure that it is an indispensable area of instruction in ESL/EFL
writing courses.
Error analysis of L2 production. Many studies to date have continued to address the
significance of teaching English verb tense and aspect system by referring to L2 learners’
production errors (Amsberg, 1984; Wyatt, 1972; Ghandessay, 1976; Ubol, 1981; Kim, 1983;
Bryant, 1984; Richards, 1985; Hantrakul, 1990; Pongsiriwet, 2001).
Pongsiriwet (2001) identified 12 types of most common grammatical errors in NNS
compositions and reported that errors occurred most frequently in the verb usages, in subjectverb agreement, verb formation and tense. He claims this result to be consistent with previous
studies which have acknowledged the use of tenses, verbs, and articles as the most erroneous
areas in L2 writing production (Arani, 1993; El-Sayed, 1982; Ghadessy, 1980; Kroll, 1990;
Santos, 1988; Scott & Tucker, 1974; Yang, 1994, as cited in Pongsiriwet, 2001). Similarly, Reid
(2000) stresses the importance of learners’ ability to accurately understand and use verb tenses
by saying that verb tense errors can cause interference in communication. Hantrakul (1990) also
emphasizes the importance of L2 learners’ tense usage by saying, “In writing, non-native

9

speakers, regardless of their academic levels or ethnic and linguistic backgrounds, share
problems in tense and aspect usage” (p.47).
Also, Roberts and Cimasko (2008) examined the errors in L2 writing on the basis of
“error gravity” research (Johansson, 1978; Khalil, 1985; Sheorey, 1986; Santos, 1987;
Tomiyana, 1980, as cited in Roberts & Cimasko, 2008, p. 137) and conducted an investigation to
see which errors in L2 written production were most identified and corrected by the professors of
various fields, who played the role of the essay raters. The results revealed the salient
“makedeness” of verbal errors and concluded that “native readers are most sensitive to verb
errors (p.137)” by referring to the responses of the rating faculty. Their findings reinstate the
significance of verb errors by emphasizing the meaning disruptions made by “poor word choice
and verb tense or modal errors” (p.137).
Avoidance in L2 writing production. While it holds true that L2 production errors may
provide invaluable resource for analyzing learners’ interlanguage and a critical data for
understanding with which language features the learners are having difficulty, criticisms of error
analysis should not go unnoticed.
Gass and Selinker (2008), in their introductory SLA course book, mentions four concerns
against error analysis, among which is the phenomenon of “avoidance” noted by Schachter
(1974). She examined how English restrictive relative clauses would be used by five groups of
NNS with different L1 and showed that Japanese and Chinese L1 learners of English produced
far fewer errors than the Persian and Arabic L1 groups (p.104-106). She attributes this error-less
production of Japanese and Chinese students to avoidance, which is closely related to the L1
interference. Louda (1981) have also agreed that avoidance is more likely to occur when there is
a similarity between learners’ first language and the target language.
10

On the other hand, in a more recent study, Mattar (2003) proposes that avoidance may
not be directly and solely influenced by the similarity between L1 and L2 but possibly related to
learners’ language proficiency level. Kleinman (1983) also diverts from the L1 interference
theory and focuses more on confidence, which “reflects the learner’s perception of his
knowledge rather than his knowledge of some structures” (as cited in Mattar, 2003, p. 104-105).
These findings in sum have stimulated more in-depth analyses on the phenomenon of avoidance
in consecutive studies, for avoidance in L2 production may hold the key not only to understand
L2 learners’ errors but also to providing practical guidelines and directions to language teachers
and curriculum designers of second language teaching.
As briefly mentioned in the previous section, L2 learners’ use and non-use of certain
grammar structures in their production, which is generally referred to as avoidance, is not a rare
phenomenon in second language learning, and many researchers have attempted to figure out its
patterns, causes, and effects in relation to various language features. In such attempts,
frequency—the raw number counts or the percentage rates of target language usage features—
has been often used to examine the possible causes of avoidance as it provides the opportunity to
detect any avoidance phenomenon in L2 production. In fact, it may be a more efficient way to
focus on the actual cases of L2 learners’ non-use of particular language features because it can
offer a comparative picture of avoidance across multiple variables (e.g. L1 variation, L2
proficiency level, age, etc.); by looking specifically at the exact numbers and/or rates of the
target language structures used by the different groups of subjects, it is possible to find out which
factor may affect the language usage. For instance, the frequency comparison between the native
speakers of English (NS) and the non-native speakers (NNS) has been widely used in studies on
L2 learners’ use and non-use of certain language features such as words, cohesive devices,
11

idiomatic phrases, phrasal verbs, and so forth in both oral and written production (Irujo, 1986;
Gary, 2011; Hubert, 2011; Saito & Poeteren, 2012; Mattar, 2002, 2003). And many studies
showed interesting findings on L2 learners’ avoidance, generating more debates about what it
really is, why it occurs, and how it influences second language acquisition and teaching
approaches.
Although avoidance could easily be attributed to L2 learners’ lack of confidence and/or
knowledge, different views have been suggested by current studies, such as its strategic function,
language proficiency variables, unconscious non-use, paraphrase, and so on. Mattar (2002)
studied avoidance of the subordinating adverbs by comparing 89 Arabic speakers in 5 different
groups according to their English proficiency. The results showed that the frequency of using the
subordinating adverbs was much higher in the group of advanced students while the lower-level
students relied on avoidance strategy by using similar phrases. Consequently, he claims his study
to challenge the domineering L1 interference theory and suggests that L2 learners’ avoidance be
viewed from various factors including learners’ language proficiency and the strategic inventions
of writing process.
The causes of NNS writers’ use and non-use of English verb tenses and aspects.
Avoidance of complex verb structures. Not only in terms of words, phrases, or
expressions but also of grammatical aspects has avoidance been a common phenomenon.
However, relatively few studies seem to have focused on L2 learners’ avoidance when analyzing
their usage patterns of English verb tenses and aspects; even when they did notice a clear case of
avoidance in certain language structure, it has been perceived as a natural, obvious response to
encountering a “difficult or complex” language structure, whether the difficulty derives from L1L2 morphological, semantic, and syntactic difference or not (Hinkel, 2004; Collins, 2007).
12

Avoidance being simply attributed to students’ lack of understanding or their inability to use
proper verb forms in writing contexts, little research has suggested the negative relationship
between NNS writers’ language proficiency and their tendency to avoid using difficult verb
structures. Instead, a number of studies related to English grammar seem to have taken it for
granted that the students’ knowledge on grammar would match with their ability to use it in a
proper, fluent, and accurate way. Therefore, studies up to date have focused mainly on the
learning of “inflected forms of verbs associated with particular tenses, aspects, and the passive
voice markers,” (Rutherford, 1984; Sharwood Smith, 1991, 1993 as cited in Doughty &
Williams, 1998, p.17) and the method of these studies often relied on “controlled experiments
and tasks” such as fill-in-the-blank type of questions (Hinkel, 2004, p.6). Although these studies
may have been and may still be informative, it is likely that the controlled experiments, which
mainly investigate L2 learners’ grammar knowledge, may not fully reveal how NNS writers
actually employ these linguistic features in writing as Hinkel (2004) points out in his
introduction.
In an attempt to overcome this problem and investigate how NNS writers actually use
English verb tenses and aspects in their writing, Hinkel (2004) conducted a comparative study
using academic essays written by 631 NNS students of 8 different mother tongues and 115 NS
students during placement and diagnostic tests at four universities. He set 7 verb tense/aspect
markers (past tense, present tense, future tense, predictive would, progressive aspect, perfect
aspect, and passive voice) and compared the frequency rates of each marker’s occurrences for
each NNS L1group. Then he found some common factors among the NNS groups and compared
it to the use of each marker in the NS students, which led him to conclude that non-trained NS
students write with more accurate and natural use of verb tenses and aspects than their advanced
13

and trained NNS counterparts. From his study, he noticed the following patterns of verb tense
and aspect usage in NNS writers’ academic essays:
1) NNSs used the past tense much more frequently (e.g. in some groups, the frequency
rates were two to three times higher) than any other tenses and aspects. They relied
heavily on narration of personal stories using the past tense in explaining and supporting
their points (p.16).
2a) NNSs rarely made generalizations structured in the present tense; NSs, on the other
hand, extended beyond a past-time recount and “project a sense of general
applicability…and depersonalization” even when using only personal experiences and
opinions to support their arguments (p.16).
2b) NNSs’ use of present tense was limited in its meanings and functions (e.g. broadbased generalizations, self-evident truths, off-topic metaphors, etc.) even when the L1 had
clear morphological tense markers, as noticed in the case of Arabic speakers (p.18).
3) NNSs’ use of future tense was either significantly more frequent than that of NSs (e.g.
Chinese and Indonesian speakers) or significantly less frequent than that of NNs (e.g.
Korean and Japanese speakers). In either case, however, they did not seem to fully
understand the appropriate uses of the future tense, which is often used to predict the
“inevitable, definite outcomes of the future” (p.19). On the contrary, NSs used a balanced
mixture of the present and future tenses depending on the possible outcomes of the future.
4) NNSs, in all L1 groups except for the Arabic speakers, “simply chose to avoid the
predictive would with its syntactic, lexical, and pragmatic complexities” (p.20). Instead,
they usually used the simple present-tense generalizations or used a predictive would in an
incorrect way. On the other hand, NSs frequently developed a set of hypothetical
situations using predictive would.
5) NNSs rarely used the progressive and the perfect aspects (median frequency rates of
0.00 in all L1 groups) and used the simple past tense instead (p.21).
6) NNSs used much fewer passives while NSs often used passive constructions for
“objectivity and detachment” (p.22).
From these findings, he concludes that even the NNS writers who are considered to be
“advanced and trained” from their past language instructions and experiences do not employ
English verb tenses and aspects as accurately and naturally as their NS counterparts. He notices
the general phenomenon of avoidance in NNSs’ uses of complex verb tenses and aspects such as
the perfect aspect, passive voice, or predictive/hypothetical would (p.23), but he does not prove
that the avoidance results from the “complexity of the language structure.” He also attributes the
14

cause of avoidance to the lack of explicit L2 grammar instruction within academic context and
emphasizes the need for overt instruction of the grammar features that are preferred in academic
writing. He also suggests that ESL/EFL students be instructed to expand their “accessible range
of grammatical structures” so that they would not rely only on the “personal narrative,” which is
generally expressed in the simple past tense but would learn to vary their language structures
based on the “fundamental skills” as true intellectuals (p.26).
Influence of the academic writing discourse
Influence of discourse in writing. It is not difficult to notice that academic composition
differs greatly from casual conversation, for there is an “academic register,” which make up the
“language of academic discourse” (Coxhead & Bryd, 2007, p.130). Gray (2011) claimed that
“there is a general consensus, even outside the academic community, that academic writing has
distinct characteristics that set it apart from other types of language” (p.8). According to Gee
(2003), the term Discourse with capital D (as cited in Osborn, 2009, p.22) is “how language is
used ‘on site’ to enact activities and identities” (p.7) and indicates that “we craft what we have to
say to fit the situation or context in which we are communicating” (p.11). Academic discourse, in
this sense, is “the discourse naturalized in the academy” (Batholomae, 1985, 2001, as cited in
Osborn, 2009, p.27) or “the preferred values, discourse conventions, and knowledge content of
the academy” (Canagarajah, 1999, p.147 as cited in Osborn, 2009, p.29).
Despite the variety in wording, there has been a general agreement on the concept of the
academic discourse, and following studies have investigated on how it may influence students’
actual production of academic prose. Shaughnessy (1976, 1977)’s work has implies that college
students’ writing errors are not “random or illogical” but may result from “being beginners in
[academic writing]” and suggests that teachers as well as students be more attentive of the
15

academy’s expectations (as cited in Osborn, 2009, p.32, 33). Similarly, Bartholomae (1985,
2001) has considered students’ “mistakes” as a byproduct of their attempts to “appropriate (or be
appropriated by) the specific discourse” and to “assemble and mimic [the academy’s] language”
(p.273) (as cited in Osborn, 2009, p.33-35). Rose (1989) has also looked at how academic
discourse affects’ students’ writing practice and concluded that “discourse of academics is a
foreign language for students…as well as the linguistic exclusion” (as cited in Osborn, 2009
p.34). More recently, Gray (2011) has conducted a large-scale study based on a corpus of 270
research articles from 56 academic journals, which provides various aspects of linguistic
characteristics.
Conventional uses of English verb tenses and aspects in academic writing discourse:
What academic discourse disciplines suggest for students. In terms of using a language, a
number of studies have acknowledged the importance of the “discourse-level” meaning. CelceMurcia and Larsen-Freeman (1999) dedicated an entire chapter on “The Tense-Aspect-Modality
(TAM) System in Discourse” to show how authentic uses of TAM in discourse may differ from
the sentence-level accounts of TAM (p. 173). They emphasize that “the uses of the TAM forms
can be fully grasped only when we consider their discourse-pragmatic and interactional features
as well as their formal and semantic features” (p.174). In this light, an abundance of research has
investigated specifically on the linguistic features of academic writing discourse. In English in
The Longman Grammar of Spoken and Written English (LGSWE), Biber et al. (1999) describe
the distributions of a full range of lexical and grammatical structures by comparing academic
writing, conversation, newspaper writing, and fiction. As briefly mentioned in the previous
section, Gray’s (2011) corpus-based study also show the variety of grammatical features
including the distribution of tense and aspect marking. According to his data which is based on
16

the frequency of each marking’s use per 1,000 words, the present tense is used most frequently
(except for in history articles), followed by the past tense. He explained that the present tense is
used to “describe the present state or nature of [a construct]…and/or establish relationships and
describe the effects of a phenomenon” while the past tense is used to “establish previous findings
and claims in situating the research study within the canon of knowledge to some extent” (104105). Coxhead & Bryd (2007) also note the grammatical features in academic prose and claim
that simple present tense should be used for verbs in generalizations and theoretical statements
(p.134), which is prevalent in academic papers.
In addition to the research papers and scholarly articles on academic discourse, a great
number of ESL/EFL references/guidebooks on academic writing have mentioned the usage of
English verb tense and aspect expected in academic papers. For example, in his recent book,
English for Writing Research Papers, Wallwork (2011) gives very detailed guidelines on using
verb tense and aspect in academic discourse as below:
-

The general convention of tense usage in Results and Discussions sections is that you use
the present simple, present perfect or past simple to refer to other authors but only the
past simple to refer to your work. The present perfect should not be used to refer to work
that you have carried out. (p.114)

-

The present simple is generally used to begin the Introduction in order to describe the
general background context, i.e. what is known already. The present perfect is then used
to show how the problem has been approached from the past until the present day. At the
end of the Introduction, the present simple is used again when the authors state what they
will do in the rest of their paper (p.203)

-

The past simple is used for the findings you use; the idea is to use the present simple for
what is already accepted in the literature, and the past simple for your new contribution
(p.204)

-

The usage of the future tends to be confined to where authors set out to prove a
hypothesis, rather than to give hard results (p.204)

-

The present simple or present perfect are generally used to introduce the literature
review. Use the present perfect to refer to ongoing situations, i.e. when authors are still
investigating a particular field (p.212).
17

-

The past simple cannot be used when there are adverbials of time; they are typically used
with the present perfect because they indicate something that began in the past (i.e. when
research first began in this area) and continues into the present. They represent
unfinished situations (p.212).

-

The past simple must be used when: (p.212)
• The year of publication is stated within the main sentence (i.e. not just in brackets)
• You mention specific pieces of research (e.g. you talk about initial approaches and
methods that have subsequently probably been abandoned)
• You state the exact date when something was written, proved etc.

-

Use the present simple to discuss previously published laws, theorems, definitions,
proofs, lemmas etc. Such published work is generally considered to be established
knowledge and the use of the present simple reflects this. (p.213)

-

Most Methods sections are written in the past simple using the passive form. The past
simple is required because the actions you describe took place in the past (i.e. before you
started to write your paper). The past simple also helps to distinguish what you did from
what others have done (which is often described in the present simple) (p.220).

-

Many tenses and constructions are used in the Conclusions – the future, conditionals,
modal verbs, etc. (p.268)

-

In conclusions, one distinction that many authors make is between what they did during
the research (simple past) and what they did during the writing process of the manuscript
(present perfect) (p.268).
Problems of applying academic writing discourse for NNS writers. While some NNS

writers may successfully learn to use English verb tense and aspect in academic papers by
referring to the findings of the literature and to the guidelines offered by books on academic
writing discourse, some would still struggle in the actual writing process. Even if what academic
writing discourse recognizes as the “conventional uses of English verb tense and aspect” is
accurate, it could be challenging for the L2 writers to apply theory into practice. In fact, there are
some possible aspects of academic writing discourse that may hinder NNSs’ language
improvement.

18

Smit (2004) implies that academic discourse as a genre may not be practically useful
after the students leave campus as the practitioners in various fields (as cited in Osborn, 2009,
p.27). Moreover, Rose (1989) studied the effects of remedial writing courses that focus on
teaching academic writing discourse to college students (as cited in Osborn, 2009). Although he
only analyzed remedial writing, he gives insights to consider limitations of academic writing
discourse by arguing that such focus on academic discourse may be “limiting growth in writing”
in five ways: (as cited in Osborn, 2009, p.35)
a) The writing is self-contained, with little connection to the larger academic writing
environment
b) Topics are not conducive to the production of academic prose
c) Vigilance for error restricts students’ composing process
d) Processes of reading and thinking are separated from writing
e) The expressive and exploratory possibilities of academic writing are limited
In addition, Hantrakul (1990) indicates that “NNS writers tend to base their choice of
verb forms on the sentence level but focus on the linguistic factors in the immediate context, the
local level” (p.137). This finding implies that NNS writers may possibly decide which verb
structure to use only by referring to any one of the academic writing disciplines which they
learned or heard about. In other words, because academic writing discourse often seems to offer
what looks like “straightforward rules” for using verb tenses and aspects, NNS writers who are
supposedly without intuition are likely to make “sentence-level” choices even when discourselevel grammar knowledge has been emphasized., In particular, when academic writing discourse
does not give very detailed information on using certain verb structure, there is a high chance
that students may apply any one of the academic writing disciplines; this may explain why
neither use of the correct form (Hantrakul, 1990) nor non-use of the incorrect form (Hinkel,
19

2004) of certain verb structures would directly mean that students know how to use that verb
form.
Although most of the references on academic discourse guide students to use the simple
past when referring to a study done by other people, not many of them provide detailed
explanations and/or examples considering the specific context but usually resort to
generalizations. For example, compare how references on academic writing discourse usually
make generalizations about using the simple present/past tense and how a sample excerpt from
Wallwork (2011) indicates that it can be not as simple as a pre-set mathematical formula:
a) Common generalizations:
“Use the simple past for the findings you use” (Wallwork, 2011, p.204)
“[Simple past tense is used to] establish previous findings and claims in situating the
research study within the canon of knowledge to some extent” (Gray, 2009, p.104)
“In general, the simple past is the ‘normal’ one for [referring to] the past events; we use
it if we do not have a special reason for using one of the other tenses” (Swan, 2005,
p.421)
b) Wallwork’s (2011) sample (p.212):
Sample Sentence: Smith and Jones ([1990]) developed a new system of comparison. In
their system two languages are / were compared from the point of view of … They
found that ….
Explanations: The first verb introduces the author and is typically used in the past
simple. Note that the first verb could also be in the present simple. However, generally
when the present simple is used the construction is slightly different: first the reference
and then the author. In the second verb, both past simple and present simple are
possible. However, it is common to use the present simple when describing a system,
method, procedure etc. functions. Present simple underlines that Smith and Jones are
still using their system and that it is still valid. The use of the past simple (were
compared) would probably imply that Smith and Jones’ system is not in use anymore
and it was just a step in this road of research that has subsequently been superseded.
The third verb (found) indicates what the author managed to do, and typically such
verbs are used in the past simple. Again, however, some authors use the present simple
in such cases.
20

From these samples, it is possible to infer that academic discourse disciplines may not be
providing information on verb tense and aspect usage in writing that is detailed and
contextualized enough for NNS students to apply properly. In other words, the complexity of
verb tense and aspect usage revealed in the sample (b) reiterates the significance of discourselevel analysis and the need for understanding each verb form’s meaning and functions, which not
so many references on academic writing discourse offer to provide.
Furthermore, the existing references and guidebooks on academic writing seem to be
addressing either limited range or meaning of the verb tense and aspect combinations compared
to what descriptive grammar books actually informs of each verb form (Celce-Murcia & LarsenFreeman, 1999; Hewings, 1999; Azar, 1999; Swan, 2005; Coxhead & Bryd, 2007; Gray, 2011;
Wallwork, 2011). First, only limited range of verb tense and aspect combination usage appears to
be guided by the academic discourse references. For instance, Hinkel (2004) cites from Quirk et
al. (1985) that “unlike generalizations that are marked by a predominant use of the present tense,
and narratives of past-time experiences, in academic writing, the future tense is employed only
occasionally” (p.19). Even in one of the most detailed references, Wallwork (2011) simply
mentions that “Many tenses and constructions are used in the Conclusions – the future,
conditionals, and modal verbs” without any further discussion (p.268). However, Celce-Murcia
& Larsen-Freeman (1999) argues that despite the gaps in the frequency of uses among the tenseaspect combinations, it is not fair to conclude that less frequency of usage directly implies less
significance (p.111). Also, many academic discourse references seem to be referring to only
limited meanings, functions, and usages of the verb tense and aspect combinations. Specifically,
the usage of perfect aspect is generalized in most academic discourse compared to what
descriptive grammar informs of its various meanings, functions, and usages as below:
21

a) What is generally said of using the perfect aspect:
- The present perfect tense indicates something that began in thepast (i.e. when
research first began in this area) and continues into the present.
- The present perfect is used to show how the problem has been approached from the
past until the present day (Wallwork, 2011, p.203, 212).
b) Other ways to explain the perfect aspect:
-The core meaning of the perfect is “prior,” and it is used in relation to some other
point in time (Celce-Murcia & Larsen-Freeman, 1999)
- The perfect is used for an indefinite query (Inoue, 1979)
- The function of the present perfect is to change the nature of the relationship
between subject and predicate—it emphasizes the predicated event’s result on the
grammatical subject (Knowles, 1976)
-A speaker of English uses a sentence in the present perfect when the information he
is giving appropriately exemplifies or explains the topic of discourse (Inoue, 1979, p.
585)
By comparing the two sets of information on the use of perfect tense/aspect, it is possible to
notice that only a partial usage of the perfect tense has been emphasized in the academic writing
discourse whereas it has multiple and distinctive meanings and functions.
Rationale of the study
The above literature has reviewed the topics of L2 grammar teaching, error analysis,
avoidance, and academic writing discourse in relation to the patterns of English verb tense and
aspect employed in non-native speaker students’ academic writing production. While some
previous studies have investigated NNS writers’ usage patterns of English TAM, not many have
questioned why they would show such tendency to overuse, misuse, or avoid certain verb
structures in their writing. Instead, many have jumped into the conclusion where the necessity of
grammar teaching was emphasized. However, several findings on the features of NNS’ use of
verb tenses and aspect remain questionable in terms of making generalizations. Moreover, in
22

addition to providing more evidence for the necessity of grammar teaching in ESL courses, it
would be critical and beneficial to specifically designate which topics of English grammar—in
this case, which verb structures among various English tenses and aspects—should be taught in
university ESL writing courses.
First, this study aims to find detailed patterns of L2 writers’ actual usage of English verb
tenses and aspects in the academic discourse. It is without doubt that L2 learners’ production
errors have revealed much about the problems and difficulties they have in writing. Error
analysis has proved that verb form and usage is one of the most problematic areas in English
(Amsberg, 1984; Ubol, 1981; Kim, 1983; Meziani, 1984; Bryant, 1984; Richards, 1985;
Hantrakul, 1990; Pongsiriwet, 2001) and that it significantly affects L2 writing (Johansson,
1978; Khalil, 1985; Sheorey, 1986; Santos, 1987; Tomiyana, 1980, as cited in Roberts &
Cimasko, 2008; Arani, 1993; El-Sayed, 1982; Ghadessy, 1980; Kroll, 1990; Santos, 1988; Scott
& Tucker, 1974; Yang, 1994). However, simply analyzing L2 production errors may provide
only partial information of the learners’ ability to understand and use the target language because
the learners may choose to avoid using certain language features (Schachter, 1974). In this light,
Hinkel (2004) analyzed NNS writers’ patterns of using verb tense and aspect in formal academic
writing by comparing the frequency rates of each tense/aspect marker of NNS and NS writers.
His study has shown how NNS writers frequently avoid using “syntactically and semantically
complex” verb structures such as the predictive would, the perfect aspect, the progressive aspect,
and the passive voice (p.20, 22, 23), which could make their writing less natural and academic.
Although his research provides insights about how NNS writer’s avoidance affects their
performance in academic discourse and thereby proposes more in-depth grammar teaching, he
does not prove that their non-use of certain verb forms is due to the lack of knowledge or the
23

lack of instructions on the “complex” verb structures. In addition, he does not compare NNS
writers in terms of their level of language proficiency, so it is even more difficult to verify
whether more instructions would actually prevent their tendency to avoid what he considers as
“more difficult” verb structures. Thus, the current research intends to explore whether less
advanced students show more tendency to avoid using more complex structures (e.g. perfect
aspects, predictive modality, etc.) by comparing the verb usages of NNS of different language
levels instead of comparing their usage patterns to those of NS writers, who are likely to have
received a completely different instruction on English grammar.
Second, based on the findings about NNS writers’ patterns of using and avoiding certain
verb structures, this study will propose other factors that may significantly affect the
phenomenon. In other words, if a positive relationship is clearly observed between the students’
language proficiency and the range or variety of verb structures they employ in their writing, it
would be logical to conclude that conspicuously low frequency rate of certain verb tense and
aspect usage reflects the phenomenon of avoidance. It would also indicate that students in higher
level courses are less likely to avoid using any verb structures regardless of their complexity.
However, if there is little or no relationship observed between the students’ language levels and
their verb usages—that is, if all of the students across language proficiency levels show a similar
pattern in using or avoiding certain verb structures—a different factor should be taken into
consideration. In this case, more attention should be paid to the context in which each verb tense
and aspect structure is being employed, which will then eventually result in discussing the
influence of the academic discourse. Academic discourse has been claimed to significantly
influence university students’ writing (Shaughnessy, 1976, 1977; Bartholomae, 1985, 1977;
Rose, 1989; Gee, 2003; Coxhead & Bryd, 2007; Gray, 2011), and a great number of references
24

and textbooks have provided specific guidelines on how to use verb tenses and aspects in
academic writing. Taking this abundance of references to which ESL/EFL students must have
had access, it would be meaningful to closely investigate how such disciplines and conventions
of academic discourse may have affected, or even limited, their verb using patterns.
Finally, as a summative conclusion, this study intends to make meaningful implications
applicable to actual teaching and testing of English writing in university ESL courses. The
necessity of teaching English verb tense and aspect should first be verified. More importantly,
however, suggestions should be made on which specific verb tense and aspect structures to be
taught and how and when they should be implemented to the writing courses, considering the
level of courses and the types of language the students need to use. Furthermore, the English
Placement Test, which is currently used as the only assessment tool for deciding every NNS
student’s ESL course placement at UIUC, may benefit from the findings of this research. If
necessary, more suggestions will be made for modifications or addition to the current
assessment.
Research Questions
1. What are some patterns of NNS students’ verb tense and aspect usage observed from the
English Placement Test essays? Does NNS usage of verb tense and aspect have a relationship
with their L2 writing proficiency?
- Is there any salient difference in these patterns among the three groups (ESL 500, ESL 501, and
Exempt group) in terms of their use and non-use?
2. Does academic writing discourse influence NNS writers’ use and non-use of verb tense and
aspect?
- How does academic writing discourse make difference in NNS writers’ verb usage?
3. Is it necessary to teach English verb tense and aspect system in university ESL writing
courses? What benefits will it have in improving NNS students’ writing proficiency in academic
setting?
25

4. What are the implications/suggestions for future grammar teaching in ESL/EFL courses?
Which verb tense and aspect structures in particular need specific instructions? How should these
grammar teaching be implemented to the ESL writing courses? What are the implications and
suggestions for the development of the EPT written exam?

26

CHAPTER 3
METHODOLOGY
Data Collection
To analyze NNSs’ patterns of using English verb tense and aspect in academic prose, 120
de-identified essays written by non-native speakers of English were retrieved from the English
Placement Test database at University of Illinois-Urbana Champaign. All the test takers were
degree-seeking graduate students (either in the M.A. or PhD. degree programs), and they were
obligated to take this test based on their TOEFL or GRE scores submitted for university
admission. The task was to write an argumentative essay based on a lecture and a reading
passage on a topic assigned for the day. The students were given chances to discuss the topic in
small groups, take notes, write an outlined draft (20 minutes), and have a peer-review workshop
before actually writing the final draft (60 minutes). Then each essay was rated by two ESL
writing instructors who were trained for rating and recalibration, and if there was any
disagreement between the two, a third rater would rate and recalibrate to make the final decision.
Finally, based on a holistic rubric, each student was placed in either ESL 500 (lower level) or
ESL 501 (higher level), and those who excelled in the exam were exempted from taking any ESL
writing courses.
In this study, data was collected only from the essays written by students whose L1 was
either Korean or Chinese, in order to decrease potential variable caused by the L1 influence on
English verb usage patterns. 30 essays written by Chinese speakers and 10 essays written by
Korean speakers were collected from both ESL 500 and ESL 501 groups, and 35 and 5 essays
were collected from Chinese and Korean speakers respectively. Every essay was provided by the
EPT RA in de-identified format, that is, without any student identification information (e.g.
27

name, ID number, etc.) from the EPT database. For the sake of clarifying and simplifying the
identification of samples retrieved and used for this study, every sample sentence was labeled as
“(Assigned course_K/C Order of the essay by page number_A/D).” The abbreviation of K/C
indicates the student’s L1, Korean or Chinese, because the essays were grouped and numbered
separately by the EPT RA. A/D shows whether there was any discrepancy between the first two
raters; that is, A indicates that the two raters agreed on the placement decision, and D indicates
that there was a third rater for the prior disagreement between the two raters. For example, a
sample sentence retrieved from the Chinese student who was assigned to ESL 500 without any
discrepancy between the two raters, and whose essay started from page 31 of the collected data
would be labeled in parenthesis as:
S1: sample text (ESL500_C31_A)
Data Analysis
For this study, a taxonomy was created for the English verb tense and aspect usage of
NNS students in academic discourse based on two inflected tenses (present and past) and four
aspects (simple, perfect, progressive, and predictive) as organized in the table 1 below. Each
essay was analyzed in terms of these 10 combinations of tense and aspect; however, instead of
merely counting the raw numbers of the total use, every time a student used a verb tense and
aspect combination (the term “verb structure” or “verb form” will be used interchangeably in this
paper), it was analyzed and categorized in one of the three groups: “correct use” or “should have
not been used/ misused” or “should have been used/ avoided.” Consequently, for each “should
not have been used” case, what would have been the correct use was suggested in the last column
of the table as the “should have been used.” However, the total numbers of “should not have
been used/ misused” cases were larger than those of “should have been used/avoided” cases
28

because some verb forms required completely different verb forms such as gerunds, to-infinitive,
and participles, which could not be revised in terms of the tense and aspect system.
In order to increase the usability and benefits of the current study, both raw frequency
counts provided by Tables 2-4 and error analysis of actual student writing samples were closely
analyzed. In other words, for each essay, every verb tense and aspect use was analyzed as one of
the 10 categories, and the total numbers of each category were counted and written down in the
taxonomy table. When all 40 essays had their own taxonomy tables in each group (ESL 500,
ESL 501, and Exempt), they were again combined and computed for the total number of uses
and the average number of uses. This frequency count method was used so as to see if each
group showed a distinctive feature in the frequency of verb tense and aspect uses and if there
were any salient differences among the groups. However, more important than the analysis of the
total numbers provided in the tables were the actual samples of misused and avoided cases of
certain verb tense and aspect combinations in all 120 essays, which accounted for the phenomena
shown in the tables. More specifically, in presenting the results of the analyses of the samples,
general explanations are provided in the result section as “Findings” for each aspectual category
of the taxonomy; these “Findings” sections include presenting the distinctive patterns of the verb
usage in each group to address the relationship between ESL students’ language proficiency and
their patterns of using verb tense and aspect. They also include any phenomena that are similar
or common among the three levels of students in order to discuss the role that academic writing
discourse may play in NNS writing. Furthermore, in addition to these general findings, more
specific explanations or clarifications are provided for some of the sample texts in order to
clarify the results and to emphasize their significance in showing the results—these specific

29

“Clarification” sections are not obligatory and are presented right below the corresponding
sample texts only when necessary.
Taxonomy of English verb tense and aspect combinations. Prior to analyzing the data,
it would be necessary to explain the taxonomy of English verb tense and aspect combinations
invented and used for the current study.
Table 1
Taxonomy of verb tense/aspect usage in NNS academic writing
Aspect

Tense

Simple

present
past
present
past
present
past
will
would
may/might
should/can/could/
have to/must

Perfect
progressive
Predictive

Etc.

Used in place
(correct use)

Shouldn’t have been
used (overuse/misuse)

Should have been
used
(avoidance/non-use)

The English verb tense, aspect, and modality (as many have referred to TAM) system has
been described by the previous literature in multiple ways; among these attempts to establish a
comprehensive yet simple picture of the TAM, various taxonomies of tenses and aspects have
been proposed for the learners and teachers of English. Celce-Murcia and Larsen-Freeman
(1999) in their The Grammar Book has discussed the tense-aspect system by using a taxonomy
that groups tenses in terms of its forms, characteristics, semantics, and usages. Unlike many
other descriptions that simply categorize tenses by time (present, past, future), they assert that
understanding the meaning of the tenses (tense and aspect combinations) in time-frame is
“inadequate” and needs “a deeper level of abstraction than that of temporal meaning” (p.112).
30

In line with their argument, the tense and aspect system was organized primarily in terms
of the aspectual frame in the current study, which would determine and explain the internal
nature of the action or phenomenon that may apply to any time using the table above. However,
all future markers were removed from the three conventional (simple, perfect, and progressive)
aspects and were embedded into the “predictive” aspect, for the future time cannot be expressed
in verb inflections and is closely related to making future predictions (Celce-Murcia & LarsenFreeman, 1999; Biber et al., 1999). Therefore, the taxonomy used in this study invented a new
aspect that deals with ways to express future predictions which vary in the degree of certainty
and probability; however, modals used and analyzed for this purpose were limited to will, would,
may/might, to consider only the extrinsic function of modality, which is related to making
predictions in the future (Biber et al., 1999). Cases of using modals and semi-modals used to
express necessity (e.g. can, could, should, must, have to, ought to, etc.) were also counted
because they were expected to be used frequently in argumentative essays; however, the usage of
these modals was labeled as “Etc.” in the taxonomy and was not analyzed in depth because they
added the meaning of necessity or obligation to genuine “predictive” purpose of the future.
Further, some tense and aspect combinations were not implemented in the taxonomy due to their
rare usage in academic discourse; for instance, perfect progressive aspect was excluded, for it
was reported to be used less than 0.5% in all registers (Biber et al., 1999). Below is the brief
summary of the four aspects’ core meanings, functions, and usages, explained by previous
research and existing grammar textbooks; these explanations provided the basic guidelines to
analyze and categorize students’ verb usage patterns in the study.


Simple aspect

31

The core implications of the simple aspect are its “complete or unchanging nature” (Hirtle,
1967) and “factuality” (Celce-Murcia & Larsen-Freeman, 1999, p.112-113).


Perfect aspect

One of the most problematic areas, and supposedly avoided usages of English verb tense
would be the perfect aspect because it does not necessarily mean and function in one single
time frame. In fact, the core meaning of the perfect is “prior,” and it is used in relation to
some other point in time (Celce-Murcia & Larsen-Freeman, 1999). It has various effects to
the action or state (e.g. “current-relevance”, “extended now”, “up-to-now effect”, “recent
completion”, “indefiniteness”, etc.).


Progressive aspect:

The main point of progressive aspect is being “imperfective, incomplete, limited, and/or
temporary” (Celce-Murcia & Larsen-Freeman, 1999, p.116). It generally indicates an action
in progress, temporary and specific situations, and ongoing states.


Predictive aspect:

This usage of English verb tense and aspect is notorious for being confusing and difficult for
NNS students to learn and for teachers to explain not only because it is complicated in its
nature but also because linguists and grammarians have defined, categorized, and described
them in such diverse ways. Thus, it will be critical that this study sets a clear taxonomy of
this aspect in terms of its meanings and functions prior to analyzing the data.
First of all, as briefly mentioned above, this paper includes only partial functions of English
modals. According to Hofmann (1996) and Palmer (1990), modals have at least two
distinctly different functions expressing either “logical probability (epistemic or extrinsic)”
or “social interaction (deontic or intrinsic)” (as cited in Celce-Murcia & Larsen-Freeman,
1999, p.141).Only the former function of modals will be discussed in this study as it has
great relevance to making predictions. The usages of will, would, may/might will be
scrutinized in terms of the degree of certainty in predicting possible results.
In addition to the unconventional inclusion of future tense in this aspect, future conditional
aspect will be taken into account as it is traditionally referred to as “predictive aspect” in
contrast to the factual and imaginative conditional aspects (Celce-Murcia & LarsenFreeman, 1999, p. 545-553). Although few are aware of its complexity in semantic
variations, what are simply referred to as “conditional sentences” can be divided into three
types in terms of its semantic relationships: factual conditional, future (predictive)
conditional, and imaginative (subjunctive) conditional relationships (See figures 27. 1 and
29. 2 in p.548, 552). Unlike how imaginative conditional sentences imply the “un-realness”
32

or “counterfactual” conditions of the current events or states, predictive conditional
sentences express “possible or potential” future plans and outcomes.
Therefore, this study will include predictive conditional as a part of the predictive aspect, not
as a distinctive category and investigate how different modal auxiliaries are used in the main
clause to indicate the degree of assertiveness or probability of the future result. Furthermore,
in this paper, the use of will, would, and may/might will be treated with particular
significance in relation to the hedging strategy, which is often considered to have an
important role in the academic writing discourse.

33

CHAPTER 4
RESULTS
Below are the tables (Tables 2-4) that show how frequently each verb tense and aspect
combination was correctly used, overused or misused, and avoided in each group.
Table 2
Total Frequency Counts of Verb Usage in ESL 500 Group
Aspect

Tense

Simple

present
past
present
past
present
past
will
would
may/might
should/can/could
/have to/must

Perfect
progressive
Predictive
Etc.
Total

Used in place
(correct use)

Shouldn’t have been
used
(overuse/ misuse)

Should have been
used
(avoidance/non-use)

897 (63.9%)
10 (0.7%)
34 (2.4%)
0 (0%)
10 (0.7%)
0 (0%)
115 (8.2%)
49 (3.5%)
79 (5.6%)
208 (14.8%)

81(54.7%)
36 (24.3%)
3 (2.0%)
0 (0%)
5 (3.4%)
0 (0 %)
9 (6.08%)
13 (8.8%)
0 (0%)
1(0.7%)

32 (25%)
8 (6.2%)
18 (14.0%)
0 (0%)
7 (5.5%)
0 (0%)
39 (30.5%)
17 (13.3%)
0 (0%)
7 (5.5%)

1402

148

128

Table 3
Total Frequency Counts of Verb Usage in ESL 501 Group
Aspect

Tense

Simple

present
past
present
past
present
past
will
would
may/might
should/can/coul
d/have to/must

Perfect
progressive
Predictive

Etc.
Total

Used in place
(correct use)

Shouldn’t have been
used
(overuse/ misuse)

Should have been
used
(avoidance/non-use)

957 (59.6%)
31 (1.9%)
49 (3.0%)
0 (0%)
15 (0.9%)
0 (0%)
174 (10.8%)
60 (3.7%)
103 (6.4%)
215 (13.4%)

94 (56.9%)
42 (25.4%)
6 (3.6%)
0 (0%)
6 (3.6%)
1 (0%)
7 (4.2%)
5 (3.0%)
3 (1.8%)
1(0%)

37 (26.6%)
16 (11.5%)
20 (14.4%)
2 (1.4%)
6 (4.3%)
0 (0%)
25 (17.9%)
26 (18.7%)
4 (2.9%)
3 (2.2%)

1604

165

139

34

Table 4
Total Frequency Counts of Verb Usage in Exempt Group
Aspect

Tense

Simple

present
past
present*
past
present
past
will
would
may/might
should/can/could/
have to/must

Perfect
progressive
Predictive

Etc.
Total

Used in place
(correct use)
897 (58.4%)
63 (4.1%)
88 (5.7%
0 (0%)
23 (1.5%)
1(0%)
89 (5.8%)
30 (1.9%)
62 (4.0%)
282 (18.4%)

Shouldn’t have been
used
(overuse/ misuse)
74 (46.3%)
75 (46.9%)
2 (1.3%)
0 (0%)
4 (2.5%)
0 (0%)
5 (3.1%)
0 (0%)
0 (0%)
0 (0%)

Should have been
used
(avoidance/non-use)
45 (29.0%)
19 (12.3%)
35 (22.6%)
5 (3.2%)
9 (5.8%)
0 (0%)
30 (19.4%)
6 (3.8%)
4 (2.6%)
2 (1.3%)

1535

160

155

Note. There is discrepancy between the total numbers of overuse/misuse cases and those of avoidance
cases in each group of students because some of the “misuses” did not have corresponding tense/aspect in
the “should have been used” column; they were merely “ungrammatical” uses of the verbs which needed
to be revised into other verb structures (e.g. verb clauses, participles, gerunds, voice change, etc.) not
categorized in this taxonomy.

While it is worthwhile to see the raw frequency counts of each verb use and to compare
the numbers in the three groups, the numbers themselves are preliminary for analyzing the
patterns of NNS usage of verb tense and aspect without actual samples. Therefore, the tables
should not be over-interpreted but used as a complementary data when scrutinizing the
phenomena case by case with the actual student samples because only then the numbers counted
for each use would make sense. The following findings provide in-depth analyses of NNS
students’ use of verb tense and aspect in terms of not only the frequency of each verb form but
also of the factors attributable for the cases of overuse/misuse and avoidance. The findings are
presented in terms of the categories in which verb usage patterns were classified and analyzed,
and they include the discourse-level analyses as well as. An asterisk (*) before a verb indicates

35

that it is ungrammatical, and a question mark (?) indicates that the use of the verb is questionable
and/or awkward to be grammatically correct.
Findings on the Simple Aspect
Table 5
Frequency Rate of Usage Patterns for the Simple Aspect
Rate of total use
(of the total verb combinations)

ESL500
66.0%
63.0%
(present)

Rate of correct use

21.7%
(past)

4.1%
(past)

91.0%
(present)

42.5%
(past)

12.1%

78.2%
(past)

31.2%
25.0%
(present)

59.4%
(present)
87.5%

11.4%
8.3%
(present)

Rate of avoidance/non-use
(of the total avoidance cases)

2.9%
(past)

88.6%
98.9%
(present)

Rate of overuse

ESL501
63.5%

8.9%
(present)

57.5%
(past)

38.1%
6.2%
(past)

26.7%
(present)

11.5%
(past)

Exempt
65.4%
57.3%
(present)

8.1%
(past)

86.6%
92.4%
(present)

45.6%
(past)

13.4%
7.6%
(present)

54.3%
(past)

41.2%
29.0%
(present)

12.2%
(past)

As the total numbers and rates reveal in the tables (Table 2-5), the simple aspect, simple
present in particular, was the most frequently used by NNS students in all levels (over 60% in
every group), which is consistent with the findings of previous literature including Biber’s
(1999) corpus-based research. In order to investigate the reasons for such phenomenon, the
current study focused on four functions and/or usages of the simple aspect to account for its high
frequency. Then, the findings were further analyzed in terms of the two major factors relating to
the research questions: (a) the relationship between the students’ verb usage patterns and their
language proficiency and (b) the role of academic discourse in students’ use of the simple aspect.
36

Celce-Murcia and Larsen-Freeman (1999) has classified the core meanings of the simple
present in 8 categories and the simple past in 6. Although a couple of the usages of the simple
present are limited to conversational context (e.g. present speech acts & conversational historical
present), the rest are regularly and frequently employed in all circumstances. One of the most
likely reasons why the simple aspect, simple present in particular, dominated every essay written
by the NNS in this study could be due to its multiple roles and the difficulty of applying these
conceptual roles to actual production. In other words, because the simple aspect has diverse
meanings which consequently result in multiple functions and usages, it seemed inevitable that
students used this verb structure most often. However, raw number counts from the Table 2-4
revealed that students in all groups overused the simple present most frequently as well, which
could imply that students were more likely to use the simple present as a substitute for other verb
forms that are less familiar or more difficult.
It was also striking to find out that students in lower levels actually produced more
correct uses of the simple present as shown in the Table 5, which may suggest that students’
knowledge on using the verb tense and aspect is irrelevant to their writing proficiency. However,
the analyses of students’ actual writing samples suggested such speculation to be misleading.
Because the simple present has multiple functions that are distinct in their meanings, there was a
high chance that students with less knowledge on verb tense and aspect choose one simplest verb
form and use it everywhere. Therefore, it was critical that the actual uses of simple aspect, the
simple present in particular, be scrutinized in terms of the functional categories in order to see
how students in different levels used this verb form respectively.
The simple aspect used for general facts or universal truths. One function widely
expressed by the simple aspect is presenting the general facts or universal truths. It is obligatory
37

to use this structure due to the implication of completeness, factuality, and unchanging nature
that the statements convey as in S1-S6.
S1: Approximately 60 to 80 percent of processed foods in a
typical American grocery store have GM ingredients.
(501_C29_A)
S2: It is a natural tendency to feel the fear about unknowing
and unexperience. When they know about their enemy,
however, they doesn’t feel a fear. (500_K15_A)
S3: In certain countries such as China, the test score is
extremely critical to get enrolled into a good college. And
graduate from a good college always have a good job.
(501_C59_D)
S4: Technology sharing, capital movements, and the movement
of people are factors that contribute to the development of
global economy and products of globalization…For instance,
nearly 340 companies of world’s top 500 have their own
branches in China. (Exempt_C25_A)
S5: Schools play a significant role in establishing students’
identity, personality, and social skills which influence on
a student’s life more than academic success of a student.
(500_K13_A)
S6: Consumers have decent rights to know about all the
ingredients of the product that they will take. (500_K17_A)
For this function of the simple aspect, students seemed to show the tendency to overuse
the simple present regardless of their language proficiency levels. In all three groups, students
misused the simple present even for statements that could not be taken as facts or general truths.
This problem was also closely related to the phenomenon of avoiding the use of the modals will
and would of the predictive aspect, leaving no room for possible uncertainty or hedging their
38

assertions. It also applied to the cases where uncertainty of the event was evidently proposed
with the conditional “if-clause” or time adverbials, which will be closely discussed as one of the
four functions in the later part of this section.
In brief, no clear distinction could be presented among the three groups in using the
simple present instead of modality, which would mitigate their assertiveness and enable them to
hedge as in S7-15. Even essays written by the advanced students in Exempt group showed
random cases of overusing the simple present when proposing arguments or predictions as
“factual statements” even when they lacked reasonable evidence to support the certainty of the
statements as in S13-15 while sometimes they did make proper use of modals as in S16. Thus, it
would be reasonable to conclude that non-native students’ tendency of “not hedging” and
“making strong arguments or predictions” contributed to their overusing simple present, which
could then be closely related to their lack of understanding academic discourse disciples.
S7: Some companies may claim that both labeling and
advertisements cost a lot, but if they can refresh the
recognitions of the consumers on the products by these
methods, the cost *becomes trivial. (500_K4_D)
S8: However, if the reforms of teacher pay implement at
school, they *start to compete, not working together.
(500_K19_A)
S9: The American *does not worry about food quality / safety,
which *leads them to be notorious for its high obesity
level and bad consumption of food. (500_K17_A)
S10: First of all, it ?is difficult to measure the so-called
performance…it is likely that the performance-based pay
system is not practical. Furthermore, [it] will lead to
less collaboration and more competition among teachers.
Every teacher ?wants to become the best among his or her
39

colleagues in order to get extra reward. [They] ?tend to
deceive in order to get extra reward. There are cases that
teachers correct answers for their students after the
exams, so the scores of their students ?is better and they
could be rewarded by the performance-based pay system.
(500_C35_A)
S11: GMO labeling ?is more beneficial for economy
eventually. Let me think about disadvantages of unlabeling.
Stress has a tremendous problem in our lives. So, when
people worry about how many percent of GM food, they should
feel anxieties. It ?arouse their stress level. As a result,
they ?are not only get an illness, but also ?reduce
efficiency in their work. Furthermore, the cost of
labeling ?is not important. You should think about the
benefit that reduced the cost by using GMO product and the
benefit of making a good image by unusing GMO product.
Eventually, it ?is economical benefits to both. (500_K15_A)
S12: First, labeling GM foods can trigger people to increase
anxieties and to mislead. Some people said that labeling GM
foods should be necessary. But by labeling GM foods, people
may think there is something problems with GM
foods…Labeling GM foods *lead people to increase anxieties
because people are not biological experts and researchers.
They *think that labeling GM foods means GM foods ?may do
not intake because of lack of knowledge. Only labeling GM
foods *makes societies or people to be unstable. Therefore,
labeling GM food *create people make wrong decisions.
(501_K3_A)
Note. In this case, most of the verbs marked with (*) should be used in the predictive
aspect because the first part of the paragraph clearly indicates that the writer is predicting
what will happen in the future; there is no proven evidence strong enough to make such
assertions with the simple present.
S13: In the first place, globalization ?is an opportunity to
the economy growth…In addition, globalization ?shows people
a better life condition because people ?have more money to
buy food, pay for their entertainment and medical service,
which ?result in life expectancy increasing…In the last,
40

globalization ?makes benefits to the spread of culture,
knowledge, and technology….To sum up, although
globalization might have a few shortcomings, people ?gain
much more benefits from it than loss. More workers find
jobs, price of food and grocery are cutting down,
people ?live longer and now technology can spread to
different countries, which ?proves globalization ?bring a
better life for us and really ?make the world a whole
family. (Exempt_C11_D)
S14: Although animal testing seems to be unethical, it ?is in
effect, inevitable to human beings. Apparently, not only
human beings but also animals *benefit from the results of
animal testing. The results of testing can apply to animals
themselves, which *reduces the risks of flu spreading among
animals. On the other hand, no alternatives ?are found to
study medical effect in research. Given that preventative
vaccines are created and tested on human beings, it
*suffers severe risks of killing the human beings before
the success. Scientists as a human being would not do this
to violate the law. (Exempt_C1_A)
S15: I have to say that strict regulations ?are not enough
and even *doesn’t work in most circumstances. Scientists
and doctors will decide that in what situation people can
conduct animal testing because animal protectors ?don’t
have the necessary knowledge. (Exempt_C59_A)
S16: Actually, unemployment in the industrial countries has
not increased systematically over the last 10 to 20
globalization years…Another important influence
globalization has brought about is capital movement and
work flow…Overall poverty in the Third World has
diminished. In other words, globalization makes the rich
richer and the poor poorer. (Exempt_C47_A)
The grammatical function in subordinate clauses of time or condition. Another
function of the simple aspect, the simple present in particular, which influenced its high
frequency rate in NNS essays is the grammatical function in the subordinate clauses of time or
41

condition when the main clause is making a future prediction as in the “if-clause” and time
clauses. Even if the subordinate clause might sound like making reference to the future,
conditional statements in English require the simple present to be used in the subordinate clause
when the entire sentence is referring to the “predictive conditional” as opposed to the
“hypothetical or unreal conditional” which requires the simple past in the subordinate clause as
in S1-5. This distinction between “possible” conditionals with reference time in the future and
“unreal” conditionals with reference time in the present was problematic in some cases,
especially in lower groups of students (S6-S7), but there were only few cases of them.
Otherwise, for this grammatical usage obligatory in the subordinate clause, most students in all
three groups seemed to have fairly good understanding, and it just increased the total number of
simple present being used.
However, students seemed to have more difficulty in using the right verb tense and aspect
for the main clause, which may overlap with the previous cases where students overused the
simple present for potentially uncertain predictions. In fact, when students used conditional
statements with “if-clause,” the uncertainty of the prediction became more explicit because the
prediction would be achieved only under certain conditions, but they failed to use the future
modality in the main clause. Instead, they used the simple present even when time adverbials and
if-clause confirmed the uncertainty of their predictions as discussed in the previous section.
S1: If we start the GM labeling, it will increase the price
of the foods and damage the productivity of the foods.
(500_K9_A)
S2: If after labeling the price is not profitable, then there
will be little necessary for GM food to exist.(501_C21_A)
S3: If a child is well-educated at home, he will behave well
42

at school and get a high score. (Exempt_C15_A)
S4: As time goes by, this technology will grow and become
mature enough to be used. (Exempt_C27_A)
S5: Let us imagine what will happen if advertising using
young beautiful girls is shown to a critical thinking
housewife who usually focus on the features, prices, and
promotions of the advertisement. And on the other hand, an
artist may become confused if an advertisement contains
only the numbers indicating the features of the good.
(Exempt_C52_A)
S6: If there *were financial rewardings, the teachers would
become more competitive so that they must work harder to
improve their own skills to get the bonus. (500_C53_A)
S7: If the experiment *showed that the policy has beneficial
effects on improving the test scores, it may also benefit
other kinds of student performance as well. (500_C53_A)
The simple aspect used to indicate states. The simple aspect is also used to describe
states as opposed to actions. This function cannot be expressed in other aspects (e.g. progressive
or perfect aspect) because progressive aspect is used for “limited action in progress” and perfect
aspect implies “continuation of an event that allows room for change” (Celce-Murcia & LarsenFreeman, 1999, p.124-126). In this light, students in all three groups seemed to have no choice
but to use the simple aspect to present their opinions or beliefs as in S1-S6. However, in the
Exempt students’ essays, there were fewer cases of using the simple present with the first
pronoun “I,” possibly indicating their awareness of the language disciplines of formal academic
discourse which usually discourage the use of first person or second person pronoun in cliché
statements (e.g. I think, I believe, etc.). In other words, although the Exempt group also used the
simple aspect most frequently, this function of providing personal opinions as a part of showing
43

“states” did not seem to be the most influential aspect of the simple present that is attributable to
its high usage for this group.
S1: I admit that performance based pay for teachers has a
bright side to encourage teachers to be creative,
resourceful, and more prepared. (500_K19_A)
S2: This means that people don’t need to worry about GM
foods’ effect on human body. (500_K9_A)
S3: We know that there are lots of countries in the world
which have laws to force manufacturers to label whether
their products are genetically modified or not. (500_C23_A)
S4: What’s more, I think this narrows the curriculum as the
lecture says. (500_27_A)
S5: As a result, I believe it is not wise to separate GM
foods with non-GM foods by labeling them. (501_C6_A)
S6: This implies that performance based pay program can be
one of the solutions to guide the students to do better in
school and also satisfies the parents. (501_K11_A)
The simple aspect used to integrate academic articles or personal anecdotes.
Probably the most problematic aspect associated with using the simple aspect is making
reference to other sources by quoting, paraphrasing, or summarizing what other people have
found or claimed. Almost all grammar textbooks suggest that the simple present tense be used for
“unchanging” facts or general truths while the simple past tense is used for actions that are
completed and thereby “remote” from the present time (Celce-Murcia & Larsen-Freeman, 1999);
however, not many of them explicitly address or guide writers how to choose the right tense for
the reporting verbs. As a consequence, many students, especially those whose first language is
not English, have a hard time choosing the right tense, and there were many incorrect uses of
44

verb tense and aspect noticed from the EPT essays in the current study when students cited from
the given references.
According to Celce-Murcia and Larsen-Freeman’s (1999) elaboration on the tense usage
for reporting verbs, simple present tense is used to convey immediacy and vividness and to
report statements that would hold true in the present or that has some kind of present relevance
(Thompson, 1994). Batstone (1995) also claims that the simple present is chosen over the past
tense when the “writers may convey personal stance with regard to a reported proposition” (as
cited in Celce-Murcia & Larsen-Freeman, 1999, p.706-707). Further, they conclude that the
simple present tense is used for reporting purposes whenever the reference source is a written,
published one, for “once something is published, it is permanently ‘present’ to the public as
general truths” (p.707). In this sense, the reference source given to the students for the EPT
written exam could be expected to be reported in the simple present tense. Nevertheless, many
students seemed to use the simple present, simple past, and present perfect in a random way,
without clear understanding of how choosing one over the other verb structure would influence
the meaning and function of the reported statement as shown in S1-S8. This random use of verb
tense and aspect resulted in ambiguity in meaning, and difficulty in understanding. This
phenomenon will be further examined in the last part of the results section in regard to the verb
shifts and discourse-level verb inconsistency.
In terms of the difference among groups, all three groups seemed to lack extensive
knowledge in employing the correct usage of the simple present tense when referring to the
given sources and integrating them into their own writing. However, compared to ESL 500
students, who integrated other sources only minimally, Exempt group students seemed to make
better use of given references. That is, whether they used the verb tense and aspect structure
45

accurately or not, advanced students tended to refer to the given article more often than less
advanced students, who usually supported their arguments with more abstract concepts or
theories than concrete and tangible evidence. Although the total cases of misusing simple past
tense in the Exempt group outnumbered the other two groups, the large number could actually
have resulted from their frequent attempts to use the references. Thus, it would be possible to
conclude that implementing references and citing sources within the text attributed to the cases
where the simple present tense “should have been used but avoided” in all three groups, and the
simple present tense was replaced by a variety of verb forms; it showed that Exempt group
students almost always replaced it with the simple past tense (S7-8) while the other two groups,
especially ESL 500 group, used any random verb tense and aspect combinations, which could
result in ambiguity of the sentence (S1-6). Furthermore, it could be inferred that most students
were not aware that the academia shared this “overt secret” of using simple present tense (or the
present perfect in some cases) for implementation of other sources because the problem
remained similar across all levels despite their language proficiency or fluency.
S1: Although there *is a federal food petition to advocate
the mandatory labeling of GM food recently as Erical
Holdsworth ?mentioned in his article: Heating Up Over
Labeling Genetically Modified Foods, we still have no
federal or national laws about it. // There are 18
countries in the world including China and Russia have such
kind of laws, *written by Erica Holdsworth in the same
article above. (500_C23_A)
Note. In this sample, it is difficult to understand which part belongs to the reference of
the article. The student used simple present (is, are, have), past present (mentioned), and
perfect participle (written) all together in one paragraph, which causes confusion and
inconsistency.
46

S2: In the article, Nicholas Whitman *said it is possible to
label “no GMO” on non-GM food as well as “this contains
GM.” (500_K11_A)
S3: Also, CAE’s recent report *mentioned that the higher
students’ test scores is, the more incentive teacher gets.
(500_K19_A)
S4: This also be *mentioned in the passage, Alice Dunerray, a
science teacher at Harte Charter School in New Orleans, is
the one benefits from this system. She get a $2,000 bonus
for her excellent work in improving children’s performance
and now she is fully passioned about caring more about
students. (500_C27_A)
S5: The article *mentioned that fifty countries have approved
the mandatory labeling, even including many developing
countries. (501_C15_A)
S6: “Most Americans, it’s not an issue for them,” ?said
Gregory Schmidt, the president and CEO of the council in
the article. The statement above *has been recorded in the
article “Heating Up Over Labeling Genetically Modified
Foods” by Erica Holdsworth (2012). Finally, according to
Erica Holdsworth (2012), the FDA claimed that foods have
genetically modified ingredients proven non health risk for
consumers. (501_C23_A)
S7: In the lecture, the speaker ?stated that workers *can
seek better employment opportunity globally and upgrade
their working skills. (Exempt_C35_A)
S8: According to the reading passage, Richard Doerflinger
regards therapeutic cloning as immoral and he *thought
cloning makes us treat life as instrument and object.
However, Daniel Perry *said that there is a clear line that
separates therapeutic cloning from reproductive cloning
which duplicates an organism. (Exempt_C33_A)
47

S9: And as is mentioned in the lecture, opponents for the
performing-based pay program maintain that curriculum
become narrower because teachers want to get the high test
scores. (501_C11_A)
S10: What’s more, the personal experience of Alice Duveray
shows the bonuses do help them performance better. (501_C41_D)
S11: In the article of “The Eight Losers of Globalization,”
Guy Preffermann explores the eight myths about
globalization. (Exempt_C3_A)
S12: As the lecturer says, developing countries as a whole
increase its share of world trade, making private capital
flow to it which is important to the world economy market.
So people can buy more stuff, which is very essential to
GDP growth. The article also gives the same information.
(Exempt_C11_D)
S13: As Randi Carter, a middle school math teacher in Cobb
Country says in Newman’s paper “Good Scores Good Money,”
teachers can’t control which kids walk into their
classrooms and that makes sense. (Exempt_C15_D)
Similarly, stories or anecdotes of the past are another type of reference that is often
expected to be used as supporting evidence in essays. According to Hinkel (2004) whose study
also analyzes the frequency rates and patterns of verb tense and aspects in NSS written texts,
NNS essays tend to employ the simple past tense as the writers often recounts personal narratives
or anecdotes that are “not generalized to other similar situations” (p.16).
However, in the current study, there were few cases of using the simple past tense. In
particular, in ESL 500 group, there were only 10 cases of using the simple past tense correctly
and only 8 cases which required using this form, possibly indicating the simplicity of their
language and the lack of variety in the types of evidence they used. To elaborate, Exempt group
48

used the simple past tense 6 times more frequently than ESL 500 group and twice as often as
ESL 501 group, which suggests how the advanced students tried to use more variety of contents
to support their arguments in more diverse sentence structures as in S14-15. On the other hand,
the lack of samples using the past tense for stories implies how lower level students have limited
range of contents, and even when they did, sample S16 reveals how lower level students made
poor use of verb tense and aspect choices to use stories and anecdotes as illustration.
Nevertheless, there were not many personal anecdotes of their own stories in all groups, which
could be seen as the influence that academic writing discourse may have imposed on the
students’ tendency to always use more logical, factual, and academic evidence rather than
personal stories.
S14: On the other hand, going against globalization can
seriously hinder the development of a country. Before the
policy of economic reform, our country was one of the
poorest country in the world. The politicians of China at
that time were strongly against capitalism and the trade
with developed countries. It was a dark time in China’s
history. (Exempt_C19_A)
S15: Can cloning really, truly, improve an individual’s life?
It may expand a person’s life span, but it may not make him
or her happy. The graduate in my discussion group told us
how the medical clinic she volunteered once expanded this
young patients’ life by providing medicine three times a
day. Their counterparts died around six years old, but they
are now 18 or 19. But she said “I can hardly tell who are
happier.” (Exempt_C73_A)
S16: Take a high school teacher in my country as an example.
There is a policy in my high school that the teachers are
paid only by students’ test scores. So my
headteacher ?works very hard, and because of his rich
experience, he ?has very good teaching method. My class
49

always ?ranks the first in my school, and he ?is well paid.
But when some young teachers ?turn to him for suggestions,
he ?is not willing to share his teaching method because
he ?doesn’t want other classes beat us in test. (501_C53_D)
Findings on the Perfect Aspect
Table 6
Frequency Rate of Usage Patterns for the Perfect Aspect
Rate of total use
(of the total verb combinations)
Rate of correct uses
Rate of avoidance/non-use
(of the total avoidance cases)

ESL500
2.4%

ESL501
3.1%

Exempt
5.3%

91.9%
14.0%

89.0%
14.4%

97.8%
22.6%

One of the most salient differences observed from the three groups was the use of perfect
aspect. The total numbers of the perfect aspect used strongly indicates that ESL 500 students
tended to avoid using the perfect tense; Exempt students made use of the perfect aspect about
twice more frequently than both ESL 500 and ESL 501students, and this group’s percentage of
using it correctly was also the highest among the three groups (see Table 5). As S1-S3
exemplifies, it was likely that ESL500 students lacked understanding of perfect aspect and
thereby avoided using it even when adverbial phrases were clearly used to indicate what scholars
have called the “up-to-now” or “extended now” effect such as: for years, for decades, since then,
for five years, until now, etc. (see Meyer, 1992 ; Inoue,1979; McCord, 19782). In contrast, many
ESL 501 students and most Exempt group students were able to employ perfect aspect to indicate
“up-to-now” or “extended now” effect as in S6-S9. However, as can be seen in examples S4-S6,
2

“A speaker of English uses a sentence in the present perfect when the information he is giving appropriately
exemplifies or explains the topic of discourse.” (Inoue, 1979, 585)
“The ‘up-to-now´ perfect: used with an adverbial of duration to bring the situation up to the speaker’s present.”
(Meyer, p.226-)
“two broad interpretations of the perfect: ‘current relevance’ (Comrie, 1976, 52) and ‘extended now’” (McCord,
1978, 18)
50

a generalization could not be made for the ESL 501 students’ ability to use the perfect aspect
correctly; there were both correct and incorrect uses of the perfect aspect even within the same
essay written by the same student, so it was difficult to tell whether they truly understood the
concept of the perfect tense let alone the usage of it.
S1: For decades, this paying method *is always
controversial in the U.S. society. (500_C25_A)
S2: For years, teachers *will get higher salary when they
teach for a long time. (500_C39_A)
S3: In recent years, large number of teachers *resigned
from the schools to seek better opportunity with higher
salaries. (500_C39_A)
S4: So far, the petition for mandatory of labeling has
gained 850,000 support around the country….The discussion
about the necessity to label GMO *became more and more
heated recently…People who disagree with this mandatory
*showed their concern on price rising and discrimination of
GM foods in the food market. (501_C15_A)
S5: The FDA has long been convincing people that the GMO
foods are “substantially equivalent” to natural foods.
(501_C3_A)
S6: In the past decades, teachers in the United States
usually *paid based on their working age. However, such
situation has been challenged since some policy makers
tried to find a performance-based way to pay for them.
(501_C41_D)
S7:Cloning has become a hot topic ever since it was
invented. (Exempt_C33_A)
S8: Animal testing has long been a controversy issue that
attracts attention from different walks of society…Each
year, many kinds of new pharmacy has been developed based
on the research on animals. Apart from the medicine
industry, animal testing has also been used to judge
whether the painting or other chemical materials will
51

effect human health. (Exempt_C5_A)
S9: Actually, unemployment in the industrial countries has
not increased systematically over the last 10 to 20
globalization years. Another important influence
globalization has brought about is capital movement and
work flow. Overall poverty in the Third World has
diminished. In other words, globalization makes the rich
richer and the poor poorer. (Exempt_C47_A)
While ESL 501 students varied in the consistency of using the perfect aspect correctly,
most students in the Exempt group did use the perfect aspect accurately and for more diverse
functions (S7-S12). Some students from Exempt group even used simple perfect when referring
to information provided in the given article or when mentioning what they have done in their
own essays as in S10, which seemed distinctive from the other groups. Moreover, students in the
Exempt group were able to use the perfect aspect not only to indicate duration but also for
“current relevance” or “indefinite past” effect (Swan, p.427; Meyer, p.226) in S11-S15. On the
contrary, lower-level students did not seem to be aware the basic difference between using the
present perfect and the simple past in terms of the relevance of the results to the present time
(S16).
S10: By far, I have discussed about the dark side of
globalization. (Exempt_K5_A)
S11: I have read an article about the application of
therapeutic cloning. It was a young lady who got skin
injury in an accident…. (Exempt_C33_A)
S12: In conclusion, I think Britain’s laws regard cloning,
which has banned reproductive cloning while allowing
research on therapeutic cloning is reasonable and wise.
(Exempt_C33_A)
S13: The GM industry has brought great benefit to people and
largely reduced staring…As an answer to the request of
52

consumers who want the right to know, the European
Commission has asserted a fairly strict rule on labeling GM
food even a single ingredient. (501_C3_A)
S14: There are no evidence GM foods are unhealthy and
harmful for people. FDA has concluded GM foods are same as
natural foods about safety. Thus, food manufacturers do
not need to label for GMO as FDA said. (501_K7_A)
S15: In addition, poverty, one of indicators of economy, has
diminished all over the world (Pfeffermann,
2002)...Moreover, because of globalization, medical
equipment and technology has been improved much in
developing countries. As a result, life expectancy has
been increased and child mortality has also decreased
(Pfeffermann, 2002). (Exempt_C43_A)
S16: Although the FDA *introduced the GMO food as a safe
food and actually there is no evidence that GMO is not
good for our health. (500_K13_A)

In short, Exempt group students had better understanding of the perfect aspect in terms of
its meanings, functions, and usages while their counterparts in ESL 500 were not familiar with
using the perfect aspect, which resulted in avoiding using it and resorting to the simple aspects.
As for ESL 501 students, not a clear generalization could be made for their inconsistent use and
relatively low rate of correct uses. Nevertheless, although there were differences in using the
perfect aspect among the three groups, the total ratio of using this verb form was relatively small
compared to other tenses and aspects; the percentage of using the present perfect was only about
5% of the entire verb tense and aspect uses, and the number of avoiding the present perfect was
actually highest in the Exempt group. Such shortage of using the present aspect could have been
caused by the lack of knowledge but could also have been attributable to the context of the
writing. In other words, because of the limited sources and disciplines of the academic discourse,
53

even the students who were familiar with the perfect aspect may not have had a chance to use it
in their essays.
Findings on the Progressive Aspect
Table 7
Frequency Rate of Usage Patterns for the Progressive Aspect
Rate of total use
(of the total verb
combinations)
Rate of correct use
Rate of avoidance/non-use
(of the total avoidance cases)

ESL500
0.01%

ESL501
0.01%

Exempt
0.01%

66.6%
4.4%

71.4%
4.3%

85.2%
5.8%

Progressive aspect was the least used aspectual structure in all groups; the total number of
progressive aspect used by ESL500, ESL501, and Exempt group students was only 10, 15, and
23 times respectively. Students in ESL 500 failed to use the progressive aspect even when there
was a clear phrase indicating the continuation of an event as shown in S1-S2. On the other hand,
they sometimes used the progressive aspect improperly when they should have used simple
present tense instead as in S3 and S4, which made their essay inappropriate for formal academic
prose.
S1: More and more people *concern about the genetically
modified organisms in America. (500_C7_A)
S2: More and more people *concern about this issue now.
(500_C33_A)
S3: The lecture and article ?are both talking about the
debate between seniority-based pay and performance-based
pay. (500_C37_A)
Note. In this case, the student used simple progressive to refer to his own writing process,
which seemed inappropriate for formal academic writing style as opposed to informal
conversational style.
54

S4: As far as I’m concerned, I?’m opposing the performancebased pay system due to the following reasons. (500_C35_A)
Note. In this sample, the student misused the progressive aspect to propose his/her own
claim in S4, which would probably be a state, rather than an action.
In contrast, more advanced students seemed to employ this aspect accurately from time to
time as in S5-S9. Nevertheless, it was not clear whether they understood the difference between
the progressive aspect and the perfect aspect when indicating the duration of a state or an action.
For instance, unlike S7, where the ongoing action was clearly indicated to be in the present
timeline with the adverb “now,” samples S5 and S6 show cases where it was not clear to
determine whether the underlined verbs referred to actions whose beginning point was before the
speech time; if the action began prior to the present reference time, verbs should have been used
in here the perfect aspect to indicate that the event has been ongoing from the past to now. In
other words, even though there were less than 10 instances of misusing or avoiding the
progressive aspect in each group, whether students really understood the differences between
progressive aspect and perfect aspect remained questionable.
Finally, in terms of referring to ongoing actions or states in the past, there was almost no
case of using past progressive in all levels. There was only one case of an Exempt student using
the past progressive appropriately in the context. Since almost every student did not make one
single use of this verb form, it would be reasonable to conclude that students considered it
unnecessary or inappropriate to include an on-going event in the past as their evidence or
supporting argument when writing a formal argumentative paper. In S10, the student
successfully used the past progressive, but the chance of using it in a timely and effective way
would heavily rely on which reference (e.g. reading articles) was given to the students for the

55

EPT writing exam. Thus, it was difficult to investigate from the EPT test what NNS students
actually knew about this verb form in terms of its meaning, function, and usage.
S5: These days consumers seem to know almost everything about
their products. A lot of countries are also labeling if a
product contains GMO. (501_K5_A)
Note. In this sample, although the use of progressive aspect is possible, it is impossible to
find out whether the student actually intended to express the ongoing event in the present
time or he or she meant to express an event that began in the past but continues to the
present, in which case, the use of the perfect aspect would have been more appropriate.
S6: The FDA is highly recognized throughout the world, so we
can pretty much rely on what they are saying. (501_K5_A)
Note. In this sample, it is also unclear whether the student was trying to say that FDA is
“now proposing something new” or that FDA “has been” saying something from the past
to now. In regard to the context and common sense, it would make more sense if the
perfect aspect were used, but this usage of the progressive aspect cannot be marked
“incorrect” because there is no way to prove it.
S7: Many scientists are now working on to find how to keep
the animals away from distinction by studying their most
appropriate hiring environment, breeding, and so on.
(Exempt_C5_A)
Note. This sentence, unlike the previous samples, clearly signals that the reference time
is the present by using the adverb “now.” Therefore, it is clear that the student is using
the progressive aspect as for the desired function as opposed to the perfect tense.
S8: It is true that cloning is changing our world, but we can
also feel that the controversy is rising. (Exempt_C31_A)
S9: Some people in the world—especially poor African
countries—are suffering from lack of nutritious food. //So
many people including children and women are suffering from
absolute poverty and food shortage in North Korea.
(Exempt_K4_A)
S10: When the government of Haiti was planning to lower the
duties on import cooking oil and focus on more profitable
56

industry, 300 employees were deemed to lose their jobs.
(Exempt_C47_A)
Findings on the Predictive Aspect
The use of predictive will.
Table 8
Frequency Rate of Usage Patterns for Will
will
Rate of total use
(of the total verb
combinations)
Rate of correct use
Rate of avoidance/non-use
(of the total avoidance cases)

ESL500
9.9%

ESL501
10.2%

Exempt
5.5%

92.7%
30.5%

96.1%
18.0%

94.7%
19.3%

When proposing their ideas and arguments, students in all three groups used the simple
present tense or will more frequently than other predictive modals; they tended to overuse either
of these two verb forms, which made their statements assertive. However, there were some
distinctions among the three groups in terms of using the modal will, which has multiple
functions in the English language. Based on the LSWE Corpus, Biber (1999) categorized the use
of modals into three groups according to the meaning—permission/ability, obligation/necessity,
and prediction/volition—and studied how each modal is used in various discourse. According to
their research on the usage of will, they mentioned its function of prediction and volition but
concluded that will is the most frequently used modal for making predictions in academic prose.
Palmer (1990) also distinguished “epistemic will” from “will of futurity” and categorized the
former into the meanings of volition, power, and habit (p. 137-142). Although it may often be
ambiguous to find one single category per use, the general distinction between “will of futurity”
and “epistemic will” seemed significant in understanding the modal usage patterns of NNS
students in this study.
57

Table 2, 3, and 4 reveal that every group used the simple present much more often than
any predictive modal auxiliaries. However, the total numbers in the table do not show which
function of the simple present each group used it for, which was why the current study offered
function-by-function analyses in the previous section on the use of the simple aspect. Likewise, it
was impossible to make any kind of assumptions or generalizations about the use of modal will
just by referring to the frequency counts in the table; rather, the actual student samples showed
how and why the students used this verb form in their writing.
Although the total number of using will of ESL 500 students (total number of 115) was
larger than that of Exempt students (89) and little smaller than that of ESL 501 students (174),
their usage of will was different from the other groups. They seemed to be unaware of or at least
very uncomfortable with using the modal will as an aspectual device to refer to what Palmer
(1990) called “what is reasonable to expect” or “a judgment made with a reasonable explanation
(from previous knowledge)” in contrast to the “will of futurity.” In other words, their use of will
was mostly limited to making “future” predictions, not as a way to refer to the present time.
Consequently, the 39 cases of avoiding use of will in ESL 500 group most often resulted from
replacing “predictive will” with the simple present when they were making predictions whose
reference time could be in the present timeline. To elaborate, if what they saw as the result of the
prediction seemed also possible to be applicable in the present as well as the future, they often
used the simple present tense instead of will. Notice how students used the simple tense instead
of will in S1-4. Consequently, when ESL 500 students used will, they intended some kind of
connection to the future time and/or the meaning of volition as in S5-S6, which Biber et al.
(1999), claims to be rarely used in academic writing discourse.

58

S1: Some of the consumers ?feel uncertainty with GMO foods
and ?don’t want to eat these foods. If food companies don’t
reveal these information, they ?don’t have choices to rule
out GM foods. (500_C1_A)
S2: Even though in someday when the GMO *has been classified
as risky, the right thing to do ?is to handle this problem
with science, not to paste “not safe” on it. (500_C3_A)
S3: If teachers’ salaries are closely connected to students’
performance, students’ test scores *become the only
criteria to measure teachers’ quality. For example,
teachers may teach the easy part in class because they
assure if the easy part is tested in the exam, it *is much
easier for students to get higher scores…In addition,
performance-based pay will cause a competitive environment
so that teachers ?are reluctant to share ideas with others.
(500_C25_A)
S4: Although in some extent, such kind of salary given method
may help improving the quality of education, attracting
more skillful teachers and making them paying more
attention on students’ performances, it also *cause more
problems when putting it into practice.// In addition, the
teachers’ over focusing on the results of test ?lead to
mess up of the purpose of education with encouraging
teachers. For they ?prefer to pay more attention on
students’ scores rather than the proceding of learning
itself. (500_C47_A)
Note. In this case, the student indicates that he or she is making predictions, not
presenting facts, by the use of may in the first paragraph; however, he or she uses simple
tense in the same paragraph, which makes the prediction sound like facts without any
clear reasonable evidence. Thus, it is likely that the student does not understand the
function of “predictive will.”
S5: For example, when I choose to buy some food, such as
rice, I cannot figure out which brand of the rice is GMO,
then, I will buy the one that has clearly labeled it. In
the following days, I will buy the same brand, for I ?have
59

confidence in this brand and it makes me feel safe and
reliable. (500_C9_A)
Note. In this example, the student uses will with the meaning of volition, showing his
willingness or determination of future actions.
S6: According to the lecture, it’s because labeling can
reduce the anxiety of buying foods. Consumers who concern
about GM foods will not buy the food. On the other hand,
consumers who don’t care about GM foods or who believe GMO
is safe will still buy GM foods. (500_C19_A)
Note. In this example, will is used to predict events that are clearly going to happen in the
future, and it also conveys the volition of the customers who can decide whether they
would buy GM foods or not. Therefore, the use of will in this case is not for the
“predictive” function.
On the other hand, students in ESL 501 and Exempt groups used will often as the
“predictive” function, to propose what they consider as a reasonable judgment or strong
predictions. However, many times their use of will seemed inappropriate or inauthentic, because
they did not provide sufficient logical reasons to rationalize the use of will over would or may. In
other words, most students used will instead of would or may, failing to mitigate the degree of
assertiveness in their predictions or arguments. Some students from ESL 501 or Exempt Group
were able to employ the appropriate use of will with logical reasons when making strong
prediction or assertion, but there was no clear distinction observed between groups. Compare
how S7-8 made strong assertions and predictions by using will without sufficient logical reasons
while S9-11used will only minimally when it was based on logical reasons from previous
sentences.
S7: On top of the customers’ perspective, the food industry
could be promoted through the requirement of labels. The
reasons behind it are the customers’ awareness of food
60

safety and the improvement from food companies. Customers
will be smarter on choosing food, which will drive the
quality competition between companies. On the other hand,
companies will spend more efforts to investigate.
Therefore, the food manufacturers will ensure quality of
foods and customers will choose the foods wisely, enhancing
the health of food industry! (501_C9_A)
S8: If a teacher will be paid more with higher scores his
students get, the teacher will pay his attention more to
students’ test score. If a teacher merely focus on
students’ score, the other aspects of students, such as
skills, attitude, spirits and so on will be ignored, and
the consequence will be extremely terrible. Besides,
focusing on students’ test score is likely to lead to
improper competition between teachers which will also
destroy the collaboration. (Exempt_C15_D)
S9: According to the demand theory, the prices of GM Foods
would go down if there are less demands. So people who
don’t care about GMOs could buy foods with lower prices. At
the same time, people who care about GMOs would pay extra
money for non-GMOs out of their will. In a word, labeling
of GM Foods will be economically beneficial. (501_C13_A)
Note. In the last sentence, the use of “will” makes sense because he has already presented
the reason/evidence for his idea in the previous sentences using the demand theory.
S10: Indeed, there are problems brought by globalization,
but these problems are tangible and can be solved…Actually,
some solutions are already emerging. For example, the US
department of homeland security has new policy to
facilitate the process of Chinese scholars to come to
U.S…With certain efforts, I believe globalization can be
solved in the end…It has benefited the countries involved
and will continue to benefit the life of human beings. As
long as we continue to work on the problems in this
process, more and more people will support globalization.
(Exempt_C19_A)
S11: …Globalization will give rise to international
investment…To illustrate, international investment will
61

help developing countries grow more rapidly with the
capital input from old industrial. For instance, the
globally convenient transportation system can link the
world together…I believe in the long term the whole world
will benefit from financial system because finance is a way
to make people live better with wise investment.
(Exempt_C35_A)
The use of would & may/might.
Table 9
Frequency Rate of Usage Patterns for Would
would
Rate of total use
(of the total verb combinations)
Rate of correct use
Rate of avoidance/non-use
(of the total avoidance cases)

ESL500
4.0%

ESL501
3.7%

Exempt
1.8%

79.0%
13.2%

92.3%
18.7%

100%
3.9%

ESL500
5.1%

ESL501
6.0%

Exempt
3.6%

100.0%
0%

97.2%
2.9%

100.0%
2.6%

Table 10
Frequency Rate of Usage Patterns for May/Might
may/might
Rate of total use
(of the total verb combinations)
Rate of correct use
Rate of avoidance/non-use
(of the total avoidance cases)

For expressing possibility with epistemic modals, “making judgment about the
possibility that something is or is not the case” (Palmer, 1990), the use of will was investigated in
the previous section. However, there are other modals which Biber’s (1999) corpus-based
research proved to be used frequently in various registers of English; it showed that would was
the second most commonly used modal to express prediction. Palmer (1990), paraphrased the
meaning of would as “I should think that…” or “It would be reasonable to conclude that…” and
added “tentativeness” to its meaning (p. 58-59). Hinkel’s (2004) research on NNS’ use of verb
62

tense, aspect, and modality which compared academic writing samples of non-native speakers
and those of native speakers also indicated that using the “predictive would” was a trait
commonly found in NS writers’ essays. Therefore, using modals other than will was investigated
as a criterion for adding more authenticity and native-like fluency to the NNS writing.
Not surprisingly, NNS writers did not employ the use of would as often as they used will;
the total number of using would by the students in ESL 500, ESL 501, and Exempt group was
49, 60, and 30 respectively, which was approximately one third of the total number of using will
in each group. Samples S1-S4 show how ESL 500 group students rarely used would but resorted
to will when making predictions. As for the use of may/might, most of the students did not seem
to be aware of the fact that may is paraphrased as “it is possible that…” and usually infers having
no specific reasons or rationale for making that prediction. Consequently, in many cases,
students used may/might instead of will in places where they did not originally intend to hedge;
when they misused may/might for will or would, their opinions sounded too weak as in S5.
Moreover, when students used one of the modals, they seemed to be using it interchangeably
with other modals, and there was no clear distinction noticed among groups in using the modals
with desired meaning distinctions. Within the same group, modals will, would and may/might
were used sometimes with proper reasons, but other times without any rationale as in S6-S10.
S1: However, in my opinion, the traditional payment ?is
reasonable if teachers devote their whole life into
teaching career, that means they ?are somewhat successful
in this field and worth of high salary, or they ?will quit
their job to find another work or be fired. Performancebased pay is not so effective as it seems. In fact, it
even ?will bring some side effect to both students and
teachers. (500_C41_A)
S2: However, GMO labeling is a significant sign of the
63

threat to basic requirement of society, which ?will make
the customers much anxious about the market. Expected
investment in the market would be delayed or even
canceled, and this will effect many areas, where destroy
power is much greater than so called suspected genetic
threat to people. (500_C3_A)
S3: First of all, when people evaluate teachers’
performance, they ?will relate it to students’
performance, but they ?are not equal things actually.
Students’ performance ?will not represent their teachers.
No teacher can determine which kids ?will be in the class
or control what unlucky things ?will happen on their kids.
Many outside things ?will effect kids’ performance thought
those external facts have nothing to do with teachers.
(500_C41_A)
Note. All predictions made in this passage is not based on logical reasons, so hedging is
necessary.
S4: I believe in some cases, in order to obtain higher
salary, some teachers ?will give out some information about
the exam to make sure their students *will get higher
scores. Furthermore, particular teacher ?tend to tolerate
cheating in his exam. // And once the teachers salary
depends on their students’ performance, one teacher ?will
only care about his own students. And if some teachers find
some better methods to teaching, they ?will not share these
experience with each other. (500_C51_A)
Note. These predictions made with will do not seemed to be based on sufficient logical
reasons.
S5: First of all, under the performance-based circumstances,
the teachers ?may work hard to teach students. According to
Duvernary in the reading passage, under this circumstances,
they have much more motivation to teach more in order to
make sure the students totally understand the knowledge and
they can gain more skills.//In a word, under the
performance-based salary system, the teachers ?may work
hard and it can avoid bad attitudes from teaching. (501_C59_D)
64

Note. In this case, it does not seem that the student intentionally used may instead of
would both in the first and the last sentence. When the student presented an opinion, it
should have more certainty than mere possibility indicated with the use of may, for he or
she must have some arguments for the claim.
S6: However, there would be some consumers who ?doesn’t want
the labeling and *has to pay for the significant cost for
labeling…On the reading passage, they say if they label
GMOs, the consumers would have negative attitude toward the
GMO products and as a result, the GM products would lose
their position and as a result, GM products would lose
their position on the market. However, I think it ?might be
right in a short term. (500_K13_A)
S7: Mandatory labeling may lead to no GM products in the
market. Producers would prefer to label non-GM foods rather
than GM foods. In this case, labeling indirectly push GM
products out of the market. (501_C20_A)
Note. In this example, the use of may and would seems appropriate because the first
sentence is merely making a prediction that is “possible” to happen while the second
sentence is an opinionated claim which would be supported by some evidence in the
following texts.
S8: Although in the lecture, it says that it may motivate
teachers to over-focus on test scores, instead of students’
learning, it is really difficult to find one specific way
to judge students learning, and test score may be proved as
one discerning path to find students’ learning is good or
not. (501_C43_A)
Note. In this case, using may is logical because the writer is deliberately using this
reference from the lecture not as a supporting evidence but as an opposing evidence; that
is, the writer probably intended to make the given statement less certain because it
opposes his or her own claim.
S9: When medicine is opened to the market, patients and
doctors can get the benefits. For patients, they would get
better information about drug and disease they have. For
doctors, they would take more advanced medical knowledge
through globalization. (Exempt_K1_A)
65

S10: Reproductive cloning which intends to make babies is
ethically unacceptable and has potential health risks to
children conceived in this way. First, since the new born
organism has the same genetic code as its original one,
this may cause endangered animals to be wiped out by the
same disease. Second, reproductive cloning ignores the fact
that human is unique. It would result in all the cloned
people lose their own features. Third, the technology of
reproductive cloning is not mature. It may have potential
risks that we don’t know. (Exempt_C29_A)
The use of “if-clause” conditionals. As one of the most confusing concepts in English,
conditional aspect of “if-clause” remained problematic in essays. In this study, the “unreal” or
“imaginative” conditional statements with “if-clause” were excluded from the predictive aspect,
because they are not making predictions that can happen but rather refer to “impossibilities with
reference to the present or the past” (Celce-Murcia & Larsen-Freeman, 1999, p.551). Instead,
predictive aspect indicated the “if-clause” conditionals which were used for predicting future
events or states under certain conditions and with possibilities to happen. Not surprisingly, very
few instances of properly employing imaginative conditionals were observed from all three
groups. However, many incorrect uses of “if-clause’ were noticed from all groups. In ESL 500
papers, there were minimal uses of conditionals as most statements are based on current
situation, facts, or simple generalizations using present tense; even when there were attempts to
use hypothetical situations, they seemed to misuse tenses both in the independent and dependent
clauses. In more advanced groups, there were more uses of conditions; however, many students
in ESL 501 also seemed to confuse these two distinctive usages of the “if-clause” and mix the
tenses as in S5-S8. In contrast, Exempt group was much better at using “if-clause” without tense
discrepancies than the other two groups, and there were actually a few cases of accurately using
“imaginative” conditionals for unreal situations as shown in S9-S11.
66

S1: If teachers’ salaries are closely connected to students’
performance, students’ test scores *become the only
criteria to measure teachers’ teaching quality…If we
neglect those practical reasons, this performance-based pay
system *is actually unfair. (500_C25_A)
S2: Even if there’s a way to define these achievement, *but
the achievement itself *is generally complicated. (500_C59_A)
S3: Once there *were financial rewardings, the teachers would
become more competitive so that they must work harder to
improve their own skills to get the bonus. (500_C53_A)
S4: If the experiment *showed that the policy has
beneficial effects on improving the test scores, it may
also benefit other kinds of student performance as well.
(500_C53_A)
S5: For the long-term development, labeling GMO will become
the future trend. We ?do not need to worry about food safety
when importing food from other countries if all the
countries use GMO labeling. (501_C27_A)
Note. In this case, using the simple present in the main clause as presenting a fact is
logically awkward because the overall context, which can be seen from the previous
sentence and the if-clause, indicates that there is uncertainty to the event which will
happen in the future.
S6: If GMO labeling becomes mandatory, people will become
aware about GMO food quality. Consumers may feel that there
may be something wrong with GMO food. If that *were the
situation, with less and less customers buying GMO foods,
the whole GMO industry may start to shrink and disappear.
(501_C31_A)
S7: If the U.S. *were having the mandatory labeling laws,
the increased expenses to food companies will lead to
increased prices for consumers to pay. (501_C29_A)
S8: I check the food label for the production date, nutrition
in it and ingredients. I ?will feel very relieved if the
table *indicated this food is organic one. (501_C27_A)
67

S9: If I hadn’t been influenced by globalization, I wouldn’t
have taken advantage that I have had. (Exempt_K4_A)
S10: If you saw the movie “Island” which is about human
cloning, you would have an opportunity to think about the
issue. (Exempt_K4_A)
S11: Imagine that if we had many our own copies and someday
one copy had severe illness, would we want to donate our
organs to the copy? It’s really hard to face this
situation. (Exempt_C71_A)
From these analyses, it was noticed that students had difficulties using the predictive
aspect more than other aspects regardless of their language proficiency groups. Possibly, the
subtle meaning differences among English modals contributed to the trouble; especially
distinguishing will and would seemed almost impossible for most of the students. The use of
conditional statements was the only feature of using predictive aspect in which Exempt group
students outdid the other groups in terms of using the structure both more frequently and
accurately. Thus, it would be possible to claim that language proficiency of students does have an
influence on students’ using of verb tense and aspect, especially for a grammatical structure like
the “if-clause” which requires certain structure. Nevertheless, students’ language proficiency
assessed by the EPT did not seem to clearly reflect the students’ ability to provide logical reasons
for choosing a modal and to distinguish the subtle meaning differences each modal conveys.
Findings on the Discourse-Level Consistency
Many scholars have emphasized the significance of discourse-level grammar for its
influence on the “connectedness” (Givon, 1984, p.269) or coherence (Pongsiriwet, 2001) of
sentences within wider discourse context. Celce-Murcia and Larsen-Freeman (1999) have also
dedicated one full chapter on TAM in discourse so as to address potential incomprehensibility of
sentence-level grammar and to show how tense shifts and sequences matters in authentic
68

discourses. However, their observations focused more on conversational discourse and were
limited to “intra-paragraph” discourse, meaning that they consider tense and aspect uses of the
sentences within one paragraph, not within the entire paper or story. Going further, this study
considered discourse-level tense/aspect consistency in NNS students’ academic writing in two
folds: intra-paragraph discourse and inter-paragraph discourse.
Verb consistency in intra-paragraph discourse. Among the three groups of students,
ESL 500 group’s writings had the biggest problem of keeping consistent verb tenses and aspects
throughout a paragraph; they maintained using the same tense (usually the simple present) or
inserted an inconsistent tense and/or aspect in the middle of the paragraph for no logical reason
as shown in S1-S7. In contrast, a majority of Exempt group students showed natural shifts of
verb forms within a paragraph as in S8-S9; in particular, S9 showed an excellent example of verb
tense and aspect uses, which made the paragraph coherent and consistent within a paragraph.
Nevertheless, it was noticeable how students across levels of writing proficiency showed
inconsistent patterns of using verb tense when referring to the given article, citing from
authorities, and shifting aspects; S10-S13 show the discrepancies among the verb tenses and
aspects in all three groups .
S1: In the lecture, opposite side people were worry about
that consumers *will consider GM foods as something harmful
and wrong after labeling even if there *are no risk.
(500_K11_D)
S2: As it is already discussed in the lecture, people *will
assume that GM food might cause health problems. However,
in the reading, the FDA confirmed that GM food *is no more
safe than natural food. A study *conducted in Seoul
National University. There *was no health issue about GM
food for almost 20 years. With the positive reputation of
health safety, the GM food industry is growing fastly. It
69

is needless to say that labeling GM food has not to be done
in order to protect GM food industry. (500_K5_A)
S3: According to the article, the biotechnology industries
and manufacturers *argued that it *is very complicated if
we label all the GMO foods. This may increase the cost and
the complexity of labeling process, but in my point of
view, it’s the consumers who pay the money. (500_C19_A)
S4: Since most of the instructors will become competitively
and they ?don’t want to share their own ideas with others
for they are all scared of losing money through teaching
others how to instruct well. (500_C47_A)
S5: As is known to all, the number of teachers in the United
States has increased, so *is the payments. (500_C59_A)
S6: In CAE’s recent report *mentioned that the higher
students’ test score is, the more incentive teachers gets.
However, their test scores do not demonstrate teacher’s
performance. (500_K19_A)
S7: Traditionally, teachers get paid by seniority, that is,
with the growth of their working experiences, their salary
increases correspondingly. However, some groups *think that
the traditional payment cannot attract or retain good
teachers…Therefore, a reform *is carried out. CAE *decides
to pay teachers based on their performance, considering
their teaching assignments, skill and their students’
achievements. The new way seems to motivate teachers to
improve themselves. (500_C41_A)
S8: According to the reading passage, there are negative
impact of economic. They said that since imports wouldcut
the price sharply, it would have the impact on GDP growth.
However, I have a different point of view. (Exempt_K1_A)
S9: Globalization clearly benefits all countries involved.
The rapid and significant economic growth in the developing
countries is largely a consequence of globalization. Asian
countries like China and India, American countries like
Brazil and Mexico, and even African countries like Ghana
70

and Tanzania have shown tremendous economic successes (T.
Harris, 2008). Developed countries also greatly gain
benefits from globalization. According to researchers at
the University of Chicago, cheap imports from developing
countries have benefited developed countries (T. Harris,
2008). It has been pointed out by Tylor Harris that the
exchange of ideas (knowledge) during globalization is the
real prize. I, myself, is a good example of this process. I
did my undergraduate training in China and now I am in U.S.
for my graduate training in Biophysics. (Exempt_C19_A)
S10: First of all, regulation in car manufacturing and use
cause critical damage to developing countries... According
to the article, Global Carbon Project report that the
number of cars and trucks will increase from 850 millions
to 2.5 billions by 2050. And it *said this increase will
mainly related with developing countries. Also, as a
lecturer *said, this rules *cause problems in developing
countries’ economy because it *impede car manufacturing
industries. (501_K19_A)
S11: First, there are no evidence GM foods are unhealthy and
harmful for people. Tom Blanchard, a Monsanto spokesman
said that FDA has concluded GM foods are same as natural
foods about safety. They cannot find any difference of both
foods. Thus, food manufacturers do not need to label for
GMO as FDA said. (501_K7_A)
Note. Every tense should be in either present or past alike, but here it is inconsistent.
S12: According to the reading passage, there are negative
impact of economic. They said that since imports would cut
the price sharply, it would have the impact on GDP growth.
However, I have a different point of view. (Exempt_K1_A)
S13: But not like the
didn’t consider the
article, he come up
globalization, “the
down. (Exempt_C37_D)

opinion of Guy Pfefferman, which he
situation of market crashed. In the
with a example that due to
country’s only oil plant” was shut

71

Verb consistency in inter-paragraph discourse. Most of the times, verb tense and aspect
usages in discourse are reviewed by a paragraph-level analysis. Although one paragraph usually
reveals sufficient information on the writer’s tendency to use verb tenses and aspects, it was even
more revealing to revisit the issue in a bigger frame. That is, as the final product of the task, the
essay itself would provide even more information in terms of verb consistency of the writers. It
was interesting to find that some students showed consistency in using verb tenses and aspects in
the topic sentences of each paragraph while some had no pattern at all. It could be generalized
that the more advanced students were able to keep paragraphs coherent not only in terms of
contents but also for verb tense and aspect. Compare S1-S2 and S3-5 for their consistency in
verb structures and the resulting style of the voice and tone. In the following examples,
paragraph distinction is marked by a double hyphen (//)
S1: First of all, the GM foods is same to the natural food.
// Second, the total cost of labeling will be tremendous.//
Last but not the least, putting a label on the GM product
makes some people be suspicious about the safety of the
food. (500_7)
S2: Some people disagree with the requirement of labeling the
GM food, for it will dismiss the consumers. // First,
labeling of GM food provide a freedom for consumers to make
choices. // Second, labeling of GM food will help GM food
stay in a safe place. // Third, labeling GM food can
establish a feeling of confidence between consumers and
food industry. (500_C9_A)
S3: I would definitely disagree with the mandatory labeling
and my reasons are as follows. // First of all, labeling GM
food may cause serious yet unnecessary anxiety among
people.// Secondly, mandatory labeling may cause prices to
increase for GM product.// Thirdly, it could impede the
development GM industry. (501_C3_A)
72

Note. In this sample, all the three reasons are similar in its degree of assertiveness and
hedging
S4: I agree with the author of this article who claims that
boycotting should be judge by its result.// First,
everybody has the right to express one’s opinion…In this
point, I agree with Walter Goodman.// However, Goodman
follows too much absolutism. One example is the boycotts of
USA political association against the art performers…// In
sum, Goodman’s claim does not consider the consequences of
a boycott. (Exempt_K3_A)
S5: Accordingly, therapeutic cloning in terms of human
cloning and general cloning research in terms of non-human
cloning should be allowed while reproductive cloning should
be prohibited.// First, therapeutic cloning can help some
people who are suffering from incurable disease.// Second,
general cloning research can solve the poverty problem in
the world.// Some people might believe that cloning
technology is not fully developed and cause grave health
risks to humans. But these worries can be solved by
regulating reproductive cloning…Accordingly, some side
effects can be minimized by banning reproductive cloning.
(Exempt_K7_A)

73

CHAPTER 5
DISCUSSION AND CONCLUSION
In an attempt to verify the grand assumption that grammar teaching benefits language
learning and to see if it applies specifically to improving second language writing proficiency,
this study focused on English verb tense and aspect system and posed four major questions that
would give direction to the study and ultimately suggest how grammar teaching should be treated
in ESL writing courses. In order to find answers to these questions, the current study examined
120 argumentative essays collected and retrieved from the English Placement Test database: 40
essays from each of the three groups, namely ESL 500, ESL 501, and Exempt group. Although
previous studies have done some research on the patterns of NNS’ use of English verb tense,
aspect, and modality (TAM) system, the current study aimed to provide more in-depth analyses
of NNS writers’ tendency to overuse, misuse, and/or avoid using certain verb tenses and aspects
by extending the methodology, comparing groups of NNS writers by their language proficiency
levels, and considering discourse-level analysis of the observed phenomena.
To begin with, the current study developed the methodology by using both frequency
counts and error analysis of each essay, which increased the validity and adaptability of the
results to be viewed in twofold. To elaborate, for the frequency counts, 10 verb tense and aspect
combinations (“structures” or “forms” for the same meaning) were identified by the taxonomy
created specifically for this study, and each combination of verb tense and aspect used by each
student was counted. Then, the total number of each combination used by the students in the
same group was computed and labeled in a table. At the same time, the actual writing samples
that provided clear case-by-case explanations for the overuse, misuse, and avoidance of each
verb structure were documented as the error-analysis part of the methodology. Consequently, it
74

was possible to retrieve both general and specific findings from the data. In fact, the close
analyses of students’ actual writing samples enabled this study to extend the scope of research
and look for factors that caused such patterns in students’ writing. This aspect of the study is
unique in the sense that most studies in the past have not paid focal attention to the cause of
errors in verb usage and/or the phenomenon of avoidance. To my understanding, they have either
focused on the patterns without questioning the causes or immediately attributed the source of
problems to the complexity of certain verb structures and the students’ lack of knowledge.
However, by comparing the NNS students in three groups of ESL 500 (less advanced),
ESL 501 (more advanced), and Exempt group (most advanced—not required to take ESL
courses), this study intended to see if students’ proficiency level of L2 writing was directly
related to NNS writers’ verb usage patterns, which would be the answer to the research question
1. On the other hand, if there was no salient difference found among the three groups, it was
considered as an indicator of having factors other than the language level to influence their verb
usage patterns, and academic discourse was pinpointed as one of the most likely factors
generating the phenomenon as in research question 2. Finally, based on these findings, more
practical implications and suggestions were made both on the teaching of verb tense and aspect
in university ESL writing courses and on the English Placement Test as in research questions 3
and 4.
Research Questions Revisited and Summarized
1. What are the patterns of NNS students’ verb tense and aspect usage observed from the English
Placement Test essays? Does NNS usage of verb tense and aspect have a relationship with their
L2 writing proficiency?
2. Does academic writing discourse influence NNS writers’ use and non-use of verb tense and
aspect?
75

3. Is it necessary to teach English verb tense and aspect system in university ESL writing
courses? Will it benefit NNS students’ writing skills in academic setting?
4. What are the implications/suggestions for future grammar teaching and the placement test in
ESL/EFL courses? How should these grammar teaching be implemented to the ESL writing
courses?
Responses to the Research Questions
1.

Based on several studies that claim avoidance to have a positive relationship with the

language proficiency level (Mattar, 2002, 2003; Hinkel, 2004) or with the complexity of
language features being used (Hinkel, 2004), it was possible to hypothesize that students with
higher proficiency in L2 writing would show fewer cases of avoiding any verb tense and aspect
structures and make use of more variety of tenses and aspects in their essays. Hinkel’s (2004)
research has compared NS and NNS writers’ use of verb TAM in academic prose, and he has
concluded from his findings that NNS writers showed much lower frequency of using the present
perfect, predictive ‘would,’ and if-clause conditionals and very high frequency of using the
simple past in their papers compared to their NS counterparts. However, the results of the present
study agreed with such assumptions only partially, as the question of how often students use one
verb structure and how often they avoid it did not always represent their ability to use it in an
accurate and natural way for the desired meaning and function.
First of all, for many of the 10 tense and aspect combinations, there was little or no clear
difference in the total numbers of correct use, misuse, and avoidance as shown in Table 2-4. The
results of the study was consistent with those of Biber’s (1999) corpus-based research in that the
progressive aspect was rarely used in the written prose as the total numbers of it usage were 10,
15, and 23 in ESL 500, ESL 501, and Exempt group respectively. Also, none of the three groups
employed the use of the past perfect although there were a few cases where it should have been
76

used. Most importantly, unlike Hinkel’s (2004) study, all three groups evidently used the simple
present most frequently, which took more than 60% of the entire verb usage in every group.
However, although every group used the simple present most often, it did not mean that all three
groups showed the same patterns in using it. That is, students in the three groups used it different
functions within the simple present.
In terms of using the simple present to express “general truths or factuality,” all three
groups tended to overuse this verb structure even when using predictive modality would have
made their statements more natural and less assertive. It was found that most of the students
across the levels did not choose to hedge but to directly propose their claims in the simple tense,
making their predictions or assertions sound factual. However, as for the function of the simple
present as “an indicator of states,” which was often used in formulaic phrases like “I think,” “I
believe,” “we all know,” and “there are,” lower level students showed higher frequency of using
it. On the other hand, higher level students made more use of the simple present for the
grammatical function in subordinate “if-clauses” or time clauses as they were more familiar with
using the conditional statements. Finally, Exempt students used the simple present more often in
reporting verbs as they referred to the given articles more often by quoting, paraphrasing, and
summarizing the references. In fact, there was a meaningful difference between the ESL 500
group and the Exempt group in using different types of evidence for their arguments; while more
advanced students relied on “facts from what they heard or saw”, less advanced students used
“their own ideas or beliefs” for their supporting evidence more often. Such difference also
resulted in an unexpected result of the Exempt group students making more misuses in the
simple past and more avoidance in the simple present. That is, they made much more frequent
attempts to implement other sources by often using the simple past instead of the simple present
77

for the reporting verbs while the lower level students, who merely chose to stay on the “safe”
side by not implementing other sources, made fewer errors in using the verb choices. Therefore,
it was possible to infer that although the frequency of using a verb structure may seem similar
across levels, there could be differences in the meaning and function of the verb tense and aspect
used by students of different levels.
Meanwhile, for some verb structures, especially those claimed to be “more complex” or
“difficult to learn” verb structures (Hinkel, 2004), different usage patterns were more noticeable
among groups even from the frequency counts. The frequency of using the perfect aspect was not
very high in all three groups, but the most advanced students (Exempt group) made correct use
of this aspectual structure for 88 times, which was approximately double the number of the
present perfect accurately used by ESL 500 (34 times) and ESL 501 group (49 times). Besides
the total number of the perfect aspect used correctly, the actual writing samples of each group
showed how advanced students used this aspect for more variety of purposes. That is, Exempt
group students in particular used the present perfect not only to express the “up-to-now ” effect
of an event or state that has continue from the past to the present, but also to indicate the “current
relevance” or the “indefinite past” as opposed to the simple past. Nevertheless, it was difficult to
find any salient difference between the two other groups, for in each group, there were many
“exception cases,” and sometimes the line between ESL 500 and ESL 501 distinction seemed
blurred.
Another verb structure that was often expected to be “avoided” by less proficient students
was the “predictive” or “hypothetical” would, which was known to be frequently used in NS
writing. In the current study, using English modal auxiliaries will, would, and may/might were
grouped and defined as the “predictive aspect”, which were to be used either to make reasonable
78

predictions of an event based on logical thinking or sufficient evidence (epistemic function of
modals) or to simply assume what will happen in the future (modal of futurity). Again, similar to
what happened in the usage of the simple present, the total counts of the predictive aspect did not
reveal much about the differences among groups. In fact, it was found that ESL 500 students
made use of modals will, would, and may/might more often than Exempt students; however, the
total numbers of use did not indicate students’ fluency or accuracy in using the modals for the
desired effects. To elaborate, students in ESL 500 used will, would, and may/might
interchangeably and almost randomly when they made predictions without sufficient evidence; it
was not likely that they were aware of the differences in degree of certainty among the three
modal auxiliaries. On the other hand, Exempt students showed many cases of distinguishing will
from the rest, according to the degree of certainty they could impose on each statement with
logic and/or evidence. For this case, too, it was difficult to make any generalization on the
patterns of ESL 501 students in comparison to both ESL 500 and Exempt group students,
because there seemed to be too many overlapping factors and exceptions to make any concrete
conclusions.
All the findings combined to respond to the research question 1, it would be best to
conclude that the frequency of using a verb tense and aspect combination does not necessarily
correlate with the level of the student’s L2 writing proficiency except for only one verb structure,
present perfect. However, students who were exempted from taking any ESL courses did have
several common features among themselves, which indicated how being able to use a variety of
verb tenses and aspects could positively influence their writing to be treated as “near nativelike.” In other words, the Exempt group students seemed to know the various meanings and
functions within one verb tense and aspect structure and were able to use it appropriately in
79

context while the lower level students seemed to know only basic concepts of the verb structure.
Therefore, even in cases where Exempt students made less use of certain verb structures or
showed more cases of avoiding them, it could have resulted from their attempts to use more
complex sentence structures (e.g. if-conditionals) or to use more diverse types of evidence while
ESL 500 students kept using simpler sentences and less variety of evidence for their arguments.
In short, the level of students’ L2 proficiency revealed in their writing samples did have a
relationship with the patterns of using verb tense and aspect structures, but the distinction
between ESL 500 and ESL 501 were not clear, and it was almost impossible to justify any
generalizations without referring to the actual student samples.
2.

In continuation of the discussion from the research question 1, the results of the present

study showed that frequency of using and/or non-using a verb tense and aspect structure had
little relationship with students’ L2 proficiency, and this finding invited other factors that may
have significantly influenced their verb usage patterns to be applied to the analyses. As a large
number of scholars have emphasized the importance of “discourse” in writing, the current study
hypothesized that the academic discourse could have influenced NNS writers’ patterns in using
verb tense and aspect as it expected, disciplined, or even limited the use or non-use of certain
verb structures. From the results of the frequency counts and the analyses of student writing
samples, it was found that academic writing discourse did have influence on NNS writers’ verb
using patterns.
While it is true that higher level students were more aware of the various meanings and
functions of a verb structure, they would not have used it unless the context in which they wrote
the essay encouraged its use in the first place. For instance, although students in Exempt group
understood the multiple meanings and functions of the simple perfect, there were only 88 cases
80

of using this verb structure while the simple present was used for 897 times, which was
approximately 10 times more than the perfect. In fact, when the simple perfect was used in the
essays, its use was usually limited to the introductory paragraph of the essay, providing
background information of the topic or the issue. Since this pattern of using the present perfect in
introduction or review of an issue was observed across the level of students’ L2 writing
proficiency, it was possible to assume that there were some disciplines that the students were
following either consciously or unconsciously.
Also, the progressive aspect was rarely used in the same manner in all three groups.
These phenomena of overusing the simple present and avoiding the progressive and perfect
aspects could be attributed to the discourse genre of the paper that they were asked to write. In
other words, because the task of the EPT written exam was to write an academic essay on a
debatable topic based on a reading passage and a lecture, students were limited to using more
formal academic language, making many assertions, predictions, and claims and providing
factual evidence to support their arguments. There was very little chance for them to include
personal anecdotes in a vivid manner, insert lively dialogues, or everyday conversations in their
essays. Consequently, there was less use of progressive or perfect aspect. In addition, compared
to Hinkel’s (2004) study in which the simple past was used most frequently by the NNS, the
simple present was used most often in all essays of the current study because EPT written exam
required the students to somehow integrate given sources (e.g. news articles) into their essays.
Therefore, since Hinkel’s (2004) subjects were asked to write about their personal opinions
without having any academic sources to refer to, they used more of personal anecdotes for which
NNS writers often employed the past tense.

81

Nevertheless, while some of these patterns may have guided NNS writers to use verb
tense and aspects appropriately so as to be welcomed by the academic discourse, there were
underachieved features of the academic discourse that NNS writers often failed to apply in their
writing. For instance, the use of predictive aspect in all three groups seemed random and
inconsistent to certain extent. The frequency rates of using “predictive modals” such as will,
would, and may/might did not seem to show any meaningful patterns, except for the fact that
Exempt group students used will for the purpose of employing “if-clause conditionals” more
often than of merely making future predictions or expressing volition as was the case with the
lower level students. However, all three groups did not show clear distinction among the modals
in terms of the degree of certainty, and their choice of using will over would indicated that they
were not familiar with hedging strategy. Although the cause of this phenomenon is not
thoroughly investigated in this study, it was likely that the NNS students did not have sufficient
knowledge on the concept of hedging because most ESL/EFL textbooks do not offer clear
guidelines on when, what, and how to hedge.
In short, the influence of academic writing discourse was clearly noticed from the NNS
writers’ patterns of using verb tense and aspect, and it even had close relationship with the tone,
voice, and style of their writing. However, despite its influence on NNS writing, it could have
limited the use of certain verb tense and aspect combinations, the use of certain sentence
structures, and even the contents of their writing.
3.

Going back to the grand question of whether grammar teaching is necessary or not,

responding to this research question of whether or not to include teaching English verb tense and
aspect in university ESL writing courses offered a chance to think about the role of
understanding and using the verb usages in the development of L2 writing proficiency. While
82

many ESL instructors as well as the test raters prioritize “content” and/or “structure” as opposed
to “grammar” in both teaching and testing English as a second language, the results of the
present study suggested that “grammar” could not actually be set apart from thecontent nor
organization of the content in any kind of authentic writing tasks. From the findings on research
question responses 1 and 2 above, it was shown that how NNS students used verb tenses and
aspects in their writing was not a mere reflection of their grammar knowledge, but rather a
combination of their ability to understand multiple meanings, functions, and usages of various
verb tenses and aspects, select the most appropriate structure for the intended purpose, and apply
it to the academic discourse coherently. In other words, teaching English verb tense and aspect in
this case would not refer to simply teaching the grammatical concepts of each verb structure by
defining each form, listing its usages, or having practice drills for multiple choice questions.
Rather, it would be linking verb usages to the discourse-level features and the content of the
essay so as to help students build up logical and well-organized ideas in a coherent manner.
The benefits of teaching verb tense and aspect system with emphasis on discourse-level
coherence and content-development were actually suggested in the current study as more
advanced students made better use of the simple present, simple past, and the present perfect
structures when implementing more variety of sources to their essay. In other words, as they
were more familiar with using verb structures for more variety of functions, they were able to
develop their supporting details based on frequent references to given articles and occasional
references to past-time stories, which all together strengthened their arguments; having more
variety of resources and expressing them with appropriate verb structures made their essay more
logical and reliable. In addition, as they employed conditional statements to predict results of a

83

future event by using “if-clause” and modality, they added more flavor to their essays and gave
readers chance to imagine and extend their thoughts.
Moreover, although not all data may correspond to this generalization, Exempt group
students seemed successful in having coherence and unity in their essays by keeping their verb
structures consistent in discourse. As Witte & Faigley (1981) claims writing quality to depend
greatly on the discourse-level features that “lie beyond sentence boundaries involving the
underlying relations between ideas which allow a text to be understood” (as cited in Pongsiriwet,
2001, p.87), it would be beneficial to teach improving coherence of the paper by explicitly
addressing how discourse-level verb consistency may influence coherence of ideas in terms of
time, aspect, and tone, especially when the students are writing in strict academic discourse
settings. Thus, giving explicit instructions on understanding and using English verb tense and
aspect would not be “teaching grammar” but would rather be a practical application of
integrating grammar with content and structure which will improve NNS students’ writing
proficiency.
4.

The last question this study aimed to answer was to specify which verb tense and aspect

combinations to focus on when implementing grammar teaching to university ESL writing
courses. Based on the previous findings and discussions of the present study, it would be
reasonable to suggest focusing on verb structures that had more influence on the academic
writing discourse. First, it would be necessary to teach the use of the simple present, the simple
past, and the present perfect as they are used to report findings from other sources and make
references. Although many grammar textbooks and writing guidebooks inform students of how
to report and cite from other sources, the differences among the three verb structures in terms of
meaning and function are often blurred or not explicitly explained. Nevertheless, it was
84

perceived as one of the most problematic areas in NNS writing, especially for advanced students
who made frequent attempts to strengthen their arguments with more authoritative evidence from
reliable sources. Therefore, teaching how to choose verb tenses and aspects for reporting and
referring to other sources would enable NNS students to increase reliability and validity of their
arguments, thereby improving the overall quality of their papers.
Second, it would be critical to include teaching English modality and hedging strategies
in ESL writing courses. Among the four major aspects categorized in the current study’s
taxonomy of verb tenses and aspects, the predictive aspect was found to be most challenging for
NNS students to accurately and naturally employ in their writing because most of them seemed
to be unaware of or at least unfamiliar with the concept of hedging. Although current ESL
writing courses do address the concept of “hedging” sometime during the course, guidelines are
not specific enough to explicitly teach students what, when, and how to hedge, especially by
using proper modals. Also, the differences among English modality are not taught in detail,
which may cause students to choose modals randomly. Moreover, some instructors even
discourage students to hedge because it may weaken their arguments. Nevertheless, rather than
just tackling the topic of hedging as a choice of writing style, it would be desirable to explicitly
teach students how using different modal auxiliaries will make differences in the degree of
certainty in their claims or predictions. Then it would be students’ responsibility and choice
whether to hedge or not depending on how strong they want their assertions to sound. However,
giving students the chance to understand the differences between the modals should be done by
the teachers a priori.
Moreover, it would be critical that ESL writing courses teach how to keep verb tenses
and aspects consistent throughout the discourse. One problem prevalent in most of the lower
85

level NNS students was the related to shifting verb tenses and aspects. When students began a
sentence with the past tense but wanted to include a general truth within the same sentence, or
when they began with an event that occurred in the past and continued to the present time, they
mix-matched the verb structures without any patterns. Such inconsistency in verb structures and
random shifts of verb structures often hindered meaning deliverance and coherence of ideas,
damaging the overall quality of the writing (Pongsiriwet, 2001), not to mention the great
confusion. Therefore, it would be beneficial to have ESL writing instructors teach and have
students practice making natural shifts and flow of the verb structures so as to make their writing
more coherent and consistent.
Finally, further implications and suggestions may be made for the testing of ESL
students’ writing proficiency. The current English Placement Test practiced at UIUC evaluates
students’ ability to write L2 compositions in an academic context using only holistic method of
assessment. Although EPT has a published validation argument3which claims strength in placing
NNS students to ESL writing courses in an orderly, controlled manner, it may not be without
potential problems. The biggest concern about the current EPT may have would be the rater
reliability in the sense that each essay is rated by two raters, ESL writing course instructors,
based on a subjective, holistic rubric that mainly focuses on the “organization and content” of the
essay. Based on the findings of the present study, the distinction between ESL 500 and ESL 501
students was often blurred and difficult to perceive. In addition, there were quite a few essays
whose placement decision was initially disagreed and still remained questionable as to how the
final recalibration was made. Thus, it would beneficial to pursue further studies in this direction.

3

The UIUC English Placement Test (EPT) Fact Sheet, 2012, retrieved from
http://www.linguistics.illinois.edu/students/placement
86

Moreover, there were limitations on the range of verb tenses and aspects that students
would use regardless of their language or writing proficiency because writing task was confined
to writing formal argumentative essays. Also, the essay prompts, articles, and other references
were limited to academic sources, which could have affected their use of supporting evidence
types and the choice of verb aspects. Although it may serve the purpose of placing the NNS
students to appropriate ESL writing courses, it may deprive the students of the opportunity to
recognize their weaknesses in linguistic and rhetoric features because most raters were trained to
ignore language features unless they “hinder the understanding to a significant degree.”
However, as the students are in various fields of studies and are likely to work in more variety of
areas, their writing ability should not be confined to using only some of the verb structures. It
would be ideal to offer students a chance to learn the overall picture of the English verb tense
and aspect system, and it could be possibly done by complementing the current EPT holistic
assessment with an analytic assessment that may check students’ knowledge on the most
essential grammatical concepts and their applications. Or, it would also be possible to develop
the essay questions to solicit answers that would have to include more variety of sentence or verb
structures. A possible example could be: “What is your own stance on the performance-based
payment for teachers? Make reference to the reading passage and given lecture as well as your
own examples. In addition, if the new payment system were put into practice, what would be the
most ideal criteria for evaluating teachers’ performance?” By extending the essay question and
giving more specific directions to use more variety of structures, the EPT would offer raters
more objective grading criteria (e.g. appropriate verb uses, inclusion of required evidence, etc.)
and students a chance to reflect on their current ability to incorporate grammar knowledge into
actual writing practice.
87

Limitations and Suggestions for Further Research
Despite its great effort to design and conduct the study in the most logical and specific
manner, the current research is not without limitations. First, the main content of the study, “verb
tense and aspect” has to be clearly defined. There was an attempt to present a new taxonomy of
English verb tense and aspect system in the phase of designing the study, and it seemed
reasonable enough to categorize the verb tense and aspect combinations into four major aspects
(i.e. simple, perfect, progressive, and predictive). However, it does not include every verb
structure used in the English language, and it may trigger objections from those who would claim
verb structures to be grouped by the timeframe rather than aspectual frame, or to be more
specifically classified into smaller groups. In particular, analyzing the “predictive aspect,” which
is used to make predictions or assertions, could be more thoroughly analyzed by distinguishing
the usages of will, would, and may/might by different functions and meanings, if it is possible.
For example, when a student used will in a sentence to address his opinion which makes
reference to the future timeline, it was difficult to tell whether it was used as an “epistemic
modality” or as a “modality of future.” Moreover, the category of “etc.” remains ambiguous in a
sense because the frequency of its use was noticeably high. The use of the modal auxiliaries in
this category (e.g. can, could, must, etc.) was not considered in this study because these modal
auxiliaries conveyed meanings in addition to the “predicting” logical outcomes; however, it
would be beneficial to have a close analysis of the other modals or to add these to a new aspect
in the future research.
Second, whether these essays written for EPT exam would truly and fully reveal students’
writing style could be questionable because of the time limit during the test and the variation of
the essay topics. Many students who took the EPT have complained that they fell short of time. It
88

is possible that students would show different tendency to use verb tenses and aspects if they
were to write the same type of essay at home or in class without time limit. In addition, some
students might feel more confident in one topic than another. EPT written exam has provided
students with various means to overcome this difficulty by offering discussion session,
opportunity to prewrite and to be peer-reviewed before writing the final draft; however, there
may still be possibility that the familiarity of the topic would affect the overall quality of the
paper. Therefore, for future studies, it would be meaningful to compare or include student
sample essays that had similar format and tasks but with no time limits and with topics that are as
similar as possible.
Third, the subjects were limited to NNS students whose L1 was either Korean or Chinese.
This choice was based on the assumption that there would be no big L1 influence on their choice
of verb structures and also on the fact that the author of the current study was more familiar with
these two languages. However, there is still room for debate about the L1 influence, so it would
increase reliability of the study to include more subjects with more variety of language
backgrounds.
Finally, the essays which were initially disagreed on their placement decision were
labeled as “D” in the student samples, but they were neither purposefully selected in even
numbers nor analyzed in separation to the other essays. If the same number of the “D essays”
were collected from all ESL 500, ESL 501, and Exempt group, and if they were analyzed in
more details, there may have been different patterns or more specific patterns observed from the
analyses. Then, it would be useful to have future studies select the subject essays based on a
more systematic category and fully utilize the data for analysis. Furthermore, the rater reliability
should also be reinforced prior to selecting and analyzing the “D essays” because there may be
89

some doubts and questions about the consistency of the standards each rater has. Not only for the
“D essays” but also for the overall essay classification process because the fine lines between
ESL 500 and ESL 501 or ESL 501 and Exempt could be ambiguous. In fact, sometimes the
differences on verb usage patterns were clearly observed between ESL 500 and Exempt,
suggesting the positive relationship between the NNS students’ verb usage patterns and their
writing proficiency; however, with the results from the ESL 501group, the relationship became
less salient. Thus, increasing the reliability of the raters and consequently that of the level
assignments (i.e. ESL 500, ESL 501, and Exempt Group) would greatly benefit the process and
results of the data analyses.

90

REFERENCES
Azar, B. S. (2000). Understanding and using English grammar (3rd ed.). White Plains, NY:
Pearson Education.
Batstone, R. (1995). Grammar in discourse: Attitude and deniability. In Cook, G. & Seidlhofer,
B. (Eds.), Principle and practice in applied linguistics: Studies of honour of G. G.
Widdowson, p.197-213. Oxford: Oxford University Press.
Bennett, T. L. (1977). An extended view of verb voice in written and spoken personal narratives.
In Keenan, E. O. & Bennet, T. L. (Eds.). Discourse across time and space, 43-50, Los
Angeles, CA: University of Southern California.
Biber, D. (Eds.). (1999). Longman grammar of spoken and written English. Harlow, England:
Longman.
Bryant, W. H. (1984). Typical errors in English made by Japanese ESL students. JALT
journal, 6, 1-18.
Celcia-Murcia, M., & Larsen-Freeman, D. (1999).The grammar book (2nd ed.) Boston, MA:
Heinle & Heinle.
Comrie, B. (1976). Aspect: An introduction to the study of verbal aspect and related problems.
New York, NY: Cambridge University Press.
Coxhead, A., & Byrd, P. (2007). Preparing writing teachers to teach vocabulary and grammar of
academic prose. Journal of Second Language Writing, 16, 129-147.
Doughty, C., & Williams, J. (Eds.). (1998). Focus on form in classroom second language
acquisition. New York, NY: Cambridge University Press.
Ellis, R. (2006). Current issues in the teaching of grammar: An SLA perspective. TESOL
Quarterly, 40(1), 83-107.
91

Gass, S. M. & Mackey, A. (2002). Frequency effects and second language acquisition. Studies in
Second Language Acquisition, 24(2), 249-260.
Gass, S. M., & Selinker, L. (2008). Second language acquisition: An introductory course (3rd
ed.). New York, NY: Routledge.
Givon, T. (1984). Functionalism and grammar. Philadelphia, PA: John Benjamins Publishing
Co.
Gray, B. E. (2011). Exploring academic writing through corpus linguistics: When discipline tells
only part of the story (Doctoral dissertation). Northern Arizona University (UMI No.
3490519).
Hantrakul, C. (1990). English tense and aspect usage in controlled written discourse by nonnative speakers (Doctoral dissertation). Illinois State University (UMI No. 9101113).
Hewings, M., & Hewings, A. (2002). “It is interesting to note that…”: a comparative study of
anticipatory ‘it’in student and published writing. English for Specific Purposes, 21(4),
367-383.
Hinkel, E. (2004). Tense, aspect, and the passive voice in L1 and L2 academic texts. Language
Teaching Research, 8(1), 5-29.
Hinkel, E. (1997). The past tense and temporal verb meanings in a contextual frame. TESOL
Quarterly, 31(2), 289-313.
Hirtle, W. H. (1967) The simple and progressive forms: An analytical approach. Montreal,
Québec: Presses de l'Université Laval.
Hofmann, T. R. (1996). Past tense replacement and the modal system. Computational
Laboratory, Harvard University, NSF Report 17.
Irujo, S. (1986). A piece of cake:/earning and teaching idioms. ELT Journal, 40 (3), 236-242.
92

Inoue, K. (1979). An analysis of the English present perfect. Linguistics, 17, 561-589.
Kim, I. H. (1983). An analysis of errors in English compositions written by selected Korean and
Thai students: its implications for teaching writing in EFL classrooms (Doctoral
dissertation). Available from ProQuest Dissertations and Thesis database. (UMI No.
19840710)
Knowles, P. L. (1979). Predicate markers: A new look at the English predicate system. Cross
Currents, 2, 21-36.
Laufer, B., & Eliasson, S. (1993). What causes avoidance in L2 learning: L1-L2 difference, L1L2 similarity, or L2 complexity?. Studies In Second Language Acquisition, 15(1), 35-48.
Li, J. (1996). Underproduction does not necessarily mean avoidance: Investigation of
underproduction using Chinese ESL learners. Pragmatics and Language Learning, 7,
171-187.
Louda, S. (1981). Avoidance ruled out by similarity. Paper presented at the Central States
Conference on the Teaching of Foreign Languages, Omaha, NE.
Mattar, H. (2002, 2003). Is avoidance ruled out by similarity? The case of subordinating
conjunctions /Adverbs in English and Arabic. Poznan Studies in Contemporary
Linguistics, 38, 103-115.
McCoard, R.W. (1978). The English perfect tense choice and pragmatic inferences. Amsterdam:
North Holland Publishing Company.
Meyer, F. S. (1992). A text-type and move analysis study of verb tense and modality distribution
in medical English abstracts. English for Specific Purposes, 11(2), 93-113.
Meziani, A. (1984). The English tense system: A pedagogic analysis. International Review of
Applied Linguistics in Language Teaching. 26(4), 281–294.
93

Norris, J. M., & Ortega, L. (2000). Effectiveness of L2 instruction: A research synthesis and
quantitative meta-analysis. Language Learning, 50(3), 10.1111/0023-8333.00136
Osborn, J. M. (2009). Intersections: Academic discourse and student identities in a community
college writing class (Doctoral dissertation). Available from ProQuest Dissertations and
Thesis database. (UMI No. 3384926)
Palmer, F. (1990). Modality and the English Modals (2nd ed.). London: Longman.
Palmberg, R. (1983). On the use of lexical avoidance strategies in Foreign-Language
communication. Psycholinguistics and Foreign Language Learning, Papers from a
conference (Stockholm, Sweden and Abo, Finland).
Paltridge, B. (2004). Academic writing. Language Teaching, 37(2), 87–105.
Pongsiriwet, C. (2001). Relationship among grammatical accuracy, discourse features, and the
quality of second language writing: the case of Thai EFL learners (Doctoral dissertation).
Available from ProQuest Dissertations and Thesis database. (UMI No. 20010828)
Pytleski, P. D. (2008). Academic discourse and cultural studies—collision, connection, or
confusion? (Doctoral dissertation). Available from ProQuest Dissertations and Thesis
database. (UMI No. 3316882)
Reid, J. (1993). Teaching ESL writing. Englewood Cliffs, NJ: Regents/Prentice-Hall.
Reid, J. (2006). Essentials of teaching academic writing. Boston: Houghton Mifflin.
Richards, J. C. (1985). Error analysis: perspectives on second language acquisition. London:
Longman.
Roberts, F., & Cimasko, T. (2008).Evaluating ESL: Making sense of university professors’
responses to second language writing. Journal of Second Language Writing, 17, 125-143.

94

Saito, K. & Poeteren, K. V. (2012). Pronunciation-specific adjustment strategies for
intelligibility in L2 teacher talk: results and implications of a questionnaire study.
Language Awareness, 21(4), 369-385.
Schachter, J. (1974). An error in error analysis. Language Learning, 24, 205-214.
Schachter, J., & Celce-Murcia, M. (1971). Some reservations concerning error analysis. TESOL
Quarterly, 11, 441-451.
Schiffrin, D. (1981). Tense Variation in Narrative. Language, 57(1), 45-62.
Shaughnessy, M. P. (1976). Diving in: An introduction to basic writing. College Composition
and Communication, 27(3), 234-239.
Smit, D, W. (2004). The end of composition studies. Carbondale, IL: Southern Illinois Press
Suh, K. H. (1992a). A discourse analysis of the English tense-aspect-modality system.
(Unpublished Ph.D. dissertation in Applied Linguistics). Los Angeles, CA: University of
California.
Swan, M. (2005). Practical English Usage. Oxford: Oxford University Press.
Swan, M. (2007). Grammar, meaning, and pragmatics: Sorting out the muddle. Teaching English
as a Second or Foreign Language, 11(2), 1-10.
Swan, M., & Smith, B. (2001). Learner English: A teacher's guide to interference and other
problems. New York, NY: Cambridge University Press.
Tregidgo, P. S. (1974). English tense usage: A bull’s eye view. English Language Teaching,
28(2), 97-107.
Ubol, C. (1981). An error analysis of English compositions by Thai students. SEAMEO Regional
Language Centre.

95

VanPatten, B., Williams, J., & Rott, S. (2004). Form-meaning connections in second language
acquisition. Mahwah, NJ: Lawrence Erlbaum Associates, Inc., Publishers
Wallwork, A. (2011). English for writing research papers. Springer Science+ Business Media.
Witte, S. P., & Faigley, L. (1981). Coherence, cohesion, and writing quality. College
Composition and Communication, 32(2), 189-204.
Zamel, V. (1987), Recent Research on Writing Pedagogy. TESOL Quarterly, 21, 697–715.
Zhou, A. A. (2009). What adult ESL learners say about improving grammar and vocabulary in
their writing for academic purposes. Language Awareness, 18(1), 31-33.

96