AL Forum

AL Forum News, Volume 29:2 (May 2009)

by User Not Found | 11/02/2011

AL Forum

In This Issue...

  • Leadership Updates
    • A Message From the ALIS Past Chair
    • A Message From the ALIS Chair
    • A Message From the Editors
  • Articles and Information
    • Two Perspectives on the Role of Cohesive Devices
    • Written Teacher Feedback on Coherence and Its Effects on ESL Student Essay Revisions
    • A Theme/Rheme Approach to Writing Instruction: From Clause to Essay

Leadership Updates

A Message From the ALIS Past Chair

Ali Shehadeh, Department of General and Applied Linguistics/TESOL, United Arab Emirates University,

Dear ALIS members,

As outgoing chair, this issue marks the end of my 2-year active involvement in our ALIS. I would therefore like to express my appreciation and gratitude to you for giving me the opportunity to serve TESOL, our IS, and you during the past 2 years. This has been an enriching experience for me: I organized and chaired an Academic Session, organized 12 Discussion Group sessions, co-organized and cochaired two InterSection sessions, and attended several IS-related meetings at the annual convention (including the ALIS open business meeting, IS assembly, and IS planning meeting). In the year to come, I will continue to serve as past chair with Scott Phillabaum as the chair, and Howard Williams as the chair-elect.

ALIS Business Meeting: The ALIS annual open business meeting at the Denver convention was held on Thursday, March 26. Several issues were discussed in the meeting. These included

  • the move to online voting: As per the board of directors’ instructions (that all ISs must move to online voting), online voting was discussed and approved unanimously in the meeting.
  • activation of the electronic discussion list and ways of encouraging members’ participation.
  • elections: A new chair-elect, newsletter coeditor, an e-list administrator, and a web content manager were elected (more on these in Scott’s message).
  • membership retention: I reported to the meeting that in a 10-month period (May08-Feb09), more than 280 ALIS members have allowed their membership to expire. Other similar figures apply to other ISs too. There was a strong feeling in the meeting that TESOL must look into the causes of the high rate of membership dropouts. As reported in my December message, three main reasons for the high rate of membership dropouts can be singled out: (a) members want to more explicitly see and feel the direct benefit and connection between TESOL and their professional needs and careers; (b) current membership fees might be unaffordable to some (unemployed or international) members; and, very important, (c) attending the annual convention, the main event of the TESOL association, is simply very difficult—if not impossible—for many members (especially when it comes to expensive lodging and travel). It was decided that a motion must be drafted to present to the IS assembly and then on to the board of directors on ways of reducing the cost of attending the annual convention.

InterSections: The Applied Linguistics IS sponsored an InterSection with the Intercultural Communication IS on Plagiarism, Culture and Ethics. Four experts were invited to speak in the session: Dilin Liu, Jane Hoelker, Leila Mohanna, and Susan Coakley. The session addressed questions such as What is plagiarism? How does the concept of plagiarism differ in different cultures? Is plagiarism an academic or ethical issue? The panelists approached these and other questions from different cultural perspectives and backgrounds based on their research and experiences in the field of applied linguistics and intercultural communication. The session was well attended. It also induced a lively discussion between the panelists and the audience.

We also cosponsored two other InterSections (as secondary sponsors). The first was with the Second Language Writing IS titled Writing Across the Curriculum and Applied Linguistics: Research and Practice. Invited speakers were Ken Hyland and Fredricka Stoller. The session was also well attended. It too involved a lively discussion between the panelists and the audience that went beyond the time allocated to the session. The second InterSection with the Materials Writing IS was unfortunately cancelled as one of the three presenters had to withdraw because of ill health and the other because of an unforeseen professional commitment. It was decided that with one panelist only, the InterSection wouldn’t be as effective or successful as it was originally planned to be.

In closing, I’d like to say once more that it was an honor to have been given the opportunity to serve the ALIS for the past 2 years. I look forward to continue to serve our association and you in one way or another.

Ali Shehadeh

A Message From the ALIS Chair

Scott Phillabaum, Linguistics & Language Development, San Jose State University,

Greetings once again, ALIS members! Having just returned from a successful TESOL convention in Denver, I’m pleased to have the opportunity to tell you about some of the highlights of the convention and update you on what’s ahead for the coming year.

ALIS Academic Session: Once again the ALIS Academic Session was a great success with strong attendance and enthusiastic participation. This year’s topic—“Interaction, Grammar, and Learner Strategies in L2 Pragmatics”—featured presentations by Marianne Celce-Murcia, Virginia LoCastro, Kathleen Bardovi-Harlig, and Andrew Cohen. Touching on a subject of growing interest in TESOL, all four presentations provoked lively discussion and contributed nicely to the ongoing conversation about what constitutes L2 pragmatic competence and how best to facilitate the acquisition of L2 pragmatics in the classroom. All four presenters have agreed to write a brief summary of their presentations for the fall AL Forum, so be on the lookout for that in November.

Interest Section Leadership Meeting: A number of important issues were raised at the IS Leadership Meeting. First, IS members are eligible to apply for a TESOL Interest Section Special Project. The purpose of these projects is to provide members of the IS and members of TESOL with information related to research, advocacy, resources, and so on. The deadline for applications for fiscal year 2010 is May 1, 2009, so a member not be able to submit an application this year, but I encourage you to consider submitting a proposal for next year. A second issue that was raised at the leadership meeting and that was stressed at other meetings throughout the convention is the TESOL Resource Center (TRC). The TRC is an online repository where TESOL members can share various resources with other members in the profession. Because the TRC is accessible around the globe, it holds the potential to be a great resource for all of TESOL’s international membership. Moreover, all of the submissions to the TRC are peer-reviewed so your contributions support your own professional development as well as that of others. The TRC is also looking for individuals who would like to serve as reviewers. Information on how to become a reviewer is available on the TRC Web site: As additional incentive to encourage members to contribute to the TRC, TESOL is running the Charles River Contest. If you submit two resources to the TRC by June 20, 2009, you will become eligible to win two free nights of lodging in Boston for TESOL 2010.

ALIS Business Meeting: A number of issues were discussed at the ALIS business meeting that I will leave for Ali’s report. We voted in a number of new officers at the meeting and I would like to take the opportunity to introduce and welcome them. Howard Williams was unanimously selected as the chair-elect. As many of you know, Howard served as chair of the ALIS a few years ago and we’re very pleased to have him back. David Olsher was also unanimously elected to continue as e-list administrator. We also elected a new coeditor for the AL Forum newsletter to replace Lorena Llosa, who has reached the end of her 3-year tenure as coeditor. Beginning with the fall issue of the AL Forum, Casey Keck will join Priya Abeywickrama as coeditor. Casey also volunteered to work as our Web content manager. I’d like to welcome both Howard and Casey and I look forward to working with them both in the coming year.

TESOL 2010: At present, the ALIS is putting together an InterSection that will focus on conversation analysis and that will be cosponsored with the Intercultural Communication Interest Section. Also, in early June we will begin vetting abstracts for next year’s convention. If you would like to serve as a reviewer, please contact me as soon as possible at

Looking Ahead: As we move forward this year, I would like to encourage you to participate in a number of projects. First, the ALIS needs to improve its overall visibility within TESOL. One way that we hope to do this is by updating our webpage and by maintaining better contact with our members. If you have ideas about what sort of Web content would appeal to our members or about how to better communicate with our members, please let me know. Second, the ALIS would like to expand its membership and to reconnect with applied linguistics researchers, scholars, and students. The ALIS is one of the oldest interest sections in TESOL, dating back to 1974; early members included Bernard Spolsky and Robert Kaplan. Over the years, we’ve witnessed a sometimes strained relationship between “researchers” and “practitioners” and a number of comments that I heard at this year’s AAAL and TESOL conventions attest to this. Let’s make this year a time of reconnecting and reinvigorating our ties with applied linguists. If you’re interested in helping out or if you have ideas about how to go about this, please let me know.

That’s all for now. Talk to you again in the fall!


A Message From the Editors

Lorena Llosa, New York University,, and Priyanvada Abeywickrama, San Francisco State University,

Greetings everyone! We hope you were able to attend TESOL 2009 in Denver!

In this issue (29.2) we are pleased to feature three articles by presenters of last year’s TESOL (New York) Applied Linguistics and Second Language Writing InterSections. The InterSection, which was entitled “Textual Coherence and Learner Writing,” provided an opportunity to highlight a topic of relevance across disciplines.The authors approach the topic in three ways: theory, research, and pedagogy—that is, a theoretical perspective of the pragmatics of textual coherence, an analysis of written work and teacher feedback, and a grammatical approach to teaching textual coherence in writing.

Howard Williams discusses recent developments in pragmatics, especially relevance theory, which views coherence in the context of text processing and thus requires writers to assess the likely cognitive resources of readers. Next, Aziz Yuldashev reports on a study that examined ESL students’ use of cohesive resources in their writing and teacher feedback. The findings suggest the need for context-based comments that relate to classroom instruction to help students develop more specific and practical ways to achieve coherence in their writing. The final article by Duane Leonard examines the pedagogy of textual coherence. Specifically, the author focuses on the theme/rheme division in functional grammar as a resource to help students develop textual coherence in their writing.

Finally, this issue marks the end of Lorena’s tenure as coeditor of the ALIS Forum. Over the past 3 years she has thoroughly enjoyed working with all contributing authors and especially with her two coeditors, Scott Phillabaum and Priya Abeywickrama. Undoubtedly, Priya and her new coeditor, Casey Keck, will continue to produce a newsletter packed with interesting and thought-provoking pieces.

We thank the authors for submitting their contributions and hope you enjoy this issue of the ALIS Forum.

Articles and Information

Two Perspectives on the Role of Cohesive Devices

Howard Williams, Teacher College, Columbia University,

ESL/FL composition textbooks above the low-intermediate level generally contain sections on coherence. Their treatment of the topic tends to center attention on the so-called cohesive devices—the various expressions that are claimed to signal the identity of one noun to another noun, or the connection of one sentence to another sentence, or the content of one paragraph to the content of another paragraph. Sometimes these are referred to as transitional expressions, discourse connectives, or conjunctive elements, and they are typically classified into types. Textbooks stress the importance of these expressions in the creation of texts that “hang together” or sound coherent.

Two learner problems with connectives are underuse and overuse. (Their completely erroneous use is a third problem that will not concern us here.) Where underuse occurs, there may be a comprehension problem. Here is an actual student example:

(1) A woman does not always believe what a man says literally. She may give him [the] benefit of a doubt. Usually she looks for a hidden meaning.

This sequence, which may require several readings to comprehend the intended juxtaposition, might benefit from the recasting of the second and third clauses with clause 2 prefaced by although. For learners at the higher levels, this sort of problem may occur only infrequently because those learners may already have been bombarded by directives to use transitional expressions: Many writers have learned the lesson well by using them often and everywhere possible.

This brings us to the second problem. Example (2) is a slightly altered version of an actual student essay:

(2) According to [Tannen’s] essay, there are behavioral differences between a man and a woman. For example, little girls think of a group of people as one natural unit. In other words, women always want to be considered as one of a group. Therefore, when one of them tries to stand out of the group, [she] may [lose her membership in the group]. In contrast to that, boys always try to be the best in any category as an individual in the group. Thus, they do not care about a group [in the same way that] girls do. In short, boys and girls are very different from each other psychologically.

This learner has overcome the deficit of the first type of composition by chaining sentences together with conjunctive adverbials as links—but in a manner that is not, in fact, witnessed often in actual published texts. One might call this passage “overcohesive.”

In what follows, I sketch two views of the so-called cohesive expressions and argue for a more enlightened vision of their place in materials development.

The “Discourse Glue” Approach to Coherence

Composition textbooks that treat discourse connectives take their inspiration—either directly or indirectly—from Halliday and Hasan’s (1976) Cohesion in English, the first book-length treatment of the question “Where does coherence in discourse come from?” and which is still quite influential in corpus studies. The authors’ answer to the question was that degree of coherence is roughly proportional to the number of cohesive devices (correctly) used. These devices, according to the authors, fall into five basic types. One type is reference elements, such as pronouns, which are indexed, mainly anaphorically, to lexical elements that appear elsewhere in the text. So in sentence (3),

(3) I bought a new car. I liked it the first time I laid eyes on it,

the writer achieves coherence through the use of the pronoun it. A similar but distinctive type is the substitution element such as the word one, as in (4):

(4) I bought a new car. My friend bought one, too.

Here, the pronoun one co-indexes with new car and even though the writer is not referring to exactly the same car, he or she still creates a pronominal bond through a set relationship. Coherence may also be achieved by ellipsis, whereby a normally required sentential element is omitted; the omission forces the reader to look backward in the text to identify the omitted phrase—and thereby to recognize the existence of a textual link. Thus in (5),

(5) I bought a new car. My friend did [ ], too.

only buy a car may fill the empty slot.

A fourth type of cohesive device is lexical cohesion, in which a lexical element that has an identifiable semantic relation to another lexical element in the text is employed. Revising the same sequence in (4), we get (6):

(6) I bought a new Toyota. The car cost me $20,000.

By interrelating Toyota with car, and by interrelating bought with cost and $20,000, the textual parts are said to achieve a bond. This is so, for example, because Toyotas are known to be a subset of all cars.

Finally, Halliday and Hasan (1976) referred to conjunctive expressions, which comprise traditional coordinating and subordinating conjunctions and conjunctive adverbial expressions:

(7) He said he bought a new Toyota, but it was a Honda. (coordinating conjunction)

(8) He adored his new car although he did not take proper care of it. (subordinating conjunction)

(9) The iron oxidized. In other words, it got rusty. (conjunctive adverbial)

For the authors, all such linguistic devices constitute what they refer to as “discourse glue.” They are explicit signals of links across various parts of the text, and without them—as the metaphor implies—the text is supposed to fall apart, to collapse, to come across as incoherent at a local, if not also a global, level. This perspective on coherence has strongly influenced ESL materials. Consider the quotation below from a chapter in Oshima and Hogue’s (1999) well-known Writing Academic English:

(10) There are four ways to achieve coherence. The first two ways involve repeating key nouns and using pronouns that refer back to key nouns. The third way is to use transition signals to show how one idea is related to the next. The fourth way . . . is to arrange your sentences in logical order [i.e., by using temporal and ordering expressions such as first...second...finally]. (pp. 40-41)

A little later, they included a chart on which the various transition signals and the temporal and ordering expressions are listed, together with their basic functions. This sort of chart is found in many composition texts. Students are encouraged to use these expressions. And the more serious students really do use them—often to the point where teachers cry “Enough!” Paragraph (2) above is a fairly typical example, and not really a bad one, because the student was more or less on target with the meanings. But the general impression is that it expresses more than is necessary. Crewe (1990) suggested that writing instructors consider deemphasizing the teaching of connectives for just such reasons.

The View From Pragmatic Theory: Connectives As Processing Instructions

Current pragmatic theory (see, e.g., Blakemore 1987, 1992, 2002) argues that the discourse glue approach to coherence is fundamentally misguided in treating text as a static entity without consideration of the interaction taking place between a reader and a writer, each aware of the other’s presence and needs. A better model asks what takes place in a hearer’s or reader’s mind in the course of processing text and what a writer can count on a reader to understand using his or her own resources. So, for example, consider a sequence like (11):

(11) We won’t be taking a vacation this year. The money just isn’t coming in the way it used to.

The sequence seems comprehensible enough as it stands, yet a model like Halliday and Hasan’s (and, by extension, Oshima & Hogue’s) predicts incoherence (or weak coherence) because there are no pronominal ties, ellipted phrases, connective expressions, or obvious lexical links; any claimed frequency collocation between vacation and money cannot take us very far toward overall comprehension, and it misses the essential point. We make sense of the sequence not because of lexical cohesion but because we are able to bring world knowledge into play in interpreting the second sentence in the context of the first. That knowledge tells us that vacations tend to cost a fair amount of money and that if a person has little of it, a vacation might be forfeited. It is unnecessary to express these facts.

The second sentence in (11) mentions money. How do I know that the money referred to is the speaker’s money? In theory, it could be anyone’s money. For those familiar with Grice’s (1975) cooperative principle and maxims, the explanation has to do with the implicit contract that exists between interlocutors—that conversational exchanges are governed by presumed adherence to a set of pragmatic norms, one of which is the Maxim of Relation. In this case, sentences like the second one will be relevant to the first. Grice’s maxims also specify that we operate with presumed norms governing the amount of information that one expects from an interlocutor (i.e., the right amount rather than too much or too little) and the quality of that information (i.e., truthful and accurate information). All in all, these constitute a backgrounded but very real contract between individuals—a contract that tends to be noticed only when it is broken in one manner or another.

Relevance theory (Sperber & Wilson, 1995), which has been developed since Grice, has taken up the idea of relevance as being at the center of all of these smaller principles. Relevance theory argues that all human communication is governed by a hearer’s/reader’s search for the most relevant interpretation of an interlocutor’s utterance. The Communicative Principle of Relevance says,

(12) Every act of overt communication communicates a presumption of its own optimal relevance. (Sperber & Wilson 1995, p. 260)

How do we know which interpretation is optimally relevant? The intuitive answer is “the first interpretation that comes to us naturally”; a more explicit answer is to call it the one most consistent with the hearer’s set of beliefs about the world and that imposes the least cost in terms of processing time and energy; an optimally relevant interpretation offers an adequate payoff for the time spent processing. But why would we expect that particular interpretation—the least costly one—to be the right one? The answer lies in the sort of contract that Grice was talking about. For example, we expect that a person will not go out of his or her way to say something in an overly complicated or obscure way that could be said in a simpler or clearer way, as that takes up more energy than it returns in rewards.

How can this be applied to example (11)? When someone says, “We won’t be taking a vacation this year. The money just isn’t coming in the way it used to,” it is not absolutely inconceivable that the speaker might be making two separate, unrelated statements. However, the scenario is unlikely. The two sentences were closely juxtaposed, and there is a mutual presumption that they will be interpreted as somehow connected; moreover, the speaker knows the hearer will seek the shortest path to a reasonable interpretation consistent with the hearer’s knowledge of the world. When someone utters the first sentence, it is natural to seek a reason for the missed vacation. The speaker knows that this query is natural. It is not so much that an explanation must be provided as that one is expected. In light of this strong mental alignment on the part of the two parties, and in light of mutual assumptions of the existence of that alignment, the second assertion comes across immediately as a comprehensible explanation for the first. Moreover, it is just as easily comprehensible to outside observers (such as to those reading a dialogue, for example) as it is to the parties conversing. The exchange is perceived as coherent. No connective like After all or You see is needed to connect the segments; the use of either would highlight what is already clear and might therefore even be seen as slightly condescending.

The upshot is that it is possible to think of coherence as independent of cohesion. (This point was made in a somewhat different way in 1982 by Pat Carrell.) Still, we do have this tendency to associate coherence with the presence of so-called cohesive devices; moreover, studies do exist (e.g., Spyridakis & Standal, 1987, Millis & Just, 1994) that show that comprehension and recall may be aided by the presence of such devices.

Blakemore (1987, 2002), who has published extensively on the topic of connective expressions, suggested that such expressions show up in languages not for the purpose of gluing segments together but because they provide shortcuts to optimally relevant interpretations. These shortcuts enable an interlocutor to signal relationships using relatively little verbiage, which is consistent with the principle of achieving maximal benefit for least processing effort. But why would we need them at all if we managed without them in example (11)?

The answer is that not every case is as straightforward as (11). We might be confronted with two sentences like the ones in example (13), with no cohesive expression supplied (let us assume that readers have no access to a larger context that would make the connection clear):

(13)The guests were coming soon. (___,) The wastebasket was empty.

(a) After all, . . .

(b) However, . . .

Each connector will give rise to different assumptions and therefore connect in a unique way. If we heard After all, we might access an inference like “We expect of the person running the household to have taken out the trash prior to guests arriving.” Blakemore suggested that after allinvites a hearer to set up a kind of informal “if X, then Y” kind of structure—a logical argument with premises and conclusion. With the (b) option (however), we have to search a little harder for a relevant connection. Perhaps the person being talked about was supposed to be preparing an elaborate dinner so that by now, we’d expect the wastebasket to be full of kitchen garbage. Blakemore called however an expression that invites us to alter an assumption that we would entertain on the basis of the first sentence alone—namely, that if the guests were coming, the wastebasket would be full of kitchen waste. The connectors lead us in two different interpretive directions; had we chosen furthermore as the connector, we would be constraining our interpretation in still another way. However, each connector constrains in a fairly consistent, conventional way, which is why we can speak of a writer’s competence in the use of connectors. Blakemore drew the distinction between semantic meaning, the type in which a word like horse is said to point to a certain real-world entity, and procedural meaning, which consists of a set of interpretive instructions to a hearer/reader; connective expressions are of the latter type. They exist in languages because they enable shortcuts to communication. Though a speaker or writer could provide complete explanations (as above) for why two sentences are juxtaposed, the result would make for highly inefficient discourse.

It should be clear by now how different this view of connective expressions is from the notion that we use them to “paste sentences together.” From the perspective of someone like Blakemore, connectors certainly do contribute to coherence, but they can frequently—maybe more often than not—be dispensed with when they contribute nothing to it; they can even detract from coherence when we have every reason to believe that our readers will be able to discern the connections on their own.

How can this view change the way we present coherence in classrooms? On the one hand it makes things harder because it reduces the value of simply throwing a list of expressions at students and saying, “Use these to make your essay hang together.” On the other hand, it may make things easier in the long run because it asks learners to do something that they should be doing anyway—namely, considering what their audience already knows or is likely to figure out for themselves from their own resources. Pragmatic proficiency, after all, entails interactional skills. Instructors need to guide writers in deciding not only what is required in the reader/writer interaction but also what is better left unsaid.


Blakemore, D. (1987). Semantic constraints on relevance. Cambridge, England: Cambridge University Press.

Blakemore, D. (1992). Understanding utterances. Oxford, England: Blackwell.

Blakemore, D. (2002). Relevance and linguistic meaning. Cambridge, England: Cambridge University Press.

Carrell, P. L. (1982). Cohesion is not coherence. TESOL Quarterly, 16, 479-488.

Crewe, W. J. (1990). The illogic of logical connectives. ELT Journal, 44, 316-325.

Grice, H. P. (1975). Logic and conversation. In P. Cole & J. Morgan (Eds.), Syntax and semantics 3: Speech acts (pp. 41-58). New York: Academic Press.

Halliday, M. A. K., & R. Hasan. (1976). Cohesion in English. London: Longman.

Millis, K. K., & M. A. Just. (1994). The influence of connectives on sentence comprehension. Journal of Memory and Language, 33, 128-147.

Oshima, A., & A. Hogue (1999). Writing academic English (3rd ed.). White Plains, NY: Longman.

Sperber, D., & D. Wilson (1995). Relevance (2nd ed.). Cambridge, MA: Harvard University Press.

Spyridakis, J. H., & T. C. Standal. (1987). Signals in expository prose: effects on reading comprehension. Reading Research Quarterly, 22, 285-298.

Written Teacher Feedback on Coherence and Its Effects on ESL Student Essay Revisions

Written Teacher Feedback on Coherence and Its Effects on ESL Student Essay Revisions

Aziz Yuldashev, University of Illinois at Urbana-Champaign,

Coherence, a concept that appears to elude teacher-friendly explanation and learner-friendly classification, tends to receive little emphasis in ESL classrooms. In an effort to effectively supplement limited instructional time devoted to teaching coherence, some teachers turn to written feedback as a means to help students improve coherence in their writing. The purpose of this study is to investigate the effects of written teacher feedback on coherence in ESL learner writing.


Coherence has been thought of and written about in a variety of ways. Some researchers viewed it as a textual property (Kintsch & van Dijk, 1978; Hoey, 1991); some other researchers underscored the role of a reader in the construction of coherence (Brown & Yule, 1983; Williams, 1985). The latter view acknowledges the significance of linguistic resources (e.g., transitions) in the text, but maintains that the realization of coherence depends on the interaction between a reader and a text. Though coherence can be explored from the point of text-based, reader-based, and other perspectives, "it is important to note that the writer, the text, and the readers all interact in the construction of coherence" (Lee, 2002, p. 138). A functional approach to research coherence in second language learning settings, then, examines this interaction by looking into the context of an ESL writing class where learners write drafts of an essay that are to be read and commented on by a teacher for subsequent revisions. This approach was employed in the present study within the context of a university-level ESL academic writing class for graduate students.

Coherence is recognized by many teachers as crucial for good writing; however, it is not easy to teach (Cerniglia, Medsker, & Connor, 1990). Lee (2002) argued that ESL writing instructors talk about coherence "in abstract terms" instead of providing clear and systematic explanation of how learners can achieve coherence. Lee noted that in an attempt to help students deal with issues related to coherence, instructors tend to resort to abstract comments, such as "your writing is not coherent; your writing lacks unity; the ideas don't hang together; the ideas are disorganized" (p. 136). The present study sets out to explore the comments teachers provide on student essays and answer the following question:

How does the nature of written feedback on coherence--abstract or otherwise--affect student revisions?


In an effort to explore how teachers give feedback on coherence and what changes students make to their drafts based on that feedback, 106 student papers produced by 44 students in an Introduction to Academic Writing course for international graduate students at a major research university in the United States and accompanying teacher comments were examined. The course was taught by 10 instructors, teaching assistants pursuing an MA degree in teaching English as a second language at the same university.

The two researchers read and analyzed the first drafts, teacher feedback, and student revisions. They first identified the comments in which teachers addressed one or more features of coherence in a particular piece of student essay. To identify comments addressing issues with coherence, researchers used major categories to establish coherence from the ESL writing literature (Hinkel, 2004; Lee, 2002). Then, they examined the revised papers to see the effects of the teacher feedback on the first student revisions. The researchers marked each revision with reference to teacher comments on the first drafts and used a subjective revision rating scale (see Figure 1) to determine what sort of changes had been made based on the comments. The analysis assessed to what extent the changes made in response to the teacher's feedback led to negative, negligible, mixed, or positive effects (for a detailed discussion of the rationale and procedures of the analysis, see Ferris, 1997, and Ferris, Pezone, Tade, & Tinti, 1997). The changes were considered, taking into account the commented part of the first draft, the functional goals of the comment, and the revised piece in the second draft.

Figure 1. Rating scale for revisions (Ferris, 1997).

0 No discernible change made by student in response to this comment

1 Minimal attempt by student to address the comment, effect generally negative or negligible

2 Substantive change(s) made by student in response to comment, effect generally negative or negligible

3 Minimal attempt by student to address the comment, effect mixed

4 Substantive change(s) made by student in response to comment, effect mixed

5 Minimal attempt by student to address the comment, effect generally positive

6 Substantive change(s) made by student in response to comment, effect generally positive


The examination of first drafts and teacher feedback with reference to coherence categories documented in Hinkel (2004) and Lee (2002) revealed that out of 841 teacher comments on student drafts, 86 were found to be comments on coherence, which is 10.22% of the total written feedback. Next, student revisions based on these comments were examined to show the nature of changes (substantive, minimal, none) and their effects (negative/negligible, mixed, positive) following Ferris (1997).

A significant number of attempts at revision (62%) showed that the changes made by students to address problems with coherence were minimal, with varying effects (negative/negligible effects, 11%; mixed effects, 16%; positive effects, 35%). In 23% of cases no attempts were made to address teacher comments. Though students made substantive changes, altogether these changes accounted for 15% of all teacher feedback-based revisions with different effects (negative/negligible effects, 2%; mixed effects, 4%; positive effects, 9%).

The results show that written feedback on coherence triggered minimal changes by students, albeit with largely positive effects. This interesting pattern, however, was not necessarily due to students' tendency to be so economical in incorporating feedback into their writing but rather can be explained by the nature of the teachers' comments.

The striking finding was that most comments that resulted in minimal changes focused on transitional words and expressions (e.g., The use of "but" is a bit confusing here) and appeared rather specific and limiting. Comments that led to no changes and substantive changes, on the other hand, were imprecise (e.g., You need to work on making your writing more coherent) and broad (e.g., Your ideas in this essay are not well-organized). Synoptically, the analysis revealed that feedback on coherence tends to be either vague and general, or very specific and restrictive, with an excessive focus on textual cohesive devices (e.g., transitional words, such as however, therefore, what's more). This finding reflects problems with teachers' feedback on coherence, which may become a source of confusion for students and may fail to prompt them to write coherently.


ESL teachers in this study did allocate some attention to coherence in their written feedback on learners' essay drafts. However, as shown in this study, the feedback on coherence tends to be on either side of the specificity pole. On the basis of students' revisions, it is suggested that teachers carefully structure specific, text-based comments that relate back to classroom instruction, and provide students with practical and concrete means to go about achieving coherence in their writing. Clearly identifying problems can be useful, but for a better outcome it might be necessary to consider providing specific and concrete suggestions on how to improve the problems on the discourse level. This solution calls for shared discourse that teachers and students can exploit when pointing out problems and discussing concrete solutions to revise them. Teachers should explicitly address issues of coherence early on in classroom instruction, and attend to coherence in their feedback with references to classroom discussions of what coherence is and how coherence can be formed. Such practice might be effective in linking classroom instruction and written comments to maximize the efficacy of feedback on coherence.


Brown, G., & Yule, Y. (1983). Discourse analysis. Cambridge, England: Cambridge University Press.

Cerniglia, C. S., Medsker, K. L., & Connor, U. (1990). Improving coherence by using computer©\assisted instruction. In U. Connor & A. M. Johns (Eds.),Coherence in writing: Research and pedagogical perspectives (pp. 227©\241). Washington, DC: TESOL.

Ferris, D. (1997). The influence of teacher commentary on student revision. TESOL Quarterly, 31, 315-339.

Ferris, D. R., Pezone, S., Tade, C. R., & Tinti, S. (1997). Teacher commentary on student writing: Descriptions and implications. Journal of Second Language Writing, 6, 155-182.

Hinkel, E. (2004). Teaching academic ESL writing: Practical techniques in vocabulary and grammar. Mahwah, NJ: Lawrence Erlbaum.

Hoey, M. (1991). Patterns of lexis in text. Oxford, England: Oxford University Press.

Kintsch, W., & van Dijk, T. A. (1978). Toward a model of text comprehension and production. Psychological Review, 85, 363-394.

Lee, I. (2002). Teaching coherence to ESL students: A classroom inquiry. Journal of Second Language Writing, 11, 135-159.

Williams, J. D. (1985). Coherence and cognitive style. Written Communication, 2, 473-491.

A Theme/Rheme Approach to Writing Instruction: From Clause to Essay

Duane Leonard, University of California, Davis, duaneleonard@gmail

As an instructor of developing writers new to the expository essay, when I focus on “flow” or cohesion within an essay I try to link it to grammatical resources in a context-embedded, meaningful way. Though there are many different ways to discuss flow, I have had some measure of success using functional grammar. Schleppegrell (2007) described functional grammar as follows:

Functional linguistics approaches to grammar that highlight the meaning making role of language are demonstrating that a focus on the value and power of different language choices can offer insights about curriculum, pedagogy, and language development. Analyzing language from a functional perspective reveals how different language choices construct more or less powerful texts, helps us see what is valued in students’ writing, and helps us explore how language develops over time. (p. 122)

I present functional grammar to students using a theme/rheme division of some of their paragraphs. The idea of theme/rheme stems from systemic functional linguistics, where theme is “the jumping off point” of the clause and the rheme is “everything else” (Locke, 1996, p. 222). I think dividing up sentences is a more accessible entry into this approach than is dividing up each clause. Thus I redefine theme/rheme as “everything up to, and including, the focal noun” and the rheme as “everything else” (see Table 1) Though my version of a theme/rheme resembles a subject/predicate sentence division, a theme/rheme division is more flexible in that it allows for multiple “subjects.” In particular, this type of division allows the instructor to focus students on noun and adverbial clauses (examples B and C, respectively, in Table 1). These deconstructed sentences are to be read normally, from left to right (one will not read a rheme before completely reading the theme preceding it).

Table 1. Theme/Rheme, Manipulated




A theme/rheme division

can be illustrative to students.


Duane Leonard feels that a theme/rheme division

can be illustrative to students.


Although Duane Leonard feels that a theme/rheme division can be illustrative, it

is a concept that has to be clearly explained or students will be lost.

After the grammatical resources expected in expository writing and a theme/rheme division are introduced, it is important to review these features by deconstructing a model paragraph written by a student. Below is a prompt followed by a body paragraph taken from an essay written in response to the prompt.

Prompt: Write an essay in which you discuss factors in homes and/or school which are, in your opinion, most important for promoting effective learning.

Besides audible factors, appearance of studying area, including cleanliness, good organization, and great illumination, is also a factor that affects students to develop efficient learning skills. In general, a good appearance of area gives people an idea of broadness and comfort, and that makes people stay calm and easily stay focused. First, cleanliness affects students’ learning mood. If the desk is dirty, covered with dust, students do not want to sit down and put their books on it. Second, a well-organized place to study make students easily know the location of every object they need to use for studying. That makes students more efficient in learning. On the contrary, if there are a lot of objects on the desk, students are not able to find a place to study. Third, illumination of studying area creates atmosphere for studying. Good illumination lets students read without exhaustion; therefore, they are able to concentrate for a long period of time. On the other hand, if the place is dark, students will feel hard to study after a while, and that might make them stop studying. Hence, appearance of studying area indirectly affects their learning efficiency but still can make a great difference on students’ learning mood, so students should not ignore it.

Breaking up a successful paragraph into theme/rheme (see Table 2) makes the writing look different; the focus is automatically split between two parts and the contents therein. This division encourages students to look at a piece of writing more analytically and more easily focuses them on grammatical resources than does the normal view of a paragraph presented above.

Table 2


Besides audible factors,appearance of

studying area, includingcleanliness,

good organization, and greatillumination,

is also a factor that affects students to develop efficient learning skills.


In general, a good appearance of area

gives people an idea of broadness and comfort, and that makes people stay calm and easily stay focused.


First, cleanliness

affects students’ learning mood.


If the desk is dirty, covered with dust, [then] students

do not want to sit down and put their books on it.


Second, a well-organized place to study

make students easily know the location of every object they need to use for studying.


Third, illumination of studying area

creates atmosphere for studying.


Good illumination

lets students read without exhaustion;


therefore, they

are able to concentrate for a long period of time.


On the other hand, if the place isdark, students

will feel hard to study after a while, and



might make them stop studying.


Hence, appearance of studying area

indirectly affects their learning efficiency but still can make a great difference on students’ learning mood, so students should not ignore it.

The flow, or transitions, in this paragraph are excellent on three levels. First, the besides audible factors situates this paragraph within a larger essay. Second, the First, Second, and Third lead the reader along from example to example. Third, Hence alerts the reader to analysis of the paragraph. Transitions are overused, but even this can be pedagogically productive; this overuse provides food for discussion with students—a discussion focused on both flow and grammatical resources.

The focus in this paragraph is also excellent, and all the nouns that show this focus are abstract and appear at the beginning of the sentence. Notice how the student has a general noun—appearance—in the first theme that is then distilled as three more specific nouns: cleanliness, appearance, and illumination. The most important noun of the paragraph, appearance, reappears in the final, analytical sentence. Though there is much more to say about the successful employment of grammatical resources to make this paragraph a good model of the expository essay, this simple theme/rheme division is not overwhelming and demonstrates resources students can later apply to their own paragraphs.

In addition, students can see how this paragraph (as the paragraph that follows Topic Sentence 2) fits within the larger argumentative development of an essay (see Table 3). Here we can see that the abstract noun appearance is one of four abstract nouns (bolded below) that develop as an even more abstract noun(important factors) of the thesis statement

Because a good essay remains incomplete without a strong conclusion, Table 4 demonstrates how this author, after unpacking the focus noun of the thesis (important factors) in each of the body paragraphs (right-hand column), repacks those same nouns in a different fashion to create a very genre-appropriate conclusion. First he repacks quiet area and appearance of area as the study environment and parents’ education levels and parents’ encouragement as parents influence, condensing four ideas into two. These two ideas are then condensed even further back to factors, but, now factorshave the weight of the essay behind them and any comment on factors is a comment on the entire essay! A key thing to note about factors is that not only was it used in the thesis statement, but it is a key noun from the prompt itself—demonstrating not only focus within the essay, but a focus directly related to the question asked. This technique of unpacking and repacking key abstract nouns, especially when this process unfolds in the theme position of key sentences, not only creates very good flow through an essay but also allows students to make a final comment in their conclusion instead of simply repeating what they have said.

To conclude, such an approach to teaching writing allows students a chance to critique the choices used to create focused development and flow in expository writing. Seeing these models also allows students to apply what they learn through this exercise to future drafts or writing assignments. Whether the focus is coherence or something else, a functional approach to grammar, one that is embedded in what students write, allows students to see how peers have successfully used grammatical resources. This approach highlights how a wide variety of grammatical resources can be used to express similar ideas. Such an approach can also strengthen critical reading skills by building students’ ability to track arguments through a text in order to assess its strengths or weaknesses. In sum, this approach shows developing writers the structure and flexibility within genres of writing: structure to guide them and flexibility to maintain their creative voice in class and the writing they will do beyond them.


Locke, G. (1996). Functional English grammar: An introduction for language teachers. Cambridge, Engliand: Cambridge University Press.

Schleppegrell, M. (2007). At last: The meaning in grammar. Research in the Teaching of English, 42(1), 121-128.