ISSN: 1756–5839 (print)

ISSN: 1756–5847 (online)

writing & pedagogy
Research Matters

Writing from Sources, Writing from

Sentences

Rebecca Moore Howard, Tricia Serviss, and Tanya K.

Rodrigue

Abstract
Instead of focusing on students’ citation of sources, educators should attend to the more
fundamental question of how well students understand their sources and whether they
are able to write about them without appropriating language from the source. Of the
18 student research texts we studied, none included summary of a source, raising ques-
tions about the students’ critical reading practices. Instead of summary, which is highly
valued in academic writing and is promoted in composition textbooks, the students
paraphrased, copied from, or patchwrote from individual sentences in their sources.
Writing from individual sentences places writers in constant jeopardy of working too
closely with the language of the source and thus inadvertently plagiarizing; and it also
does not compel the writer to understand the source.
Keywords: plagiarism, student research, composition instruction, writing
from sources, summary, paraphrase, quotation, copying,
patchwriting

Affiliations

Rebecca Moore Howard: Syracuse University, USA.

mail: [email protected]
Tanya K. Rodrigue: Wheaton College, Massachusetts, USA.
email: [email protected]
Tricia Serviss: Auburn University, USA.
email: [email protected]

WAP vol 2.2 2010 177–192 doi : 10.1558/wap.v2i2.177

©2010, equinox publishing
LONDON
178 writing & pedagogy

Introduction
Writing from sources is a staple of academic inquiry. It plays a key role in
publications in every scholarly discipline, from the literary criticism of English
studies to the literature review in scientific publications. It plays a key role as
well in the assignments given to both graduate and undergraduate students.
The research synthesis helps graduate students survey and participate in the
conversations of their discipline, and the term paper, despite criticisms, persists
as a common undergraduate genre. Hence writing from sources looms large in
composition curricula, in introductory writing courses devoted to researched
writing, critical reading, analysis, and argument.
At the same time, many educators worry that students are accomplishing
their writing from sources by illicit means. It has become commonplace for
students to be described as would-be plagiarists, with unacknowledged copying
as their primary strategy of writing from sources. Indeed, contemporary culture
– including media discourse and academic discussions – asserts that we are in
the midst of a “plagiarism epidemic.” As David Callahan tours the college lec-
ture circuit talking about what his book calls our “cheating culture” (Callahan,
2004) and as headlines announce an “‘Epidemic’ of Student Cheating” (BBC
News, 2004), the academy and indeed culture itself seem collectively poised at
a precipice over which we will surely slip.
The Center for Academic Integrity (http://www.academicintegrity.org/) con-
ducts surveys asking students about whether they have engaged in a variety
of “cheating behaviors,” including unacknowledged copying from sources.
However, little research has inquired into the range of students’ techniques for
writing from sources. Do they represent their source through copying (whether
cited or uncited), summary, paraphrase, or patchwriting – “[c]opying from a
source text and then deleting some words, altering grammatical structures, or
plugging in one-for-one synonym-substitutes” (Howard, 1993: 233)?
Although no descriptive research has studied the whole range of students’
techniques for writing from sources, some valuable research on summary-
writing has been conducted. Brown and Day (1983) report on six “rules” that
writers follow when summarizing: two involve deletion of material from the
source text; two involve generalizing from specifics in the source text; and two
require invention of sentences that capture the gist of one or more paragraphs.
Students from elementary school through graduate school use deletion tech-
niques. The more advanced a student’s education, the more he or she is likely
to apply the generalization rule for summarizing. Not until tenth grade do
students employ invention as a means of summarizing, and only in graduate
school do they do so in all appropriate cases (Brown and Day, 1983).
 Writing from Sources, Writing from Sentences 179

Three years after Brown and Day’s experimental study, Sherrard (1986) asked
ten paid undergraduates to alternately summarize or recall seven texts which
were ordered randomly. She discovered that their most common method of
summarizing is not to combine multiple sentences from the source but to
paraphrase a single key sentence.
Brown and Day, and also Sherrard, were working prior to the Internet era
and thus prior to the time when educators were consumed with concerns
about plagiarism. The more recent research in the Internet era is predictably
contextualized by these concerns. Much of the recent research on summary
explores the relationship between summary, plagiarism, and patchwriting.
Whereas many institutions’ academic integrity policies classify patchwriting as
a form of plagiarism – a moral failure – recent research indicates that it occurs
as an intermediate stage between copying and summarizing: inexpert critical
readers patchwrite when they attempt to paraphrase or summarize. Roig (2001)
finds that 22% of psychology professors patchwrite when presented with the
task of summarizing complex text from an unfamiliar field. Howard (1993:
233) posits patchwriting as a learning strategy rather than an act of academic
dishonesty. Pecorari (2003) provides empirical verification of this hypothesis in
her discovery that non-native speakers of English (L2 writers) patchwrite, even
when writing doctoral dissertations. Shi (2004) reports that the Chinese college
students in her study copied longer sequences of words when summarizing than
did their native-English-speaking (L1 writers) counterparts.
What is now needed, we believe, is a great deal more information about
what students are actually doing with the sources they cite. What source uses
are being marked by citations? Are students copying from, patchwriting from,
paraphrasing, or summarizing the texts they cite? Are they accurately repre-
senting what is in the source? Are they fully citing their sources each time they
use them? These questions have been addressed in the field of applied linguistics
by Pecorari (2003, 2006, 2008), who studied the writing of L2 graduate students
in U.K. universities. But the scholarship of composition and rhetoric, the disci-
pline in which we work, has been largely silent on these issues. Little is known,
then, about how either L1 and L2 college students use the sources they cite. Yet
only when we have such information will writing instructors be able to craft
good pedagogy for students’ writing from sources.

Our Inquiry
We began our inquiry with an exploratory hypothesis: that college students,
both L1 and L2 writers, patchwrite. Our research was an intensive exploration
of a small sample of college students’ researched writing, to discover how
180 writing & pedagogy

many of the papers drew on which of the four source-use techniques: copying,
patchwriting, paraphrasing, and summarizing.
We refer to our work as an inquiry rather than a formal research study.
We began our work in the belief that large-scale, quantified data collected in
naturalistic rather than controlled environments is needed to answer ques-
tions about students’ uses of their cited sources. Our inquiry was intended
as a means of identifying what questions should be asked and what methods
should be used to answer them. Since ours was a preliminary inquiry, we did
not quantify our results but instead worked collaboratively to decide on the
issues that should be investigated in a formal study for the future that we are
now designing.
Having secured IRB clearance and course instructors’ permission, in Spring
2007 we visited 15 sections of a required sophomore research writing class
at what the Carnegie Foundation classifies as a large, private, not-for-profit,
comprehensive doctoral university. We asked students to allow us to study the
researched writing they did in the course. At the end of the term, we collected
final researched papers from the instructors; removed students’ names from
the papers; established separate piles for each section; randomized papers
within each pile; and began working our way down each pile until we found a
paper whose sources we could retrieve. Students’ uneven success with source
documentation made this an often-challenging task, and sometimes the sources
cited were not available online or at our libraries. Once we found a paper
whose sources we could retrieve, we included it as a paper for our research.
Because we worked with full anonymity for the participating students, we did
not control for demographic factors such as race, gender, and home language.
At the university where we collected data, 10% of students are international
students and 29% are from what the university calls “underrepresented groups.”
The university does not collect data on students’ home languages.
We chose 18 papers for two reasons: first, we designed our research after
studying that of Pecorari (2003, 2006), who chose a similar number for her
study. Second, the same constraints faced us as did Pecorari: our methods are
labor-intensive. Reading not only student papers but also the sources they cite,
and then coding each source use in each student paper, is time-consuming,
involving 3–5 hours’ work per paper. Moreover, in our research, each student
paper was coded by two researchers. We therefore decided on a relatively small
sample size, 18 papers, before our research began.
Once we had found sources for 18 papers, we read the sources and the papers.
Our questions were simple:
 Writing from Sources, Writing from Sentences 181

Does the paper contain one or more incidences of patchwriting?

Does the paper contain one or more incidences of paraphrase?
Does the paper contain one or more incidences of summary?
Does the paper contain one or more incidences of direct copying from
sources?
Does the paper contain one or more incidences in which direct copying is
not marked as quotation?
For this research we defined summary as restating and compressing the main
points of a paragraph or more of text in fresh language and reducing the sum-
marized passage by at least 50%. The 266-word Gettysburg Address (Lincoln
1863), 1 for example, might be summarized (by Lincoln or another person of
his time) this way: “The civil war that we are now fighting tests the principles on
which our country was founded. We must pursue this war as a way of honoring
the men who fought and died on this battlefield.”
We defined paraphrasing as restating a passage from a source in fresh lan-
guage, though sometimes with keywords retained from that passage. Paraphrase
does not involve a significant reduction in the length of the passage. The first
sentence of the Gettysburg Address, for example, might be paraphrased this
way: “The United States was founded in 1776 on the principles of liberty and
equality.”
Following Howard (1993), we define patchwriting as reproducing source
language with some words deleted or added, some grammatical structures
altered, or some synonyms used. The first sentence of the Gettysburg Address,
for example, might be patchwritten this way: “Eighty-seven years ago, the
founding fathers created a new nation that was conceived in the principle of
liberty and was dedicated to the equality of man.” If quotation marks are used
for the copied bits, the text is marked as quotation, not patchwriting. However,
a passage may be patchwritten even when it is properly quoted and referenced.
By copying we mean the exact transcription (though perhaps with occasional
minor errors) of source text. As we categorized passages of student text into the
four of types of source use, whether the passage was referenced did not affect
its category. Copying, then, can include both quotation and unacknowledged
copying. Regardless of whether quotation marks and referenced citation were
present or absent, exact copying was classified as copying.
In searching for these four methods of source use (summary, paraphrase,
patchwriting, and copying), we were also searching for indications of source
182 writing & pedagogy

comprehension – or difficulties with source comprehension. Scholarly and

textbook literature asserts that patchwriting is a sign of uncertain comprehen-
sion of the source (Angélil-Carter, 2000; Roessig, 2007; Roig, 2001) and that
summary is a sign of source comprehension (Angélil-Carter, 2000; Brown
and Day, 1983; Harris, 2006). Copying and paraphrasing are not necessarily
a sign of either. Copying does not require comprehension of what one copies,
regardless of whether the copying is marked as quotation and cited. Paraphrase
does require comprehension, but usually only of a sentence or two.

Findings
From the 18 papers we read, we derived the following answers:
1. Does the paper contain one or more incidences of patchwriting?
- In 16 of the 18 papers (89%), the answer is “yes.”
2. Does the paper contain one or more incidences of paraphrase?
- In all 18 papers (100%), the answer is “yes.”
3. Does the paper contain one or more incidences of summary?
- In all 18 papers (100%), the answer is “no.”
4. Does the paper contain one or more incidences of direct copying from
sources?
- In 14 of the 18 papers (78%), the answer is “yes.”
5. Does the paper contain one or more incidences in which direct copying
is not marked as quotation?
- In 13 of the 18 papers (72%), the answer is “yes.”
In addition, as we read, we made two further discoveries:
6. Of the 18 papers, 17 (94%) contained non-common-knowledge informa-
tion for which no source was cited.
7. Of the 18 papers, 14 (78%) attributed information to a source that either
did not contain that information or said something different from what
the student was attributing to it.
Despite the widespread pedagogical belief that summary is important to
source-based writing, our reading of 18 undergraduate research essays, along
with the sources those essays cite, uncovered not a single incidence of sum-
mary. We found copying, paraphrasing, and patchwriting – but no summary.
A paragraph from one student paper, 8.10, 2 compactly illustrates the sorts of
 Writing from Sources, Writing from Sentences 183

writing from sources that we encountered. Before we read the sources it cites,
this paragraph looked like a good research synthesis:
Studies show that children, as well as parents, in low-income families
have very few assets, so eliminating asset tests for coverage could increase
enrollment (Cox, Ray, and Lawler). Also, states could use ‘presumability
eligibility for pregnant women and children’ covered under Medicaid or
SCHIP. Through this, children or pregnant women who seem eligible for
the programs can be immediately enrolled until a final determination of
eligibility can be produced. To determine who ‘seems’ eligible for health
care coverage, school staff could be trained to judge who should be
enrolled. Studies show that children with health insurance have fewer sick
days from school, so this could ‘yield educational benefits’ (Broaddus).
With the increasing diversity and immigration status of our society,
Medicaid and SCHIP should also provide information on eligibility and
enrollment in many different languages, and in both documentation or
letters and personal visits. In every state, many lose coverage by Medicaid
and SCHIP when it is time to renew. In order to change this trend, the
programs should change their period to a 12-month plan, rather than the
6-month plan now. Also, to eliminate confusion and difficulty for a family,
states with separate Medicaid and SCHIP programs should coincide their
renewal times and conduct renewal by mail or telephone. States with call
centers, and reminder letters for renewal should increase recertification of
coverage. Lastly, they should consider enforcing a grace period of about
one to three months for renewal (Cox, Ray, and Lawler). Some states
have finally begun to take an initiative on solving these low enrollment
problems.
The paragraph appears to handle sources well, using quotation marks, providing
in-text citations to acknowledge sources, and citing two different sources, one
of them (Cox, Ray, and Lawler, 2004) in two different parts of the paragraph.
(Both sources are Web sites, so the absence of page references is not an issue.)
A reading of its sources, however, reveals that the paragraph is extensively
patchwritten. Figure 1 below places a succession of sentences in the apparently
well-cited paragraph above side-by-side with the corresponding sentences in
the sources. (We should note that in almost every case in all 18 papers, we
were easily able to locate the exact sentence from which the student writers
were working.) Underlining indicates where the paper is using the exact or
near-exact phrasing of its source. For the sake of brevity, Figure 1 illustrates
just the first few sentences, though the remainder of the paragraph continues
in the same vein, with only one sentence that does not contain copying or
patchwriting.
184 writing & pedagogy

Paper 8.10 Cox, Ray, and Lawler (2004) Broaddus and Ku (2000)

Studies show that children, as Studies have shown that

well as parents, in low-income most low-income families
families have very few assets, have few assets. Eliminating
so eliminating asset tests asset tests…
for coverage could increase
enrollment (Cox, Ray, and
Lawler).

Also, states could use (quotation not in the cited

‘presumability eligibility for source, though the keyword
pregnant women and children’ ‘presumptive [not presumability]
covered under Medicaid or SCHIP. eligibility’ is)
Through this, children or This temporarily enrolls children
pregnant women who seem and pregnant women in SCHIP
eligible for the programs can and Medicaid as soon as they
be immediately enrolled until a apply for benefits, pending a
final determination of eligibility final eligibility determination.
can be produced.
To determine who ‘seems’ School staff could be trained in
eligible for health care how to conduct presumptive
coverage, school staff could be eligibility determinations and
trained to judge who should be how to carry out the necessary
enrolled. follow-up activities.
Studies show that children In addition to helping school
with health insurance have children gain better access to
fewer sick days from school, so health care and prevention
this could ‘yield educational services, presumptive eligibility
benefits’ (Broaddus). may yield educational benefits;
recent research suggests that
children who are insured have
fewer sick days and miss school
less often than children who
lack health insurance.
With the increasing diversity Write Letters reminding
and immigration status of our families to renew SCHIP. Go
society, Medicaid and SCHIP door-to-door to help families
should also provide information in the renewal process. .
on eligibility and enrollment . . Give families materials
in many different languages, about renewal in multiple
and in both documentation or languages.
letters and personal visits.

In every state, many lose In virtually all states, many

coverage by Medicaid and people lose Medicaid
SCHIP when it is time to renew. and SCHIP when it is time
to renew or recertify for
benefits.

Figure 1: Sentence-by-Sentence Comparison of a Paragraph from Paper 8.10 with

its Sources (Instances of exact copying, whether cited or uncited, and patchwriting
are underlined)
 Writing from Sources, Writing from Sentences 185

All eighteen of the student writers whose papers we analyzed engaged in the
sorts of textual strategies illustrated in Figure 1. A passage from student paper
A.1, interspersed with our comments in italics, demonstrates the fine level of
myriad difficulties that the writer faces in producing this text:

Medical students must possess knowledge. Aspiring to be a doctor is not

an easy career endeavor, in fact, it requires a lot of work and dedication.
According to Weinberger, the object of knowing requires more than being
right. [This misrepresents Weinberger’s definition of knowledge.] It is
necessary for doctors to be right because they do have a lot at hand,
which can be a patient‟s life and even just his or her immediate health.
[This incorrectly applies Weinberger to the paper’s investigation of a
medical students’ blog; it attempts to�������������������������������������
������������������������������������������������] Knowledge is „justified
true belief.‟ [����������������������������������������������������������
��������������������������������������������������������������������
�����������������������������������������]�So what does this have to do
with medical students and the web? From the medical students‟ side,
medical students are justified in believing that they will become doctors
because they have the capability to do so, but on the web side according
to Weinberger the web is „a hodgepodge of ideas that violates every rule
of epistemological etiquette.‟ [���������������������������������������������;�
�����������������������������.] Ideas that are posted on the web are
wrapped in individual voices that make it harder to dig out exactly what is
being said (Weinberger 139). [������������������������������������������
�������������������������������������������������������������.] The idea of
individual voices allows people to express themselves whether it is factual
or opinionated. This is exactly what blogging is focused on and what it is
basically about. In reading Anna‟s blogs, we only „hear‟ about her life,
feelings, and thoughts and not anyone else‟s. [������������������
��������������������������������������������������������������������������
apply Weinberger’s theory of the web as a social network and site of
dialogue to the medical students’ blog.]�

Figure 2: Paragraph from Paper A.1, with Our Comments in Brackets and Italics
(Our comments are based on reading not only the student’s paragraph but also its
source)

The source in question is David Weinberger’s Small Pieces Loosely Joined

(Weinberger, 2002), a 240-page complex theoretical text. Paper A.1 cites two
of those 240 pages, and its uneven representation of the Weinberger text sug-
gests the possibility that these may be the only two pages that the student read.
The paper endeavors to deploy Weinberger’s theory of knowledge, which may
186 writing & pedagogy

have been accessed by consulting the Weinberger index rather than by actually
reading Small Pieces Loosely Joined. This is our primary concern throughout
our analysis of these 18 papers: they cite sentences rather than sources, and
one must then ask not only whether the writers understood the source itself
but also whether they even read it. As teachers – and as writers ourselves – we
are not unfamiliar with the quote-mining approach to complex texts: the
search for a “good sentence to quote” – or to paraphrase or patchwrite – and
perhaps to cite. The absence of summary in these papers does not necessarily
mean that the student writers did not read the whole text being cited, nor
does it mean that they did not understand what they were reading. But the
absence of summary, coupled with the exclusive engagement of text on the
sentence level, means that readers have no assurance that the students did
read and understand.
When the source treats a technical topic or when it lists concrete items, the
writer working exclusively on the sentence level predictably struggles to write
from those sentences. Here, for example, are two passages from paper 3.6,
side-by-side with their sources:

Paper 3.6 Bainbridge (2007) source

After the materials are separated they are The obstacles of recycling plastic can be
melted down and mixed together. Then they overcome by using an elaborate monomer
undergo a complicated inverse polymer recycling process wherein the polymer
reaction from the one used to make it, undergoes an inverse polymer reaction
resulting in a mixture of chemicals which are of what was used to manufacture it. The
then synthesized to form a new polymer of end product of this procedure is a mix of
the same kind (Bainbridge). chemicals that form the original polymer,
which is further purified and synthesized to
form a new polymer of the same type.

Paper 3.6 West (2007) source

Plastic labeled number two is a high density Number 2 is reserved for high-density
polyethylene plastic, also known as HDPE. polyethylene plastics. These include heavier
These plastics are most commonly found in containers that hold laundry detergents and
containers holding heavier liquids, such as bleaches as well as milk, shampoo and motor
milk cartons, shampoo bottles, and laundry oil. Plastic labeled with the number 2 is often
detergents. The plastic is a much softer recycled into toys, piping, plastic lumber and
texture and is much more flexible that PETE. rope. Like plastic designated number 1, it is
HDPE is also very commonly, and fairly easily widely accepted at recycling centers.
recycled but can only be recycled once. HDPE
is often recycled into toys, plastic lumber, and
piping (West).

Figure 3: Comparison of Two Passages from Paper 3.6 with Their Sources (Exact
copying, whether cited or uncited, and patchwriting are underlined)
 Writing from Sources, Writing from Sentences 187

We have chosen these three papers – 8.10, A.1, and 3.6 – not because they are
extreme incidents but because they are concise illustrations of the struggles
that were in evidence in all eighteen papers. Similar struggles are documented
in prior research, especially in applied linguists’ studies of second-language
writers’ work with English-language source texts (Keck, 2006; Pecorari, 2003,
2006, 2008; Shi, 2004).

Discussion
We offer these side-by-side comparisons not to suggest that the writers are mis-
using sources (though sometimes that is indeed the case) but to demonstrate
that these students are not writing from sources; they are writing from sentences
selected from sources. That leaves the reader with the unanswered question:
does this writer understand what s/he has read? And it leaves the writer in a
position of peril: working exclusively on the sentence level, he or she is perforce
always in danger of plagiarizing. When one has only the option of copying or
paraphrasing, one can easily paraphrase too lightly, producing a patchwritten
sentence too close to the language of the original. This is a particular peril for
inexpert writers: From his review of scholarship in citation analysis, White
(2004: 105) concludes that, in general, it is only advanced writers who write
from sources without using any language from the source. Howard (1993)
argues that patchwriting should be considered a transitional stage in writing
from sources, rather than plagiarism, and the Council of Writing Program
Administrators (2005) labels patchwriting a misuse of sources rather than
plagiarism.
Still, many institutional codes of academic integrity – and indeed, many
writing handbooks and textbooks – persist in treating patchwriting as a form
of plagiarism. To complicate the matter, as Sandra Jamieson demonstrates,
the extent to which patchwriting counts as plagiarism can vary according to
academic discipline (Jamieson, 2008). Moreover, when one has only the option
of copying or paraphrasing, the copying may become so extensive that the
writer feels the need to withhold complete citation, for fear of appearing too
dependent on the source language. Or the writer may simply not know how
often to cite persistent use of source language.
Our inquiry does not answer the question of why none of these 18 students
summarized their sources, nor why so many of them patchwrote, misinter-
preted what a source said, or offered non-common-knowledge information
without citing a source. Perhaps they did not understand the sources. Perhaps
they didn’t care enough about the research project to invest themselves in the
task of source comprehension. Perhaps they did not conceive the research
188 writing & pedagogy

project as one in which they should engage with their sources, but instead saw
it as one in which they should find isolated sentences that might be useful in
their own texts. Qualitative research will be needed to answer these questions.
This work also does not indicate whether the 18 students whose written
work we studied are representative of all students at the institution in which
the data were gathered, much less whether they represent college students in
the aggregate. A larger, quantified, multi-campus study will be needed to make
such generalizations.
Our inquiry does not contradict Keck’s (2006) observation that patchwriting
occurs in most college students’ writing. Keck also found that the incidence
of patchwriting is higher among second-language writers. Because we did not
control for first language, we cannot affirm the latter observation, but all of the
college writers in our pilot research patchwrote.
This pilot study suggests that issues of source selection may be significant as
well. In the examples we have given, the students were striving to reproduce
extended information rather than argument, technical information on topics
that they may never have previously studied. These were papers being written
in a general composition class, not in discipline-specific instruction, which
means the student writers did not necessarily have any prior expertise in
the topics they chose to research. Nor, in the traditions of most composition
instruction, does the instructor necessarily have any expertise in the topics the
students are researching. Faced with reproducing extended technical informa-
tion and not wanting to copy long passages, the students might not have had
the vocabulary and background knowledge necessary to do anything but
patchwrite the passages.
Our observations also raise questions about problems students may have
with source-based writing, problems that are both prior to and foundational
to their correct citation of sources. Citation counts for little if what is being
cited is a fragmentary representation of the source. Plagiarism is difficult
to avoid if one is constructing an argument from isolated sentences pulled
from sources.
Our observations affirm the difficulties that all students have when using lan-
guage from sources – or trying to avoid doing so. In fact, Roig (2001) establishes
that all writers, even research faculty, struggle when writing from unfamiliar
sources on unfamiliar topics. In Roig’s experimental research, psychology
professors were given the task of paraphrasing text. Roig found that the more
difficult the source text, the more the professors appropriated language from it.
Twenty-two percent of the professors patchwrote: they made syntactic changes
to the original language of the source text. Twenty-four percent distorted the
meaning of the source (Roig, 2001: 315).
 Writing from Sources, Writing from Sentences 189

What we are illustrating in Figures 1, 2, and 3, then, are issues with which all
writers seem to struggle. However, despite the accumulating body of research
on writers’ intertextual struggles with their sources, these are not widely rec-
ognized as global issues; instead, they are widely regarded as malfeasance
committed by ignorant, indifferent, or unethical writers.

Conclusion
From this research, we are left with a compelling question: when writers work
from sources, to what extent are they accessing the entire source, and to what
extent single sentences from it? In the eighteen papers we examined, it is consis-
tently the sentences, not the sources, that are being written from. Perhaps some
or all of these writers had a comprehensive understanding of those sources but
chose to work only with isolated sentences within them. Or perhaps some or
all of these writers did not understand or did not engage with some or all of
their sources. Instead, they may have searched for “good” sentences and then
decided whether to paraphrase, copy from, or patchwrite from them. Again,
qualitative research will be needed to test these hypotheses.
Clearly, more research into the nuances of writers’ uses of sources is needed.
Interviewing or observing writers as they make their source-use decisions will
illuminate why they make the choices they do, and how committed they feel
to the educational ideals embedded in the task they have been given. Studying
writers in a variety of contexts will discourage fallacious overgeneralizations
about writing techniques. Do advanced undergraduates working in their
majors, for example, draw on sources in different ways than do sophomores
taking a required generic course in researched writing? Studying writers who
are reading a variety of genres will explore another possible factor. Do writers,
as Sherrard (1986) suggests, use sources differently when those sources are
narrative rather than expository? And how does source use vary according to
the genre in which the writers are working? Further issues involve approaches
to instruction that might improve students’ use of sources. What effects do
various pedagogies have on writers’ practices of source use?
All of our research questions will be pursued in the large-scale, quantified
study now called the Citation Project (CitationProject.net). But special atten-
tion will be paid to the question of source comprehension and summary and the
relationship between the two. Clearly our preliminary inquiry suggests that we
have much more to learn about whether students understand the sources they
are citing in their researched writing, whether they choose to summarize those
sources and the reasons for their choices, and the extent to which the absence
of summary correlates with a lack of source comprehension.
190 writing & pedagogy

About the Authors

Rebecca Moore Howard was awarded the PhD in English by West Virginia
University. She is now Professor of Writing and Rhetoric at Syracuse Univer-
sity, and her scholarship focuses on authorship studies, especially students’ use
of sources. Tanya K. Rodrigue earned her PhD in Composition and Cultural
Rhetoric at Syracuse University. Her doctoral dissertation examines the role and
needs of teaching assistants in writing across the curriculum. She is Andrew W.
Mellon Postdoctoral Fellow in Composition and Rhetoric at Wheaton College
(Massachusetts). Tricia Serviss was awarded the PhD in Composition and Cul-
tural Rhetoric by Syracuse University and is now Assistant Professor of Rhetoric
and Composition at Auburn University. Her dissertation reveals how the defini-
tion of and possibilities for literacy are constructed in disparate localities.

Notes
1 The entire text of the Gettysburg Address is as follows:
Four score and seven years ago our fathers brought forth on this continent,
a new nation, conceived in Liberty, and dedicated to the proposition that all
men are created equal.
Now we are engaged in a great civil war, testing whether that nation, or
any nation so conceived and so dedicated, can long endure. We are met on
a great battle-field of that war. We have come to dedicate a portion of that
field, as a final resting place for those who here gave their lives that that na-
tion might live. It is altogether fitting and proper that we should do this.
But, in a larger sense, we can not dedicate – we can not consecrate – we can
not hallow – this ground. The brave men, living and dead, who struggled
here, have consecrated it, far above our poor power to add or detract. The
world will little note, nor long remember what we say here, but it can never
forget what they did here. It is for us the living, rather, to be dedicated here
to the unfinished work which they who fought here have thus far so nobly
advanced. It is rather for us to be here dedicated to the great task remaining
before us – that from these honored dead we take increased devotion to that
cause for which they gave the last full measure of devotion – that we here
highly resolve that these dead shall not have died in vain – that this nation,
under God, shall have a new birth of freedom – and that government of the
people, by the people, for the people, shall not perish from the earth. (Lin-
coln, 1863)
2 We collected student papers with full anonymity for the writers; hence we do
not attach their names to the texts. In addition, our study works exclusively
with student texts and not students; hence we do not attach pseudonyms to
the papers. Like Shi, we are studying student texts, not students, so (again
like Shi) the texts are numbered rather than pseudonymed.
 Writing from Sources, Writing from Sentences 191

References
Angélil-Carter, S. (2000) Stolen Language? Plagiarism in Writing. New York: Longman.
Bainbridge, R. (2007) Recycling plastic. Ezine Articles. Retrieved 8 April 2007 from http://
ezinearticles.com/?Recycling-Plastic&id=354295
BBC News (30 June 2004) “Epidemic” of student cheating? Retrieved 30 June 2004 from
http://news.bbc.co.uk/2/hi/uk_news/education/3854465.stm
Broaddus, M. and Ku, L. (6 December 2000) Nearly 95 percent of low-income uninsured
children now are eligible for Medicaid or SCHIP: Measures need to increase enroll-
ment among eligible but uninsured children. Center on Budget and Policy Priorities.
Retrieved 16 July 2007 from http://www.cbpp.org/12–6-00schip.htm
Brown, A. and Day, J. D. (1983) Macrorules for summarizing texts: The development of
expertise. Journal of Verbal Learning and Verbal Behavior 22: 1–14.
Callahan, D. (2004) The Cheating Culture: Why More Americans Are Doing Wrong to Get
Ahead. New York: Harcourt.
Citation Project (2010) Retrieved 8 July 2010 from CitationProject.net.
Council of Writing Program Administrators (2005) Defining and avoiding plagiarism:
The WPA statement on best practices. Retrieved 22 June 2008 from http://wpacouncil.
org/node/9
Cox, L., Ray, J., and Lawler, K. (28 October 2004) What can consumer health assistance
programs and states DO to improve Medicaid and SCHIP enrollment and retention.
Families USA. Retrieved 16 July 2007 from http://www.familiesusa.org/issues/medicaid/
making-it-work-for-consumers/improving-medicaid-and-schip.html
Harris, J. (2006) Rewriting: How to Do Things with Words. Logan: Utah State University
Press.
Howard, R. (1993) A plagiarism pentimento. Journal of Teaching Writing 11(3): 233–46.
Jamieson, S. (2008) One size does not fit all: Plagiarism across the curriculum. In R.
M. Howard and A. E. Robillard (eds.) Pluralizing Plagiarism: Identities, Contexts,
Pedagogies 77–91. Portsmouth, New Hampshire: Boynton-Cook.
Keck, C. (2006) The use of paraphrase in summary writing: A comparison of L1 and L2
writers. Journal of Second Language Writing 15: 261–278.
Lincoln, Abraham (19 November 1863) Gettysburg Address. American Rhetoric Online
Speech Bank. Retrieved 8 August 2009 from http://www.americanrhetoric.com/
speeches/gettysburgaddress.htm
Pecorari, D. (2003) Good and original: Plagiarism and patchwriting in academic second-
language writing. Journal of Second Language Writing 12: 317–345.
Pecorari, D. (2006) Visible and occluded citation features in postgraduate second-
language writing. English for Specific Purposes 25: 4–29.
Pecorari, D. (2008) Academic Writing and Plagiarism: A Linguistic Analysis. New York:
Continuum.
Roessig, L. (March 2007) Making research matter. English Journal 96(4): 50–55.
Roig, M. (2001) Plagiarism and paraphrasing criteria of college and university professors.
Ethics & Behavior 11(3): 307–324.
192 writing & pedagogy

Sherrard, C. (1986) Summary writing: A topographical study. Written Communication 3:

324–343.
Shi, L. (2004) Textual borrowing in second-language writing. Written Communication
21(2): 171–200.
Weinberger, D. (2002) Small Pieces Loosely Joined: A Unified Theory of the Web.
Cambridge: Perseus.
West, L. (2007) How to recycle different types of plastic. About.com. Retrieved 1 June 2007
from http://environment.about.com/od/earthtalkcolumns/a/recycleplastics.htm
White, H. (2004) Citation analysis and discourse analysis revisited. Applied Linguistics
25(1): 89–116.