Thonney Te a c h i n g t h e C o n v e n t i o n s o f A c a d e m i c D i s c o u r s e 347 > Teresa Thonney A study of scholarly research artic

Te a c h i n g t h e C o n v e n t i o n s o f A c a d e m i c D i s c o u r s e 347

> Teresa Thonney

A study of scholarly research articles from six disciplines provides insights about academic
writing that composition instructors can use to prepare students to write across

the curriculum.

Teaching the Conventions of
Academic Discourse

New Voice

Given the current emphasis on disciplinary discourses, it’s not surprising that so little recent attention has been devoted to identifying conventions that are
universal in academic discourse. In this essay, I argue that there are shared features
that unite academic writing, and that by introducing these features to first-year
students we provide them with knowledge they can apply and refine in each new
discipline they encounter.

Some scholars believe that making generalizations about academic writing
is impossible. Just as there is “no autonomous, generalizable skill called ball using
or ball handling that can be learned and then applied to all ball games,” David
Russell argues, there is no “autonomous, generalizable skill or set of skills called
‘writing’ that can be learned and then applied to all genres or activities” (57, 59).
Because there are no “general” skills that students can learn and transfer to all
writing situations, some suggest that students would benefit more from learning
about the ways writing conventions vary across academic disciplines and discourse
communities (Wardle 784).

Others (such as Berkenkotter and Huckin; Freedman) believe that writing
conventions can’t be taught and that trying to teach them “assumes that one can
learn to write academic genres by adhering to a definite rule-set” (Lynch-Biniek).
But linguistic scholars (including Swales; MacDonald; Bazerman; Biber) have dem-
onstrated that patterns and formulas prevail in academic writing, and many have
described the benefits of teaching writing conventions to students (see, for example,
Williams and Colomb). By teaching conventional ways to introduce topics, iden-
tify sources, and organize arguments, for instance, we provide “a valuable tool for
clarifying academic mysteries to large numbers of students” (Birkenstein and Graff).
In fact, Wilder and Wolfe found that students who were explicitly taught language
conventions in a literature course wrote better essays and reported comparable or
higher levels of enjoyment in the course than those receiving no instruction in
writing conventions (170).

e347-362-May11-TE.indd 347 4/23/11 10:19 AM

Text Box
Copyright © 2011 by the National Council of Teachers of English. All rights reserved.

348 T E T Y C M a y 2 0 1 1

As Hassel and Giordano noted in a recent TETYC article, the need for
explicit instruction in writing conventions is particularly acute at open-admission
two-year colleges, where many students, including those testing into college-level
writing courses, are unfamiliar with rhetorical strategies expected in college writing
(25). Even freshmen at universities, when asked to write college papers, can feel like
they are being asked “to build a house without any tools” (Sommers and Saltz 131).

Studies by Carroll, Herrington and Curtis, and McCarthy reveal considerable
variety in the writing undergraduates do and in the disciplinary approaches they
encounter. Disciplines differ in modes of inquiry, in forms of proof, and in meth-
ods of research. These differences manifest themselves in writing, as documented
in corpus-based studies by Swales, MacDonald, Hyland, and others, differences
students will appreciate when they learn to write the genres of their chosen majors.

Despite this variation, some principles appear in all academic writing guides,
no matter the discipline, as Karen Bennett found in her survey of forty-one style
manuals. Some shared features, such as source citation, are, of course, realized differ-
ently across disciplines; but Bennett found “remarkable consensus as regards general
principles, methods of textual construction, and the kinds of grammatical and lexical
features to be used” (43). No first-year student is expected to write like discipline
insiders when writing in entry-level courses that are “predisciplinary in both theory
and practice” (Diller and Oates 54). But research indicates students are rewarded
when they produce prose that resembles that of experienced academic writers.

To determine what rhetorical features appear in the prose of experienced
academic writers, I analyzed twenty-four research articles—four articles from each
of six disciplines: psychology, sports medicine, biology, marketing, literature, and
engineering. The articles were randomly selected from the following peer-reviewed

American Journal of Community Psychology

American Journal of Sports Medicine

Journal of Cell Biology

Journal of Marketing Research

PMLA (Publications of the Modern Language Association of America)

Proceedings of the Institution of Mechanical Engineers

My analysis reveals six standard “moves” in academic writing:

> Writers respond to what others have said about their topic.
> Writers state the value of their work and announce the plan for their papers.
> Writers acknowledge that others might disagree with the position they’ve

> Writers adopt a voice of authority.
> Writers use academic and discipline-specific vocabulary.
> Writers emphasize evidence, often in tables, graphs, and images.

e347-362-May11-TE.indd 348 4/23/11 10:19 AM

Te a c h i n g t h e C o n v e n t i o n s o f A c a d e m i c D i s c o u r s e 349

Introducing first-year composition students to these conventions of academic writ-
ing provides them with knowledge they can use now and refine later when writing
in their chosen disciplines.

Let’s start with the standard way academic writers begin—by summarizing
what others have said about their topic.

1. Academic Writers Respond to What Others Have Written about
Their Topic

When academics write, they join a conversation. To show they understand this they
refer to what others have already written about their subject. This feature appears
in every article of the sample. Consider this passage from a report in the sports
medicine sample articles:

In the past decades, major insights have been gained into how intrinsic factors
and extrinsic signals control and guide the development of dendrites and den-
dritic spines and how patterned neural activity shapes this process (Hering and
Sheng, 2001; . . . Van Aelst and Cline, 2004). Nonetheless, large gaps still exist in
our knowledge about how all these pathways integrate and execute their function
at the molecular level. (Huang, Zang, and Reichardt 527)

By referring to what others have said about a topic, writers accomplish two things:
they show that they are addressing an issue that matters, and they establish that
there is more to be said about it.

Sometimes writers enter the conversation by taking issue with the conclu-
sions of previous researchers, as in this passage from the literature articles:

[Christopher] Lane’s thesis, linking ambivalent national-symbolic identifications
on the part of homosexual writers to specifically colonial rhetorical struc tures, is
convincing (3); however, I would position Auden’s case dif ferently, as paradoxical
to this founding paradox of colonial passion. (Christie 1576)

Others have noted that disciplines vary in the way disagreement gets ex-
pressed. Linton, Madigan, and Johnson found that in literary criticism, for example,
attacks can get personal, unlike in other disciplines where disagreements are ignored
or limited to criticizing research methods (73–74). But the writers in my sample,
including those representing literature, show respect for previous research. Under-
graduates, given their junior status, would be wise to follow suit when disagreeing
with published scholars.

Like published scholars, undergraduates write research-based papers, today
more than ever (Lunsford and Lunsford 793). But they struggle in two notable
ways. First, many students fail to contribute to the conversation. Instead of analyz-
ing, synthesizing, or adding to what others have said, they merely show they have
“done the reading.” Second, in student papers, incorrect or missing source citations
abound. Tinberg and Nadeau’s recent study of first-year students at a community
college reminds us that for students the most in-depth discussion and practice of

e347-362-May11-TE.indd 349 4/23/11 10:19 AM

350 T E T Y C M a y 2 0 1 1

writing occurs in their required writing course (128). One way we can prepare
students for writing across the curriculum is with assignments that involve sum-
marizing, synthesizing, attributing writers, and commenting on what they have said.

2. Academic Writers State the Value of Their Work and Announce the
Plan for Their Papers

One reason academics refer to what has been written about an issue is to establish
that the issue matters. Another reason is to show that their research addresses an
aspect of the issue still unresolved. All twenty-four writers in our sample explain
that their research is necessary, unique, or otherwise of value, as in this passage from
the marketing articles:

The vast majority of research that has assessed the effect of price promotions on
brand evaluation has studied the effect after product trial, rather than pretrial. . . .
Unlike previous studies . . . , we examine the effects of price promotions pretrial
to isolate their informational impact on brand quality perceptions from the
potentially moderating effect of prior personal experience with the brand.
(Raghubir and Corfman 212)

Scholars must sell their work to editors and reviewers; but students too must “sell”
their work to their professors. By explaining why their topic is important, how
their approach to a topic is unique, or even why they chose to write about a topic,
students set their papers apart from papers that lack purpose.

In addition to stating the value of their work early, academic writers help
readers navigate their texts. All twenty-four titles in our sample announce the spe-
cific topic of the article; a few (particularly in the sciences) also convey the research
results. Here is an example from the psychology articles:

Conceptualizing and Measuring Historical Trauma among American Indian

From the biology articles:

Process Outgrowth in Oligodendrocytes is Mediated by CNP, a Novel Microtu-
bule Assembly Myelin Protein

Twenty-three of twenty-four articles also include subheadings that announce the
topic of sections:

Effects of Multiple Ankle Sprains on Postural Sway
Matters of Conscience in Machiavelli and Macbeth

Another way academic writers prepare readers for what is ahead is with an
explicit statement of purpose. Here is an example from the engineering articles:

This paper describes the development of a second generation of piezoelectric
paint and the characterization of sensors made with it. (Hale et al. 1)

In some articles, writers announce their hypothesis:

e347-362-May11-TE.indd 350 4/23/11 10:19 AM

Te a c h i n g t h e C o n v e n t i o n s o f A c a d e m i c D i s c o u r s e 351

We hypothesized that there would be an increase in ankle repositioning errors
and postural sway in basketball players who had sustained bilateral ankle sprains,
under conditions in which they had to rely more heavily on ankle proprioceptive
input. (Fu and Hui-Chan 1175)

In other articles, the statement of purpose expresses the writer’s opinion:

I . . . precede my discussion of the trope of the castrato with a brief histori cal
overview of the situation and reception of actual castrati singers. I then show how
Jo hann Jakob Wilhelm Heinse (1746–1803) and Friedrich Schiller (1759–1805)
used the figure of the castrato as a privileged metaphor for the negotiation of
class conflicts, gender con cepts, and the nature of art. (Krimmer 1544)

Many students think the main claim in an academic argument must be an assertive,
polemic statement. But corpus-based analyses reveal that most academic writers
state their main claim matter-of-factly (Conrad 119–20). Statements that begin
with “This paper describes,” “We hypothesized,” and “I then show” (from the above
examples) are not argumentative; they hardly seem like opinions.

Most writers in our sample identify the paper’s organization along with the
purpose. Here is an example from the psychology articles:

First, we will provide an overview of previous work conceptualizing historical
psychological distress among American Indians. Second, we will present a sum-
mary of qualitative data from elders on two American Indian reservations in the
upper Midwest that was used to develop a measure of historical trauma. Third, we
will describe measures of historical trauma and provide measurement characteris-
tics and frequencies on the basis of a sample of 143 parents. (Whitbeck et al. 120)

From the marketing articles:

The article is organized into four sections . . . that systematically investigate the
effect of package shape on volume perceptions, preference and choice, consump-
tion (perceived and actual), and postconsumption satisfaction. (Raghubir and
Krishna 314)

Some composition instructors want students to avoid statements of purpose
that begin “In this paper” and to avoid “blueprint” statements that announce topics.
But such statements are commonplace in academic journals, and many professors
reward students who make reading easy. In their analysis of 50 graded essays (from
various disciplines), Tedick and Mathison noticed “the general pattern was that
subjects received higher holistic scores on the essays—regardless of prompt type—
that they framed well enough for readers to be able to make predictions about the
content to come” (206).

In addition to providing subheadings and overviews, many writers in the
sample stop within their articles to announce what is next, as in this example from
the marketing articles:

In the next section, we discuss relevant research on visual mental imagery in the
design, marketing, and psychology literature, present a conceptual model of how

e347-362-May11-TE.indd 351 4/23/11 10:19 AM

352 T E T Y C M a y 2 0 1 1

visual mental imagery influences the customer appeal of the product designed,
and propose a set of hypotheses. (Dahl, Chattopadhyay, and Gorn 19)

Most writers end by summarizing what has been covered and reiterating the value
of their research, as shown in this example from sports medicine:

To our knowledge, this study provides the longest follow-up in the literature
of patients undergoing meniscal repair with the arrow. . . . Indeed, this study
represents the longest follow-up in the literature on any of the available all-inside
meniscal repair devices. (Lee and Diduch 1140–41)

Every article in the sample includes a statement of purpose, preview sen-
tences, review sentences, and sentences that announce the value of the research.
Student research and writing may not be as complicated as that of the scholars in
our sample, but students write for professors who read many papers—quickly. A
wealth of research has shown that when writers signal where they are going and
how they will get there, readers read faster and remember better what they have
read (Meyer 212–16). This is an important principle for students to learn.

3. Academic Writers Acknowledge That Others Might Disagree with the
Position They’ve Taken

Because scholars recognize that others might disagree with their conclusions, they
sprinkle their writing with qualifiers, or hedges, such as “probably,” “possibly,”
“maybe,” and “it seems,” particularly when writing to colleagues. Writers use hedges
to make statements more accurate and to avoid appearing dogmatic. Examples of
hedges are italicized in the following sentences from our sample. First from the
sports medicine articles:

The onset latency to the ADM was not affected, whereas the onset latency to the
FDI was affected, suggesting, the lesion may be located in the palm, distal to the
motor branch to the ADM. (Akuthota et al. 1228)

From the psychology articles:

[Oppressed people] tend to be passive and unable to recognize their own capacity
to transform their social reality; and their existence is often accepted on the basis
of destiny, bad luck or supernatural will. (Balcazar, Garate-Serafini, and Keys 250)

Writers in the sample also anticipate potential critics by recognizing the limitations
of their findings:

More research, varying the factors previously identified, is necessary to establish
the generalizability of our findings to a broader range of product design contexts.
(Dahl, Chattopadhyay, and Gorn 27)

Professors sometimes complain that students fail to back their claims with
sufficient evidence. While this is sometimes true, the problem can be partly due to
students’ failure to qualify assertions. Some students, especially those who are not
native speakers of English, underuse qualifiers (e.g., “apparently,” “likely,” “possibly”)

e347-362-May11-TE.indd 352 4/23/11 10:19 AM

Te a c h i n g t h e C o n v e n t i o n s o f A c a d e m i c D i s c o u r s e 353

and overuse words expressing certainty (e.g., “really,” “of course,” “certainly”) in
their writing (Gilquin and Paquot 47).

By teaching students how to distinguish between statements of “fact” and
opinion, how to differentiate between generalities and specifics, and how and when
to moderate claims with hedges, we help them write better arguments in any dis-
cipline. Students readily see the difference between “Surveys prove Americans are
changing their attitudes about same-sex marriages” and “Surveys suggest Americans
may be changing their attitudes about same-sex marriage,” and with practice they
learn to moderate sweeping generalities.

4. Academic Writers Adopt a Voice of Authority

Although tentative in their claims, academic writers still write with authority.
Conveying authority is understandably challenging for student writers. David
Bartholomae describes their dilemma: Students “have to speak in the voice and
through the codes of those of us with power and wisdom; and they not only have
to do this, they have to do it before they know what they are doing . . . and be-
fore, at least in the terms of our disciplines, they have anything to say” (156). Even
graduate students have difficulty establishing an ethos of authority when writing as
initiates in their field (Blakeslee 133). But students can learn to imitate techniques
of experienced writers.

Using First or Third Person

Writers in a few disciplines, such as engineering, tend to avoid first person in for-
mal writing. A look at two passages from our sample, the first from marketing, the
second from engineering, is revealing:

In this article, we examine the effect of elongation on (1) perceived volume, (2)
perceived consumption, (3) actual consumption, (4) postconsumption satisfaction,
and (5) choice. As described in Figure 1, our model suggests that package shape
directly affects perceived volume and through this, indirectly and inversely affects
perceived consumption. (Raghubir and Krishna 323)

This paper presents a new approach to model the friction layer in brake systems
in the investigation of noise and vibration, especially high-frequency squeal. . . .
The friction layer is modeled as a coupling stiffness between the brake pad and
the rotor as a combination of the elastic stiffness of the friction layer superim-
posed on the coupling modal stiffness of the brake-pad combination. . . . By
incorporating the earlier results in a two degree of freedom model, the predicted
frequencies were shown to be close to the squeal frequencies obtained from field
tests. (Paliwal et al. 520–21)

The engineering paragraph includes no mention of who completed the research
(“predicted frequencies were shown”). In fact, the paragraph is from a journal
that advises authors: “Papers should be written in the third person in an objective,
formal and impersonal style.”

e347-362-May11-TE.indd 353 4/23/11 10:19 AM

354 T E T Y C M a y 2 0 1 1

But in the rest of the sample, nineteen of twenty writers use first person.
Writers in medicine, marketing, psychology, biology, and literature all make clear
that they formed hypotheses, collected data, and reached conclusions. From the
sports medicine articles:

We compared the results obtained from the injured ankle with those from the
uninjured ankle. (Santilli et al. 1186)

From the psychology articles:

My colleagues and I interviewed 28 adult Bosnians attending a community
mental health program. (Miller 225)

Compared to the engineers, these writers also use more active voice con-
structions—another way to convey authority. For engineers, the average number of
occurrences of passive voice within 500-word excerpts is nearly twice the average
for any other discipline in the sample (15.8 occurrences in engineering versus 8.8
occurrences in sports medicine, 4.3 in psychology, 6.0 in marketing, 7.0 in cell
biology, and 3.25 in literature).

The challenge for student writers is knowing how and when to use first
person. Many students needlessly preface statements with “It seems to me” or “I
think” (Gilquin and Paquot 48–49, 55–57). Others, attempting to convey authority,
adopt the voice of moralizing parent. With direction, however, students improve.
They can learn to judge when writing “I think” has purpose and when writing
“I think” is pointless. (McKinney Maddalena provides excellent help for students
concerning when to use first person.)

Writing Concisely

Another way writers create an ethos of authority is by using a high percentage of
meaning-carrying words. In the 1970s, Jean Ure developed a method for deter-
mining a text’s lexical density by calculating the percentage of lexical words (445).
Lexical words include nouns, verbs, adjectives, and adverbs—classes of words that
convey meaning and to which we continue to add. Grammatical words include
pronouns, auxiliary verbs, conjunctions, prepositions, articles, and other determin-
ers—classes of words to which we don’t add. Thus, the following sentence includes
seven lexical words (in bold print):

Some scientists believe that stem cells can be used to treat diseases.

While spoken language includes many grammatical words (Ure found the per-
centage of lexical words in spoken language to be below 40 percent), written texts
tend to be more lexically dense. In Ure’s study (in 1971), the lexical density of a
textbook was 50.2 percent, and the lexical density of a scholarly journal was 52.8
percent (cited in Ventola 159). The lexical density in our sample ranges between
52.8 percent (in sports medicine) and 56.5 percent (in cell biology). In other words,
more words than not are meaning-carrying words.

Writers in our sample pack meaning into sentences:

e347-362-May11-TE.indd 354 4/23/11 10:19 AM

Te a c h i n g t h e C o n v e n t i o n s o f A c a d e m i c D i s c o u r s e 355

They don’t describe “meniscal healing that was incomplete” but instead write
“incomplete meniscal healing.”

Not “sociologists and geographers who are feminists” but “feminist sociolo-
gists and geographers.”

Not “an outdoor site that is exposed” but “exposed outdoor site.”

The average lexical density rate of the sample is 54.4 percent, higher than that of
most types of writing. When we teach students how to revise for conciseness, we
teach them a sure-fire way to improve the quality and authority of their academic

5. Academic Writers Use Academic and Discipline-Specific Vocabulary

One obvious marker of academic writing is academic vocabulary. Several studies
of academic writing have focused on familiar sequences of three or more words
referred to as “lexical bundles.” They include phrases such as the following:

in order to
the presence of
the fact that
in the case of
as a result of

Lexical bundles like these account for 20 percent of the words in academic prose
(Biber et al. 995), and using these phrases is one indicator of proficiency in aca-
demic writing. But Viviana Cortes found that students rarely use them in their
writing, and when they do use them it is often not in the way published writers
do. She concludes that students would benefit from explicit instruction in lexical
bundles and their functions (420–21). For example, when an assignment involves
summarizing data from studies, an instructor could show students lexical bundles
commonly used to introduce previous research (such as “studies have shown that”
and “have been shown to”) (Conrad 134). Additional ideas for teaching academic
lexical bundles are found in Graff and Birkenstein’s book They Say / I Say.

Another marker of academic writing is specialized language. Scientists have
long been known for co-opting words and using them in new, specialized ways, as
seen in these phrases from our sports medicine and biology articles:

prolongation of the median motor latency
preactivation of the lower extremity muscles
genomic integrity

But this tendency is not unique to scientists—as additional examples from the
sample illustrate. From the engineering articles:

limits of linearity of piezoelectric paint

From the psychology articles:

estimates of construct loadings

e347-362-May11-TE.indd 355 4/23/11 10:19 AM

356 T E T Y C M a y 2 0 1 1

From the marketing articles:

expectancy disconfirmation

From the literature articles:

textual and libidinal potentials of coloniality

Technical words like these precisely and concisely convey specialized meanings to
others in the field and denote one’s membership in any academic community. In
fact, Robyn Woodward-Kron has demonstrated that “adopting the specialist language
of the discipline is intrinsic to learning disciplinary knowledge” (246).

One way to make students aware of specialist language and lexical bundles
is to have them look for recurring terms, stylistic conventions, and other patterns
in a corpus of academic writing. There are many free resources for corpus-based
research, including the Corpus of Contemporary American English (COCA), with
concordancer, available at; and the Michigan Cor-
pus of Upper-Level Student Papers (MICUSP) at
(Information about additional corpus research and analysis resources is provided by
David Lee at Students can use text analysis tools to study
the writing of a specific discipline, to learn how the writing styles of disciplines or
genres vary, or to analyze their own writing. Corpus-based research assignments
also provide students with opportunities to conduct primary research. (See Bowker
and Pearson for assignment ideas.)

6. Academic Writers Emphasize Evidence, Often in Tables, Graphs, and

Academic writing is ultimately judged on the basis of its evidence, and academic
writers use various techniques for highlighting data.

Fourteen (58 percent) of the authors in our sample include tables, graphs,
or charts. Given the prominence of data in academic writing, it is important that
students learn how to “read” quantitative data. Yet, as Joanna Wolfe recently argued,
most first-year students do not understand that writers manipulate “statistical ex-
pressions in order to make an interesting story out of their data” (459). She calls on
composition instructors to discuss quantitative arguments in their courses:

Our students should be able to quickly discern that the statements “there is a
one-in-fifty chance that a bad event will happen” and “there is a 98 percent
chance that everything will be okay” differ only in rhetorical choice between
two mathematically equivalent figures. And students should have practice making
their own arguments from quantitative data, not only so they can see the many
ways in which such claims can be manipulated, but also so they can see the role
that invention plays in statistical data, experimental results, and other quantita-
tive arguments that are often popularly perceived as nonrhetorical “facts.” (455,
original emphasis)

To illustrate the rhetorical nature of graphs, Wolfe provides four graphical
representations of raw data, each lending itself to a different interpretation of the

e347-362-May11-TE.indd 356 4/23/11 10:19 AM

Te a c h i n g t h e C o n v e n t i o n s o f A c a d e m i c D i s c o u r s e 357

data (463–64). With examples like those Wolfe offers, we can show students that
quantitative data are as much “language” issues as they are …

Looking for this or a Similar Assignment? Click below to Place your Order