Iowa State University Capstones, Theses and

Retrospective Theses and Dissertations

Dissertations

1984

Run-on sentences and sentence fragments: is the

instruction treating the problem?
Lori J. Nielsen
Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/rtd

Part of the English Language and Literature Commons, Reading and Language Commons, and
the Rhetoric and Composition Commons

Recommended Citation
Nielsen, Lori J., "Run-on sentences and sentence fragments: is the instruction treating the problem?" (1984). Retrospective Theses and
Dissertations. 193.
https://lib.dr.iastate.edu/rtd/193

This Thesis is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital
Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital
Repository. For more information, please contact [email protected].
 Run-on sentences and sentence fragments:
Is the instruction treating the problem?

by

Lori Jan Nielsen

A Thesis Submitted to the

Graduate Faculty in Partial Fulfillment of the
Requirements for the Degree of
JVT.ASTER OF ARTS

Major: English

Approved:
Members of the Committee:

Signature redacted for privacy Signature redacted for privacy

Signature redacted for

privacy Signature redacted for privacy
For the Major
Signature redacted for privacy

Iowa State University

Ames, Iowa
1984
 ii

TABLE OF CONTENTS
Page
INTRODUCTION 1

CHAPTER I. REVIEW OF THE RESEARCH 7

Run-on Sentences 7
Summary 13
Sentence Fragments 14
Summary 21
Remediation of Run-ons and Fragments 22
CHAPTER II. ANALYSIS OF STUDENTS' RUN-ONS AND
FRAG~mNTS: DISCUSSION OF FINDINGS 27
Run-on Sentences 27
Sentence Fragments JJ
Analysis of Findings and Suggestions for
Remediation )8
CHAPTER III. ANALYSIS OF HANDBOOKS' TREATMENT OF
OF RUN-ONS AND FRAGMENTS 4)
Run-on Sentences 44
Sentence Fragments 49
Summary and Analysis 54
CHAPTER IV. SUW~RY AND DISCUSSION 57
Summary of Research Problem, Method, and Findings 57
Conclusions and Implications 60
NOTES 65
BIBLIOGRAPHY 72
 iii

APPENDIX A. RUN-ON SENTENCES: STUDENT SAMPLE WITH

GRAMMATICAL DESCRIPTIONS 76
APPENDIX B. SENTENCE FRAGMENTS: STUDENT SAMPLE WITH
GRAMMATICAL DESCRIPTIONS 80
 iv

LIST OF TABLES
Page
TABLE 1: Grammatical descriptions and examples of
run-on sentences from the research and from
the student sample JO
TABLE 2: Grammatical descriptions and examples of
sentence fragments from the research and
from the student sample 35
 1

INTRODUCTION

College freshmen, unaware of grammatical rules, write

run-on sentences and sentence fragments in their compositions.
Instructors justifiably mark those structures in students'
compositions as serious errors. By way of remediation, the
instructors assign sections in grammar handbooks that tell
students both how to avoid and how to correct those errors.
But some students continue to write run-on sentences and sen-
tence fragments. What, then, are effective ways to deal with
those sentence structure errors in freshman compositions?
Handbook approaches may or may not be effective, but run-
ens and fragments need to be dealt with in some way because
they are errors; they violate the verbal contract that exists
between the writer and the reader.l That is, a reader does
not expect a comma at the end of a sentence or a period at
the end of a dependent clause. Both run-ens and fragments
appear with frequency in students' compositions. John
Higgins analyzed college freshmen's placement themes and
found that "run-together sentences" with or without commas
occurred in 50% of all papers, appearing more than once in 27%
of them; fragments occurred in 45% of all papers, recurring
in 20%. 2 Furthermore, both appeared more often in papers
from what he identified as the lower remedial group.J Run-
ens and fragments are common errors in compositions, but what
 2

is the nature of these grammatical blunders? Why do they

occur? What knowledge or skills are students lacking? In-
structors must answer these questions if they are to help
their students write correct compound and complex sentences.
In compositions, the appearance of those two types of
sentences indicates a certain level of syntactic maturity.
Studies completed by Kellogg Hunt and Frank O'Hare, for in-
stance, show that as students mature, their ability to "con-
solidate a successively larger number of single sentences into
a singleT-unit" grows, so students reach successively higher
levels of syntactic maturity.4 Hunt defines aT-unit as "a
single main clause (or independent clause . . . ) plus what-
ever other subordinate clauses or nonclauses are attached to,
or embedded within, that one main clause."5 He reported that
"T-unit coordination blooms early"--by fourth grade--and that
subordination appears by grade eight.6 At that time, students
begin to use the appositive, and by college age, he found them
producing -ing structures ("The chicken cackled, waking the
man"). 7

In another study comparing students' and adults' writing,

Francis Christensen also found that a high incidence of sub-
ordination characterizes a mature style.8 However, Christen-
sen focused on what he called "final free modifiers" or sub-
ordinate structures "set off by junctures or punctuation" that
 3

are "loose or additive or nonessential or nonrestrictive."9

He defines final free modifiers as "prepositional phrases;
relative and subordinate clauses; noun, verb, adjective, and
adverbial phrases or clusters; and . . . verbid clauses or
°
absolutes." 1 Christensen found that not only would a mature
style "have a relatively high frequency of free modifiers,"
but those free modifiers would also occur most often in sen-
tence final position.1 1
These studies show that coordination appears in the
writing of students of a fairly young age. College students,
therefore, should not produce run-on sentences but should be
able to write compound sentences correctly. However, college
students with deficient writing abilities may not have mas-
tered the principle of coordination yet; those students may
require remedial instruction. Regardless, run-on sentences
at the college level would be considered as serious errors
since coordination is a skill college students should have
acquired. On the other hand, students at this level are still
learning to create and subordinate dependent grammatical
structures, and because some of these structures are unfa-
miliar, students punctuate them as sentences. Thus fragments
are errors but also evidence that students are attempting to
write more complicated sentences. However instructors view
run-ons and fragments, though, students evidently need to
 4

review the principle~ of coordination and subordination.

In this study, the terms coordination and subordination
will refer to the joining of clauses and phrases to form com-
plete sentences. Specifically, the principles of coordination
will refer to connecting two independent clauses to create a
compound sentence; the principles of subordination will sig-
nify linking a dependent phrase or clause to an independent
clause. Students by college age presumably have mastered
coordination and have acquired some facility with subordi-
nation. The existence of run-ons and fragments may suggest
an absence of grammatical knowledge of what constitutes inde-
pendent and dependent structures. Without that knowledge,
students are often unable to punctuate their sentences cor-
rectly. Their use of periods and semicolons may create frag-
ments, and their use of commas may produce run-on sentences.
Mina Shaughnessy maintains that ignorance of sentence struc-
tures and punctuation rules will produce run-on sentences and
fragments; to her, those errors indicate that students mark
off sentences according to the "rhetorical units" they per-
ceive, which results in word groups "longer or shorter than
the grammatical sentence ... 12 Not knowing the grammatical
make-up of complete sentences, students are bound to use the
wrong punctuation and consequently will produce run-ons and
fragments, even though their sense of juncture may be
correct. 13
 5

Apparently, to avoid writing run-ons and fragments, stu-

dents need to know the elements of a complete sentence, the
principles of coordination and subordination, and the perti-
nent punctuation rules. Students must have "an analytical
grasp of the sentence" in order to punctuate correctly the
rhetorical units their intuitions identify.14 Since most
composition instructors depend on grammar handbooks to teach
correctness, how accurately and efficiently do those texts
deal with run-ons and fragments? Furthermore, does the treat-
ment of those sentence structure errors reflect the research
on them? Most importantly, do the handbooks and the re-
search deal with the types of run-on sentences and sentence
fragments students actually write?
The following chapters will examine what the research
says about run-ons and fragments, what students' errors reveal
about the problems and solutions, and what handbooks present
as explanation and remediation. Specifically, chapter one
will review the research on run-ons and fragments and outline
the remedial approaches suggested for each error. Chapter
two will present an analysis and discussion of run-ons and
fragments from fourteen narrative essays written by students
in a college freshman composition class. Next, chapter three
will analyze handbooks' treatment of run-ons and fragments.
Finally, chapter four will summarize and draw conclusions
 6

about the effectiveness of most textbooks' instruction and

will then suggest a different perspective on run-on sentences
and sentence fragments and perhaps a more accurate approach
to remediation.
 7

CHAPTER I. REVIEW OF THE RESEARCH

Run-on Sentences
In general, the research defines a run-on sentence as one
of two types: (1) a run-together sentence, or two independent
clauses joined without punctuation, and (2) a comma splice,
or two independent clauses linked by only a comma. Comma
splices receive most of the attention in the research, al-
though both types of run-ons indicate that students do not
recognize grammatically independent clauses and do not use
punctuation marks correctly. According to Shaughnessy, lack-
ing those abilities and thinking in rhetorical units larger
than sentences, students hesitate to use a period when their
thoughts go on; they use a comma instead.16 Their commas con-
nect sentences that make rhetorical sense together: "'In her
late teens my mother looked for enjoyment, I'm the end re-
sult.'"l7 Do run-ons indicate a stage in some students' de-
velopment of the ability to coordinate independent clauses
correctly?
According to a study completed by Philip DiStefano and
Robert Marzano, run-on sentences may occur when students first
attempt more complicated structures and so may be indicators
of growing syntactic maturity. DiStefano and Marzano analyzed
compositions of varying quality written by students grouped by
ages nine, thirteen, and seventeen. They sought to identify
 8

factors in compositions that would be "good predictors of

quality at different grade levels ... 18 The ability to coor-
dinate begins in grade school, but run-on sentences were
named "positive predictors of quality at all grade levels ... 19
Unfortunately, DiStefano and Marzano fail to explain why run-
ons are "positive predictors." However, as they discuss the
factors predictive of quality, they state that "there is def-
initely a logical progression in the way age and quality of
writing are connected to base clause expansion." 20 Logically,
then, as students try to write more complicated sentences--
adding independent clauses to expand a base clause, for ex-
ample--they would initially produce more errors. Run-on sen-
tences could therefore predict a developing capability of
writing correct compound sentences. As syntactic maturit~

grows and students learn to coordinate two independent clauses

correctly, the incidence of run-on sentences should decrease.
Understanding that run-ons may be indicative of syntactic
growth, instructors may recognize that the college students
who produce them need supplementary lessons to review coor-
dination. Those remedial lessons would be additionally help-
ful if instructors could focus upon the grammatical contexts
in which run-ons commonly occur. Do the independent clauses
college students run together fall into common grammatical
patterns? In a study done to answer that question, Dona Kagan
 9

notes two common patterns in run-on sentences: a short sen-

tence followed by a longer one, and a strong verbal element,
such as a compound verb or participial phrase, in the longer
sentence.21 Kagan asked remedial college students to identify
the run-ons and complete sentences on a prepared test that
listed, without punctuation, eleven run-ons and four complete
sentences. Kagan observed three main patterns in the stu-
dents' selections of items as run-ons: short sentence + long
sentence with a subordinate clause; short sentence + long sen-
tence with a compound verb; long sentence with a participial
phrase + short sentence. 22 This unit from her test illus-
trates the pattern of a short simple sentence followed by a
long sentence with a participial phrase: "The door slammed the
woman clenched her fists digging her nails into her palms."2J
(Other examples of Kagan's patterns are in Table 1 on page JO.)
Kagan reasoned that students define "'complete sentence'" as a
unit which exceeds "a certain minimal length" and contains a
verb-noun sequence.24 The students' choices seem to indicate
ignorance of grammatically complete sentence structures; how-
ever, Kagan's results reflect an artificial situation. Stu-
dent-written run-ons may not follow the structural patterns
Kagan's study described. Her study does reveal that students
lack grammatical knowledge about sentence structure, but what
other information do they need that would enable them to avoid
writing run-on sentences?
 10

Students may continue to write run-on sentences if the

principles of punctuating two independent clauses do not be-
come clear. Therefore, since students commonly link related
sentences using only commas, they evidently need to know the
correct means for connecting sentences.25 The implication is
that students misunderstand or are unaware of the accepted
uses of commas and are ignorant of how to use semicolons.
Then a run-on sentence may be merely a punctuation error, for
students are joining rhetorically related sentences at the
correct juncture, adhering to that principle of coordination.
What students need to learn is the acceptable use of punctua-
tion marks for coordinating two independent clauses. William
Irmscher explains that if students recognize a period as a
"terminal mark" and a comma as an "interrupting mark," then
the other punctuation marks can be learned as "sophisticated
variations of those two," for he contends that all the marks
perform the same functions but to varying degrees.26
From this perspective, Irmscher recommends building a
system for the usage of punctuation marks, since a run-on sen-
tence is not "a failure to write complete sentences but a
failure to punctuate them correctly ... 27 In Irmscher's system,
punctuation marks can be explained and related by their func-
tions. A period terminates or has "a stopping and separating
function;" a comma interrupts or performs a dual separating-
 11

combining function--"a break, but going on. " 28 A comma by it-

self is therefore inadequate to separate two sentences; it
must be used with a coordinating conjunction to perform that
role. However, a semicolon, while operating as a comma to
link, is also strong enough to separate two sentences by it-
self or with a conjunctive adverb. After building such a
system of punctuation marks with the comma as the base, each
mark can be comparatively explained, and Irmscher concludes
that "all the rest is illustration, practice, and applica-
tion. n29
Since run-on sentences are incorrectly punctuated, then
explaining the correct usage of punctuation marks should help
students produce grammatically correct compound sentences. As
they practice writing these constructions correctly, according
to Irmscher they will gradually learn that the means of coor-
dination used is also a rhetorical choice.JO However, the re-
search emphatically points out that while running sentences
together without punctuation or connectives is always incor~

rect, comma splices are often rhetorically effective and ac-

ceptable. Several conditions must be met, though: the clauses
spliced together must be short, of parallel grammatical struc-
ture, and closely related or logically connected in meaning.3 1
In fact, a comma splice may actually be preferable when it
creates no "syntactic ambiguity" and lends "greater fluency of
 12

expression" to the passage . .32 For example, in "He filled the

beaker with liquid, he heated it on a butane burner(,) and
he took readings at equal intervals of _time," the independent
clauses are short, simple sentences, each beginning with the
personal pronoun he as the subject; since the clauses state
steps in a process, they are logically related . .J.J
While instructors usually prohibit the use of such comma
splices in expository writing, most of the research counters
that comma splices often appear in the works of professional
writers where they are not considered errors . .3 4 Instead, in
narrative/descriptive writing, a comma splice is often "a
good and appropriate, sometimes necessary, use of the com-
ma.".3.5 However, before students are taught the conditions
that will allow comma splices, they need to learn how to coor-
dinate two independent clauses correctly. If students write
run-ons, they are linking semantically related sentences,
which indicates a rhetorical consideration. Therefore, what
these students apparently do not know is how to punctuate a
compound sentence correctly. They must learn, first, to rec-
ognize when independent clauses are joined inadequately, and,
second, to judge what punctuation mark will correctly link the
clauses under what conditions, such as a comma being adequate
only when a coordinating conjunction follows it. After they
master the correct alternatives, then the acceptable uses of
comma splices can be discussed.
 13

Summary
Comma splices receive the most attention in the research,
although both types of run-on sentences are errors because a
reader expects to stop at the end of a sentence. The lack of
punctuation between two independent clauses or the use of a
comma between them causes confusion. The research explains
that run-ons are caused by a failure to recognize two inde-
pendent clauses and by misunderstanding the rules of punc-
tuation. Because coordination is an early-developing skill,
most college students should link two rhetorically related
independent clauses intuitively. Since only one study
specifies the grammatical construction of the two clauses run
together, few conclusions can be made regarding the gram•
matical composition of run-on sentences. When errors occur
at the college level, instruction in punctuation may ade-
quately alleviate the problem of run-ons. Lessons that review
the structure of independent clauses and the principles of
coordination and teach the uses of punctuation marks should
reduce the number of run-ons in college students' compositions.
Although the research claims that under some conditions comma
splices are acceptable, college instructors usually do not
accept them in expository writing.
 14

Sentence Fragments
Sentence fragments are another common type of sentence
structure error college students make in their compositions.
Fragments, like run-ons, generally occur because students use
the wrong punctuation mark; using a period instead of a comma,
students inadvertently cut off part of a sentence and create
a fragment. In other words, a fragment occurs when students
"break a single grammatical sentence into two punctuated sen-
tences incorrectly," as in "'My mother had four children. On-
ly because she had no choice.'"J6 Shaughnessy explains that
students punctuate by rhetorical, not grammatical, units, as
in the preceding example; that is, students' sense of com-
pleteness is not a grammatical sense, so they mark off as sen-
tences units that may present a complete idea but are depen-
dent grammatical structures.37 Frequently, fragments are
grammatically dependent sentence final structures that have a
sense of rhetorical independence, as in "'We would live off
the earth, and nature. Living together to survive .... JS Here
the student uses a period instead of the comma necessary to
attach the sentence final participial phrase to the base
clause.
Students punctuate a number of dependent grammatical
structures as sentences. According to Shaughnessy, sentence
initial adverbial clauses may become fragments because
 15

students, by the time they reach their ma~n thought, forget

the "initial word that suspends the clause," as in "'But I
think if people could learn to think a little like children.
This would be a better place to live.'"J9 Shaughnessy also
claims that the second part of a compound structure often be-
comes a fragment; in this example, the second verb of a com-
pound verb in a noun clause is punctuated as a sentences "'I
agree that the little boy has seen beauty. But is confused
with the question.'"40 Sarah D'Eloia names several other
structures that commonly become fragments in students' writ-
ing: "long introductory phrases, subjects divided from verbs,
verbs separated from subjects, and adverb clauses ... 4l (Ex-
amples of these fragments are in Table 2 on page J5.) D'Eloia
supports Shaughnessy's stance that fragments occur because
students punctuate according to rhetorical units, not gram-
matically complete units. That is, to D'Eloia, fragments are
"errors of punctuation rather than grammar," yet they arise
because students, reading their sentences, have no other means
to tell how to punctuate them than by using the length of the
units or the pauses between them. 42 She explains that frag-
ments occur "because the student is unable to establish sen-
tence boundaries by distinguishing independent clauses from
all the other structures which can attach to them ... 4J
Kagan names other dependent structures as being common
 16

sentence fragments. As she did for run-ons, Kagan prepared a

test on fragments which she administered to·. the same remedial
students; they were to indicate which items were complete sen-
tences and which were fragments. This test listed "fifteen
different combinations of syntactic structures" that were
fragments and five that were complete sentences.44 As she did
for run-ons, she again found patterns in the students' selec-
tions of fragments as complete sentences: verb + subordinate
clause, verb + direct object +prepositional phrase, two prep-
ositional phrases. 4 5 (Fragments from Kagan's test that illus-
trate these patterns are in Table 2 on page 35.) Kagan there-
fore concluded that a verb at the head of a structure is ap-
parently a "common miscue" that students use when identifying
sentences, as in "Shouted loudly because he knew she was
deaf."46 Kagan also speculated that some inherent quality in
prepositional phrases--perhaps their length or function as a
complete grammatical unit--causes students to identify them as
complete sentences.47 Unfortunately, these patterns which
Kagan identified as common types of sentence fragments are
drawn from an artificial situation, not from actual students'
writing, and they may not be valid.
The results of Muriel Harris' study are more reliable, as
she analyzed student-written fragments. In order to prepare
instructional materials for a writing lab, she examined one
 17

hundred fragments from essays students brought to the lab

from their composition classes. She found that free modi-
fiers occurring after the base clause predicate often become
fragments. Specifically, she, as Christensen, identified
these fragmented final free modifiers as "nonessential prep-
ositional phrases; relative and subordinate clauses; nomina-
tive absolutes; and noun, verb, adjectival, and adverbial
phrases ... 48 (Examples of these structures as fragments are in
Table 2 on page 35.) Of the student fragments analyzed, 83%
were these "modifying phrases and clauses which should have
been included after the base clause" but which were separated
from it instead by a period.49 These final free modifiers,
as in "Her arms were long and small, but there was strength
in them. Hands rough and calloused from long hard hours of
work." are "late blooming structures ... 5° Harris concurs with
Christensen's findings that final free modifiers are evidence
of a developing mature style, because they are found in the
works of mature writers.5 1 Therefore, as college-age students
learn to handle these structures, they often punctuate them
incorrectly as sentences, not yet fully understanding the
principles and methods of subordination nor the use of the
comma for joining.
Harris, like Christensen, claims that final free modi-
fiers are signs of increasing syntactic maturity. Therefore,
 18

when struggling to produce longer, more complex sentences by

adding words and clauses per T-unit, students will likely pro-
duce fragments in the process. At first, students may punc-
tuate final free modifiers as sentences until they learn how
to control these structures by joining them to the base
clauses with which they belong. Developing this skill to sub-
ordinate begins early in grade school, according to Constance
Weaver. After studying fragments students wrote in grades one
through six, Weaver reported that students produce a fairly
stable proportion of fragments per grade level; however, the
types of fragments change from level to level as students be-
gin using more mature grammatical structures.52 For instance,
at first grade she found that students will punctuate explana-
tory "because" clauses as sentences, and then in second grade
they will cut off the second part of a compound .structure; by
sixth grade they write the different types of subordinate
clauses as sentences.53 So as students start learning about
and producing new grammatically dependent structures, initial-
ly they may write them as sentences, creating fragments. As
they continue to write from grade to grade, learning more
about sentence structures and punctuation, ideally they will
master the use of dependent clauses and phrases and not write
them as fragments.
When students at all grade levels begin transforming
 19

complete ideas into dependent grammatical structures, they

may at first punctuate the structures as sentences.54 As
Shaughnessy suggests, students may fail to attach the struc-
tures because subordinating them at the ends of sentences
requires students to remember their base sentences.55 Ap-
parently, these students have not yet developed the semantic
and syntactic abilities to do that. However, not all sentence
fragments are considered errors. Two studies--one by Harris,
one by Charles Kline and W. Dean Memering--categorize sentence
fragments as "broken sentences" and "minor sentences."56 They
define a broken sentence as a "fragmented, discontinuous, and/
or noncontinuous thought," and both studi.es concur that it is
an error and should be avoided.57 Harris gives as an example
of a broken sentence, "'In a little night club in Louisville,
a couple of my friends, Rick and Lon, the duo who were pro-
viding the entertainment that night for the club. Rick plays
an organ with three synthesizers included.'"5 8 She explains
the first unit as a broken sentence because attaching it to
the second unit will not complete it, so this type of frag-
ment is an error.59
On the other hand, Harris, Kline and Memering contend
that minor sentences "express a complete idea or complete a
previously stated idea minus one or more of the items typi-
cally present in an English sentence," so a minor sentence
 20

needs to be obviously related to an adjacent "written unit."60

Kline and Memering explain that minor sentences could be ques-
tions, answers to questions, or could occur in dialogue, but
they must be written deliberately for specific rhetorical and
stylistic purposes and. not occur unintentionally.61 Harris
suggests that minor sentences occur more frequently and are
not errors, yet this example she provides, although it meets
the prescribed conditions, does not seem to be an intentional
fragment: "'She had a very funny look on her face. As if she
was scared and just wanted to be left alone.'"62 Here the
fragment or minor sentence can be joined logically to the base
clause. A better example of a minor sentence might be "The
blue car." in answer to the question, "Which car is yours?" in
a dialogue.
However, students do not necessarily need to know that
these types of situations allow sentence fragments until after
students can recognize and correct fragments in their own
compositions. Besides, as Kline and Memering point out, sen-
tence fragments are usually not permitted in the types of
formal, expository writing done in college composition
courses.63 If students do write compositions that contain
dialogue or question-answer situations, then teachers may give
examples of fragments allowed in those contexts. Otherwise,
for whatever mode of discourse, students should first gain
 21

control of complete sentence structures, learning to embed

and attach clauses and phrases correctly. After showing that
they can consistently produce complete sentences, students
could learn to examine the style and purpose of their dis-
course to ascertain where a sentence fragment might be more
effective than a complete sentence.

Summary
A sentence fragment is an error that occurs when students
punctuate as sentences rhetorical units that are actually de-
pendent grammatical structures. The research cites a number
of those dependent structures as common sentence fragments,
and most often they are final free modifiers. At the college
level, fragments are frequently subordinate clauses. As in-
dicators of syntactic maturity, fragments occur as students
learn to subordinate increasingly more complicated grammatical
structures. Then from grade to grade, as the research sug-
gests, students fragment different structures as they struggle
to subordinate each successfully. To facilitate the process,
instructors should present the principles of subordination
and teach the difference between independent and dependent
sentence structures. Although the research amply covers the
rhetorically acceptable uses of sentence fragments, these sit-
uations would normally not occur in the types of expository
writing students do in college. Besides, before students
 22

should be allowed to experiment with acceptable uses of frag-

ments, they should indicate that they know the requirements
of grammatical completeness by writing complete sentences and
correcting any fragments they find in their compositions.

Remediation of Run-ons and Fragments

Since run-ons and fragments are common errors that occur
with some frequency in students' compositions, how can teach-
ers best deal with them in the classroom? What, according to
the research, needs to be taught? As suggested, run-ons and
fragments occur because students punctuate sentences according
to the length of the structures, the pauses they hear, or a
sense that the structures are related somehow; students fail
to distinguish independent clauses from each other and from
structures that can attach to those clauses and often closely
resemble them.64 These inabilities suggest that students must
learn to analyze and recognize sentence structures and learn
to use punctuation correctly, so students need a basic grammar
vocabulary, such as independent and dependent clause, subor-
dinator, conjunction, and semicolon. Students also need to
review coordination and subordination. While all this know-
ledge can be obtained from grammar-handbooks, Irmscher sug-
gests that teaching the grammar of sentence structure be in-
corporated with teaching punctuation usage. 6 5 Punctuation
should be taught not as rules but as a system "largely deter-
mined by syntax."66 In other words, students should develop
 2.3

an awareness of the grammatical units that require commas, for

instance, rather than learn all the rules f9r using commas.
"Unless the intuitive sense of closure is ·counterbalanced by
syntactic certainty, all kinds of blunders can occur ... 67
Aside from suggesting that students need to learn basic
sentence structures and punctuation usage and review the prin-
ciples of coordination and subordination, the research pro-
vides few specific techniques for remediating run-ons and
fragments. D'Eloia suggests that students apply a structural
analysis to items in a group of semantically related struc-
tures to discern which are complete and which are fragments;
otherwise, only Shaughnessy suggests an instructional plan for
remediating run-ons and fragments.68 First, she recommends
that students learn such basic concepts as subject, verb, and
object so their sentences can be discussed, and students
should know the names and functions of punctuation marks; then
"the study of punctuation ought to be a study of sentence
structure, not merely a definition of the marks themselves ... 69
Punctuation usage and sentence structure should be taught at
the same time because "the process whereby writers mark sen-
tences is related to the process whereby they make them."7°
After proposing this basic instruction, Shaughnessy makes
more directed suggestions for helping students avoid writing
run-ons and fragments. First, she claims that students should
learn the options for connecting independent clauses: comma +
 24

conjunction, semicolon, and semicolon + conjunctive adverb.71

She believes that students will need to practice using the
latter, however, by supplying the most logical conjunctive
adverb between two statements in prepared exercises.72
Shaughnessy also recommends that students be taught the dif-
ferences between linking and embedding independent clauses
and thereby learn the principles of subordination, too.73
To learn not to write sentence fragments, students need to
practice embedding, by which she says they will notice how
embedded units "adhere" to complete sentences and require
commas when in sentence initial position.74 Moreover,
Shaughnessy believes that practice with embedding will give
students a basic vocabulary of common subordinators, "words
that signal embedding," and as such may also signal a frag-
ment if the unit is punctuated as a sentence.75
Shaughnessy, therefore, prescribes sentence combining to
give students practice producing correct compound and complex
sentences. This activity, she claims, allows students to
generate both types of sentences while retaining grammatical
control and also lets students see how those structures can
go wrong and become run-ons and fragrnents.76 In addition,
sentence combining can "increase syntactic maturity in writers
of college age."77 When used in combination with basic in-
struction in punctuation and grammar, sentence combining will
actually help reduce the number of errors students make,
 25

including run-ons and fragments. This is the conclusion Carol

David drew after studying the effectiveness of a program she
used with remedial students. They were taught clause analy-
-
sis, coordination, and subordination, along with punctuation
usage; sentence combining exercises complemented the instruc-
tion.78 She found that, after completing the program, stu-
dents "reduced their errors by more than half."79 Elaine
Maimon and Barbara Nodine reported similar results from a
writing program that employed sentence combining. They found
that embedding errors, including run-ons and fragments, in-
creased initially, but ultimately T-unit length increased and
the number of errors decreased.so
A final suggestion for reducing the number of run-ons
and fragments in students' writing comes from Donna Gorrell.
Called controlled or guided composition, her suggestion is an
alternative or precursor to sentence combining. Controlled
composition requires students to copy and eventually manipu-
late prewritten material, asking them to employ the grammati-
cal knowledge they have but do not use.81 Since these activ-
ities demand accurate transcriptions and manipulations, stu-
dents must attend to the "lexical and syntactic forms in the
written language;" eventually students are asked to do "con-
trolled phrase and clause embeddings." 82 Controlled compo-
sition activities then provide practice in writing correctly
without direct grammar instruction. This activity plus
 26

sentence combining are the most specific exercises the re-

search prescribes to remediate run-on sentences and sentence
fragments. Otherwise, the research only specifies the in-
struction necessary to prevent students from producing those
incorrect sentence structures.
 27

CHAPTER II. ANALYSIS OF STUDENTS' RUN -ONS

AND FRAGMENTS: DISCUSSION OF FINDINGS

Although not extensive, the research on run-on sentences

and sentence fragments gives teachers of composition a better
understanding of why those problems occur in students' writing,
perhaps suggesting a new perspective for dealing with the
errors. Remediation of both sentence structure problems would
be more direct if teachers understood the grammatical contexts
in which each commonly occurs. This knowledge would allow
teachers to narrow their focus and be more precise in their
instruction. What, then, are the circumstances and contexts
under which students write run-ons and fragments?

Run-on Sentences
According to the research, run-on sentences are of two
types: a comma splice or a fused sentence. In both cases,
students connect two sentences that they perceive as one rhe-
torical unit, which indicates that they remember the principle
of coordination but use inadequate punctuation. They fail to
recognize two independent clauses punctuated as one sentence,
and they do not use the correct punctuation mark or a con-
junction at the juncture.
From the research, only Kagan's study specifies gram-
matical contexts for run-on sentences, but do those contexts
represent run-on sentences actually written by students? To
 28

find out, this study examined fourteen randomly selected

narrative essays written by students in a college freshman
composition course. The twenty-one run-on sentences found in
them indicate that Kagan's contexts may not be accurate. The
fourteen essays examined were the first assigned compositions
of a semester course and were completed outside of class.
Run-ons were explained in class as an error that would be
marked in the compositions, but no instruction was given on
coordination and punctuation. In the fourteen compositions
examined, twenty-one run-on sentences occur in only seven of
the compositions. One composition contains two run-ons, one
has three, one has four run-ons, and one composition contains
.
nine. Three compositions contain one run-on each. Eighteen
of the run-on sentences are comma splices while only three
are fused or run-together sentences, so the concentration on
comma splices in the research seems justified.
Contrary to Kagan's study, the majority of student run-
ens do not follow the "short sentence + long sentence" pat-
tern; only three sentences do. Nine run-ons are long sen-
tences followed by short sentences, and in the remaining nine
run-ons, the sentences are of similar length. Kagan also
states that in the items students selected as run-ons, the
long sentences generally contained participial phrases, com-
pound verbs, or subordinate clauses. In the run-on sentences
 29

the students in the sample wrote, only one of the long sen-
tences contains a participial phrase, one has a compound verb
(another one is also a compound sentence),· and four have de-
pendent clauses. In fourteen of the run-ons, both sentences
are simple sentences. The results of Kagan's study do not
seem to reflect accurately the types of run-on sentences stu-
dents write. While the twenty-one student-written run-ons are
not an adequate sample from which to draw definitive conclu-
sions, they nonetheless raise questions and show the need for
more extensive research.
The Harper Handbook of College Composition offers a dif-
ferent list of four grammatical contexts for comma splices
which prove more accurate than Kagan's findings. The contexts
are as follows: (1) the two joined sentences are related by
content but not by syntax; (2) the second sentence begins with
a personal pronoun whose antecedent is in the first sentence;
(J) the second sentence begins with a demonstrative pronoun or

adjective; (4) the second sentence begins with or contains a

conjunctive adverb. 8 J Half of the student sample run-ons fit
two of these categories. Ten of the second sentences begin
with personal pronouns whose antecedents are in the first sen-
tences, and in two of those, the second sentences begin with
demonstrative pronouns. Table 1 below gives ·examples of run-
ons with grammatical descriptions from Kagan's study, The
 30

Harper Handbook, and the student sample; the similarities

and differences among the three sources are apparent.

Table 1. Grammatical descriptions and examples of run-on

sentences from the research and from the student
sample

grammatical description example

Kagan
1. short sentence + long "The phone rang he did not
sentence with subordinate hear it because he was in the
clause shower" (p. 134)
2. short sentence + long "It snowed heavily people
sentence with compound watched and waited in their
verb cabins" ( p. 134)
3· long sentence with "Grinning from ear to ear he
participial phrase + entered the house the~ were
short sentence all waiting" (p. 134)
The Harper Handbook of
College Composition, 5th edition
1. two sentences related by "A meeting of the club is
content, not by syntax scheduled for tonight, many
important items are on the
agenda." (p . .328)
2. second sentence begins "The ambulance driver exam-
with a personal pronoun ined the victim carefully,
with its antecedent in he did not say a word."
the first sentence (p. 328)
3· second sentence begins "Drive carefully when you
with a demonstrative pro- approach the bridge, this is
noun or adjective very narrow." (p . .328)
4. second sentence begins "I was late for the lecture,
with or contains a con- however, Ms. James did not
junctive adverb scold me." (p . .329)
 .31

Table 1. (continued)
grammatical description example
Student sample
1. long sentence + short "Once in awhile I will bring
sentence or sentences back the wonderful memories and
of similar length shed a tear over his loss, it
was a, great one to me."
second sentence is a (see example one)
simple sentence
.3· subject of second sentence (see example one)
is first word
4. subject of second sentence (see example one)
is personal pronoun that
links it rhetorically with
first sentence

A more accurate analysis might show more significant re-

sults from an examination of the second sentence in a run-on.
The structure of that sentence, the position of its subject,
and the use of a pronoun for the subject are the most apparent
factors in the analysis of the twenty-one student run-ens.
Twenty of the second sentences are simple sentences, and in
sixteen the first word of the second sentence is the subject.
In nineteen of the second sentences, that subject is a pro-
noun (seventeen are personal pronouns, two are demonstrative
pronouns). In six of those nineteen, the pronoun is the same
one that stands as subject in the first sentence. In fourteen
of those nineteen cases, the pronoun's antecedent is in the
first sentence (the antecedent is often, however, another pro-
noun and not a noun). In the analysis of the students' run-
ens, then, five results seem significant: (1) most run-ons are
 .32

comma splices; (2) the short sentence usually follows the

longer one when they are of different lengths; (.3) the second
sentence is usually a simple sentence; (4) the subject of the
second sentence is often the first word; (5) the subject of
the second sentence is usually a pronoun that rhetorically
links it with the first sentence. (See Appendix A for the
student sample run-ens with grammatical descriptions.)
With run-on sentences, then, students do seem to be punc-
tuating according to rhetorical units. Pronouns are a link to
the preceding sentence, so students use a comma instead of a
period between the two sentences to recognize the relation-
ship. The sense of juncture is correct, but students do not
realize that a comma is inadequate to separate two independent
clauses. To proofread for run-ens, students should scan their
sentences for commas; if the word following a comma is a per-
sonal pronoun or a noun, then they should analyze the struc-
tures on each side of the comma to determine if they are inde-
pendent clauses. Instruction should explain sentence analysis
and review the principles of coordination so that students
know that they must insert a conjunction after the comma or
replace it with a semicolon or period, for students who write
run-ens have recognized the correct juncture. Remediating
run-on sentences, then, is largely a matter of clause analysis
and instruction in punctuation usage.
 33

Sentence Fragments
With sentence fragments, the students' sense of juncture
also seems to be accurate, but often punctuation incorrectly
separates sentence final dependent clauses from their indepen-
dent base clauses. The research defines a sentence fragment
as a single grammatical sentence incorrectly broken into two
sentences by a period. The period separates two rhetorical
units: one is an independent clause, the other a dependent
grammatical structure that students incorrectly sense is com-
plete by itself. Although students use a period instead of a
comma, the complete sentence shows growing syntactic maturity
because students are trying to add clauses and phrases per
T-unit. They create dependent structures but do not yet at-
tach them to main clauses. Students' intuition for joining
(commas) instead of terminating (periods or semicolons) is
not quite formed. With fragments, students are applying the
principles of subordination and have an accurate sense of
juncture. Their failure is in comprehending independent and
dependent structures and in using appropriate punctuation.
Several dependent grammatical structures are commonly
cited as being fragments in students' compositions. Shaugh-
nessy names two structures: initial adverbial clauses and the
second part of compound structures. Kagan, from her test,
cites three structures that students selected as sentences:
 J4

verb + subordinate clause, verb + direct object +preposition-

al phrase, and two prepositional phrases. Harris, in an anal-
ysis of real student fragments, reports that of those "which
got detached from their base clauses by inappropriate inser-
tions of periods," 7% were in sentence initial and 8J% in sen-
tence final position.84 Of the latter, 26% were "primarily
subordinate clauses at the ends of the sentences," and 57%
were final free modifiers: noun phrases, nominative absolutes,
and verb phrases.85 In 24% of those final free modifier frag-
ments, "some element . . . was compounded." 86
Analysis of the student sample fragments from the four-
teen student compositions supports Shaughnessy's and Harris'
findings but not Kagan's. However, ·only eight sentence frag-
ments were found in five of those fourteen compositions. No
instruction had been given on subordination prior to the com-
position assignment, and fragments were cited as being major
errors. Seven of the students' fragments are in sentence
final positionJ four are cut off from the main clause by pe-
riods, four by semicolons. All the fragments require commas
to attach them to their base clauses. Each fragment contains
a dependent clause. The fragment in sentence initial position
contains two participial phrases and an adjective clause. One
of the sentence final fragments contains the second part of a
compound verb preceded by a participle and an adverb clause.
 35

One of the fragments produced by a semicolon contains however

as a transition. (See Appendix B for the student sample
fragments with grammatical descriptions.)
In support of Harris' findings, four of the fragments
that follow the base clause are final free modifiers. None
are nominative absolutes, but two are noun phrases and one is
a verb phrase. One is the second part of a compound structure,
which Shaughnessy also notes as being frequently fragmented.
There is also, as Shaughnessy indicates is common, a long in-
troductory clause fragmented, though it is an adjective rather
than an adverb clause. To show how the grammatical structures
from the research and the student sample compare, Table 2 be-
low provides examples with grammatical descriptions.

Table 2. Grammatical descriptions and examples of sentence

fragments from the research and from the student
sample
grammatical description example
Shaughnessy
1. initial adverbial "'But I think if people could learn
clause to think a little like children.
This would be a better place to
live. '" ( p. 25)
2· second part of a "'I agree that the little boy has
compound structure seen beauty. But is confused with
the question.'" 'P· 25)
Kagan
1. verb + subordinate "Shouted loudly because he knew she
clause was deaf" (p. 131)
2. verb + direct object "Placed his hat on the hook"
+ prepositional (p. 1.31)
phrase
 36

.Table 2. (continued)

grammatical description example

Kagan (continued)
3· two prepositional "In the corner under the table"
phrases (p. 131)

Harris
1. sentence final sub- "'Playboy has a reputation for
ordinate clause getting a sophisticated and elite
group of readers. Although this is
a value judgment and in some cir-
cumstances, not a true premise.'"
(p. 180)
2. final free modifier, "'I believe that the author is try-
noun phrase ing to convey the meaning of life
to the reader. A sense of purpose
and fulfillment to life.'" (p. 180)
3· final free modifier, "'The story appealed to your sense
nominative absolute of nostalgia and proved a point.
The point being that at maturity we
have to fit into a style and be-
come responsible.'" (p. 180)
4. final free modifier, "'She opened the door and let us
verb phrase into her home. Not realizing at
the time that we would never enter
that door in her home again.'"
(p. 180)

Student Sample
1. final free modifier "In an attempt to break my nervous-
verb phrase ness, I decided to leave early for
(misuse of semicolon) school; hoping the fresh air and
sunshine would help."
2. sentence initial "After straightening myself in the
adjective clause with mens room, where I received a
two participial strange look from a guy seeing me
phrases change. I go to class and am about
five minutes late."
 J7

Table 2. (continued)
grammatical description example

J. sentence final frag- "Another nurse asked me to lay on

ment with second part a table; positioning me so that I
of compound verb was extremely uncomfortable and
told me not to move."
4. however used as a "He was traveling so fast; however,
conjunctive adverb that his car was able to turn our
pick-up truck one hundred and
eighty degrees around."
5· final free modifier, "My adventure began when Neal, Phil
noun phrase and I set out for our first attempt
of this bright and blue day to ex-
perience the sport of water skiing.
A sport which requires both skill
and endurance."
6. final free modifier, "In an attempt to break my nervous-
ve:r;"b phrase ness, I decided to leave early for
school; hoping the fresh air and
sunshine would help.
7· second part of a "Once informed, I began first by
compound structure putting on my life preserver. Then
my skiis, which were like being
fastened to the floor."

While there are only eight fragments to analyze from the

student sample, at least three conclusions seem notable: (1)
most fragments occur in sentence final position; (2) they usu-
ally contain one dependent clause; (J) the period or semicolon
preceding them indicates a correct juncture that needs to be
filled by a comma. This final conclusion again indicates that
students do not understand punctuation usage. Because four of
the fragments were created by semicolons, students specifi-
cally need to learn when to use that mark. They discern a
juncture between the independent and dependent clauses and may
 )8

recognize that a period is not warranted, yet students want a

mark stronger than a comma. For this reason, perhaps they use
a semicolon. So while technically the semicolons are creating
fragments, these seem like errors of punctuation rather than
sentence structure errors.
Learning the correct use of semicolons, however, requires
clause analysis, which is also necessary if students are to
learn the principles of subordination so that they will join
dependent clauses to base clauses instead of punctuating them
as sentences. Because subordination is a skill students begin
to acquire at an age later than that at which coordination is
mastered, the principle needs to be reviewed thoroughly be-
cause students are still learning to subordinate different
types of dependent structures, such as -ing structures.

Analysis of Findings and

Suggestions for Remediation
Of the seven students who wrote run-ons in the sample,
only two also wrote sentence fragments; the student who wrote
nine run-ons wrote no fragments. This student may have had
deficient syntactic skills and may not have mastered the
punctuation used with coordination nor have learned how to
use subordination at all. Perhaps some of the run-on sen-
tences from the sample actually indicate an attempt to sub-
ordinate. That is, when the second sentence of a run-on is
 J9

short and has a personal pronoun in initial position as the

subject, the writer may have meant to subordinate it to the
preceding sentence to which it is closely ·related. For ex-
ample, instead of writing, "I thought we had it made she just
stood there shaking," the student may have intended to say
··because she just stood there shaking." The student who wrote
this run-on produced no fragments in his composition, so per-
haps he was trying to use subordination.
Regardless, using the wrong punctuation marks causes
fragments as well as run-ons. This tendency suggests that
students need to learn when to use periods, semicolons, and
commas, and for that students must be able to identify inde-
pendent and dependent clauses in sentences. This line of in-
struction leads to a review of the principles of coordination
and subordination. Remediation of both sentence structure
errors therefore begins at the same point--clause analysis
and punctuation usage--and requires students to learn several
basic grammatical terms: subject, verb, clause, independent
clause, dependent clause, conjunction, subordinator, coordina-
tion, subordination, comma, semicolon, period. Conjunctive
adverb and subordinator may be avoided as terms if the most
common of each are simply listed on a sheet of punctuation
patterns for reference.
After learning the necessary terms and the usages of
 40

punctuation marks and reviewing the principles of coordination

and subordination, students should examine correctly written
compound and complex sentences and then produce their own in
sentence combining exercises. When they have correctly pro-
duced compound and complex sentences in guided activities,
they should write a paragraph or essay in which they con-
sciously produce both types of sentences.
At this point, instructors can introduce proofreading
techniques. The grammatical pattern that emerged from the
analysis of students' run-ons suggests that students should
scan their sentences for commas; if the wora following the
comm~ is a pronoun, they should apply clause analysis to the
group of words following and preceding the comma to determine
if the sentence is a comma splice. According to the analysis
of the fragments in the sample, to check for sentence frag-
ments, students should analyze the word groups that follow
semicolons to be sure that they are independent clauses.
Then, reading backwards, since most fragments are final free
modifiers, students should analyze structures that begin with
subordinators, relative pronouns, or participles (-ing words).
Students should also read aloud when they proofread to develop
a sense of sentence completeness.
After locating run-ons and fragments in their writing,
students should correct them by applying what they have
 41

learned about punctuation, coordination, and subordination.

Students should realize, though, that the m~ans by which they
correct their errors is a rhetorical choice. Students have
at least three options for correcting a run-on sentence:
between the two independent clauses, insert (1) a comma +
conjunction, (2) a semicolon, or (J) a period to create two
separate sentences. However, students may also choose to
subordinate the second independent clause and attach it to
the first one, perhaps with a comma. To correct a sentence
fragment, the most obvious means is to change the semicolon
or period to a comma to connect what is usually a sentence
final dependent clause to its preceding base clause. But
again the method used is a rhetorical choice, and students
may also add the missing grammatical elements to make the
fragment a complete sentence. Generally, however, a fragment
can be attached to an adjoining independent clause.
By following these procedures, students should be able
to produce compound and complex sentences and identify and
correct their faulty counterparts as well. Students need to
learn sentence analysis to identify independent and dependent
clauses. Then they need to be aware of the principles of
coordination and subordination so they can recognize when a
clause should be linked or attached. Also, students need to
know the names and functions of the punctuation marks
 42

necessary for correct coordination and subordination. The ·

instruction should enable students to proofread for run-ons
and fragments in their own writing as well, for they must
finally be able to write correct sentence structures of their
own and proofread for errors in those structures in their
compositions. Therefore, the most valuable learning practice
may be having students write, proofread, and correct.
 4J

CHAPTER III. ANALYSIS OF HANDBOOKS'

TREATMENT OF RUN-ONS AND FRAGMENTS

Based on the research and the student sample, effective

remediation of run-ons and fragments should begin with lessons
on basic sentence structures. Students need to be able to
identify dependent and independent clauses in sentences be-
fore they can understand the principles of coordination and
subordination, which should be explained next. Accompanying
this instruction should be lessons on the correct uses of
punctuation marks. Then students should not only be able to
produce correct compound and complex sentences, but they
should also be able to analyze their sentences to find and
correct run-ons and fragments.
Do grammar handbooks follow this sequence of instruction
for remediating run-on sentences and sentence fragments? The
following ten handbooks used by instructors of college fresh-
man composition courses were examined to answer that question:
The Little English Handbook: Choices and Conventions; Handbook
of Current English, 6th edition; The Random House Handbook,
Jrd edition; The Heath Handbook of Composition, 10th edition;
The Little, Brown Handbook; Writing--A College Handbook;
Harbrace College Handbook, 9th edition; The Harper Handbook Q!
College Composition, 5th edition; Writing Skills Handbook; The
Macmillan Handbook of English, 7th edition. Three points of
 44

comparison were drawn from the texts' sections on run-ons and

fragments: (1) definitions of the errors, (2) rules and expla-
nations for preventing and correcting them, (3) exercises pre-
scribed for learning to correct the errors. The texts' treat-
ment of run-ons will be examined first, followed by a review
of their coverage of fragments. After that, the handbooks'
effectiveness in dealing with run-ons and fragments can be
evaluated.

Run-on Sentences
The texts, like the research, generally categorize run-on
sentences as one of two types: comma splices or fused sen-
tences. A comma splice, sometimes referred to as a comma
fault, occurs when two independent clauses are joined by only
a comma. A fused sentence consists of two independent clauses
run together without a conjunction or punctuation. The defi-
nitions of each type of run-on sentence are fairly consistent
in the handbooks, and, with one exception, the texts cover
run-ons in sections within chapters that deal with errors that
occur in compositions.
Only The Random House Handbook deals with run-ons solely
in the glossary. While the other handbooks with glossaries
define them there, too, they deal with them in more detail
in other sections. After defining comma splices and fused
sentences, the texts usually list rules by which run-ons may
 45

be avoided. Generally, they tell students not to join two

main clauses with a comma unless the comma is followed by a
coordinating conjunction. To avoid fused sentences, students
are told not to run two sentences together without an appro-
priate connecting word and/or the proper punctuation. Whether
or not the handbooks include such dictums, however, they all
explain methods by which run-ons may be corrected, usually in
context with examples that show how those methods are applied.
For instance, Writing--A College Handbook recommends as one
alternative, "put a conjunction after the comma," and immedi-
ately gives an example: "She wore huge dark glasses, so no one
recognized her." 87 In showing corrections for fused sentences,
the handbooks either refer students to the same methods used
to correct comma splices or list them again. Most texts also
caution students that the manner in which they correct run-
ens is a rhetorical choice, and to that end, one example of a
comma splice is sometimes corrected by more than one method
to show the differences.
A few texts list three ways to correct comma splices, but
most give four. The ff~ath Handbook explains six methods for
correcting comma splices, giving, as most of the other hand-
books do, accompanying examples for each. One of the common
methods, included in The Heath Handbook, suggests coordinating
the clauses with a conjunction. The Heath Handbook's example
 46

shows how this method·works: the comma splice, "We will add
another room to the house this summer, painting will have to
wait until next year." is then revised as "We will •• · this
summer, but painting . year." 88 No more explanation is
given, although some texts refer students to other sections
for more complete explanations, such as the section on con-
junctions. The other three commonly listed methods for cor-
recting comma splices are joining the independent clauses with
a semicolon, separating them into two sentences with a period,
or subordinating one of the clauses. One of The Heath Hand-
book's additional methods is to use a semicolon or a period
plus a conjunctive adverb between the two sentences. The
o.ther method explains that commas adequately join sentences
when .they are "short, closely related independent clauses in
a series," such as "The wind blew, the shutters banged, the
children trembled ... 89
While this last method admits instances when comma
splices are acceptable, most of the handbooks do not include
it. Only three others ref~r to a similar rhetorical use of
comma splices, so generally the handbooks regard run-on sen-
tences as errors to be corrected by one of four or five valid
means. The explanations of those corrective measures, however,
are often unsatisfactory. Other than stating the method and
giving an example, explanation is often not provided, as has
 47

been illustrated by The Heath tlandbook and Writing--A College

Handbook. Furthermore, any grammatical te~s used in the
explanations are often defined only in other parts of the
handbooks to which students are referred. While writing com-
positions, if students refer to a section on comma splices,
the information they need should be there, stated simply.
In its chapter on run-on sentences, the Harbrace College
Handbook not only uses grammatical terminology but refers to
complex syntactic contexts when describing comma splices.
For example, it explains that since conjunctive adverbs are
not coordinating conjunctions and can change position in the
second independent clause, a semicolon must be used between
two sentences joined by a conjunctive adverb.9° This expla-
nation seems to provide more technical information about con-
junctive adverbs than most students need for correcting run-on
sentences. A handbook such as Writing Skills Handbook that
suggests adding a "linking" or "joining word" to a comma
splice seems more suited to students with limited grammatical
knowledge. If students knew as much as Harbrace College Hand-
book requires, they probably would not be writing run-on
sentences.
Another shortcoming of the handbooks is the absence of
clear, explicit proofreading procedures for locating run-ons
in a composition. A few of the texts advise students to read
 48

their sentences aloud because different breath pauses indicate

various marks of punctuation; this is a suggestion in The
Heath Handbook. While this technique may be valid, it is not
enough. Students also need to be able to analyze sentence
structures to determine if two independent clauses are written
as one sentence; some of the texts, such as Little, Brown
Handbook and Handbook Q! Current English, express this neces-
sity. The Harper Handbook goes one step further and explains
four grammatical contexts in which comma splices occur which
match the findings from the analysis of the student sample.
If students can analyze clause structures and recognize common
grammatical structures for comma splices, they should be able
to find run-ons in their compositions. But while some texts
suggest valid proofreading techniques, they fail to emphasize
proofreading equally with the methods to correct run-ons. And
students cannot correct run-ons unless they can find them.
The exercises most textbooks provide on run-on sentences
assume that students can recognize them even though some texts
seldom explain just how that identification can be done.
Three texts have no exercises on run-ons, but in the others,
the activities are generally of the same type: identify and
correct the run-ons in lists of sentences or in paragraphs.
The handbooks therefore assume that, after reading their sec-
tions on run-ons, students will be able to recognize and
 49

correct those sentence errors, at least in prepared exercises

if not in their own compositions. The texts do explain valid,
though often grammatically complicated, methods for correcting
run-ons, but they do not consider the principles of coordi-
nation nor the grammatical requirements for punctuation that
run-on sentences violate. Without clear instruction on
coordination and the punctuation it requires, carry-over
into students' own writing of what the texts teach may be
limited.

Sentence Fragments
Similar shortcomings characterize the handbooks' treat-
ment of sentence fragments. In explaining them, the texts
use even more technical grammatical terminology. Following
the same pattern of presentation used with run-ons, the texts
define fragments, explain and give examples for correcting
them, and prescribe exercises on identifying and correcting
fragments. However, in the case of sentence fragments, the
handbooks more often name the numerous grammatical struc-
tures that become fragments and spend more time discussing
permissible uses of fragments. Overall, the handbooks'
coverage of sentence fragments is lengthier and uses more
grammatical terms in explanations but indicates more directly
how students might proofread for fragments in their compo-
sitions.
 50

The definitions of sentence fragme~ts are of three types.

Some handbooks provide lengthy, complex definitions that in-
clude grammatical terminology, such as this one from The Lit-
tle English Handbook:

a string of words, between an initial capital letter

and a period or a question mark, that lacks a sub-
ject or a finite-verb predicate (or both) or that
has a subject and a finite-verb predicate but is
made part of a larger structure by a relative pro-
noun . . . or by a subordinating conjunction.91

Most students would require subsequent definitions of finite-

verb predicate, relative pronoun, and subordinating conjunc-
tions, but even then this definition of fragments, with its
complexity, would probably not be clear. What is meant by,
"made part of a larger structure," for instance? At the other
extreme, other texts' definitions are simple but adequate;
Writing--A College Handbook defines a fragment as "a part of a
sentence punctuated as if it were a whole one ... 92 The third
type of definition, though also simple, contains one or two
grammatical terms that can easily be explained. For example,
here is the Harbrace College Handbook's definition: "A frag-
ment is a nonsentence. It is a part of a sentence--such as a
phrase or subordinate clause--written as if it were a
 51

sentence ... 93 From this definition, phrase and subordinate

clause would have to be defined, but the fact that a fragment
is "a part of' a sentence" is clear.
That there are various definitions for sentence fragments
may indicate disagreement on how thoroughly handbooks should
deal with fragments. A variety of' grammatical structures be-
come sentence fragments, but how much detail is necessary to
explain fragments to students? Presumably the texts discuss
specific fragmented grammatical structures as a proofreading
aid so that students can recognize the structures in their
writing. There are other, more direct suggestions for proof-
reading, too. For instance, the Harbrace College Handbook
suggests that fragments can also be recognized by reading
aloud and noticing voice intonation as an indicator of' termi-
nation or incompletion; however, it plainly says that being
able to recognize the structural differences between complete
sentences and dependent structures is the better way to proof-
read for f'ragments.94 The texts generally agree on this last
point, but most of' the texts also list and give examples of'
numerous grammatical structures that can frequently become
fragments.
If these lists are intended to help students identify
fragments, then lengthy explanations of' the structures would
be included for each item. But in most cases, students must
 52

refer to other sections of the handbooks for explanations.

The grammatical structures listed in The Heath Handbook as
common types of fragments are appositive phrases, preposition-
al phrases, participial phrases, infinitive phrases, and de-
pendent clauses.95 The Little, Brown Handbook and The Heath
Handbook are two other texts that have similar lists of syn-
tactic structures as fragments; usually the structure is only
named and illustrated, with no additional explanation, as in
The Heath Handbook: appositive phrase--"The crowd that at-
tended the local track meet was the usual one. Parents,
friends of the athletes, and people looking for a good tan."96
If most fragments are sentence final structures or subordinate
clauses, as the research contends, then it does not seem nec-
essary to use so much grammatical detail to explain fragments,
especially when students must refer to other sections of the
text to discover, for instance, what an appositive is.
Fortunately the texts' suggestions for correcting frag-
ments are simpler and consistent. As they did for run-ons,
the handbooks present remediation procedures with rules that
are explained, generally, by examples. Handbook Qf Current
English explains that a fragment is often part of a preceding
sentence, which may be evident if both structures are read
aloud; if the second structure is a fragment, it should be
joined to the preceding sentence with a comma: "The next
 ------------ -- -----------

53

afternoon we made our way through the harbor of Okinawa. That

island, which had made history during World War II." becomes
"The next afternoon we made our way . . . Okinawa, the island
which had . . . during World War II ... 97 Usually the texts
suggest two ways to correct a fragment: (1) attach it to an
adjacent sentence from which it has been cut off, as in the
preceding example, or (2) add the sentence part--subject or
predicate--that is missing. Some texts, like Handbook Q(

Current English, suggest a third alternative: totally rewrite

the fragment as one or more sentences when it is "involved or
hopelessly snarled ... 98 Whatever type of grammatical structure
the fragment is, the texts consistently recommend those three
methods for correcting fragments, although the texts generally
fail to note that the method used is a rhetorical choice.
If handbooks can explain basic sentence structures,
clarify subordination and the uses of punctuation, and pro-
vide practice for identifying correctly punctuated complex
sentences, students should be able to recognize and correct
fragments in their writing. Most of the texts provide exer-
cises to give students practice; usually of the same type used
with run-on sentences, the.exercises ask students to identify
and correct the fragments in lists of sentence structures and
in paragraphs. More complicated exercises provide fragments,
ask students to name their grammatical structures, and then
 54

rewrite the fragments as complete sentences. Other exercises,

which may confuse students, ask them to differentiate, in a
list of fragments, between those that are acceptable and those
that are errors. Most of the handbooks discuss permissible
sentence fragments. Generally, however, the texts caution
students that fragments are unacceptable in expository writing,
since fragments' main use is in narrative and descriptive
essays. Besides, the texts imply that students should con-
sistently demonstrate an ability to write complete sentences
before trying to use fragments effectively in rhetorically
appropriate contexts, such as dialogue.

Summary and Analysis

Overall, the handbooks' treatment of sentence fragments
is more extensive and grammatically involved than it is for
run-on sentences. The texts' discussions of both errors,
though, omit material that the research and student samples
indicate is necessary for successful remediation. The texts,
in their sections on run-ons and fragments, fail to consider
them as punctuation errors or as grammatical problems of
coordination and subordination. Other sections of the texts
that might refer to run-ons and fragments generally do not.
In only five handbooks is there direct mention of run-ons
and/or fragments in the chapters on punctuation; fragments
 55

are referred to in the sections on periods, run-ons in the

sections on commas (The Little English Handbook, The Random
House Handbook, The Harper Handbook, Writing--A College Hand-
book, Handbook of Current English). Only twice is there
mention of fragments in the sections on semicolons (The Little
English Handbook, The Random House Handbook). Furthermore,
the exercises in the texts ask students to identify and cor-
rect the run-ons and fragments in a paragraph or list of sen- ·
tence structures by using proofreading skills that have not
been provided.
The sections that deal with run-on sentences and sentence
fragments frequently refer students to various other sections
of the texts, such as those on the correct use of the period.
Furthermore, the sections on run-ons and fragments are consis-
tently isolated from the texts' discussions of coordination
and subordination. The Little English Handbook lists no
sections on either subordination or coordination in its table
of contents, and The Random House Handbook, Handbook of Cur-
rent English, The Macmillan Handbook, and Harbrace College
Handbook have no sections on coordination. The remaining five
texts--The Harper Handbook; Writing Skills Handbook; Little,
Brown Handbook; The Heath Handbook; and Writing--A College
Handbook--contain sections on both coordination and subordi-
nation. Only in Writing--A College Handbook is there mention
 56

of run-on sentences in the chapter on coordination; the chap-

ter on subordination does not include a section on fragments,
however.
These omissions reveal serious inconsistencies with the
research and the analysis of the student samples, which indi-
cate that run-ons and fragments occur when students misunder-
stand coordination and subordination and are ignorant of the
punctuation mar~s necessary to employ those principles suc-
cessfully. The handbooks need several adjustments to help
students remediate run-ons and fragments. In their sections
on each error, the texts need to explain clause analysis so
that by identifying subjects and verbs, students can recog-
nize independent and dependent sentence structures. Comple-
mentary to that instruction would be an explanation of the
correct usage of periods, commas, and semicolons with coor-
dination and subordination. After receiving this instruction,
students should be able to write and recognize correct com-
pound and complex sentences. Then, the handbooks should
explain common grammatical patterns that occur in run-ons and
fragments as proofreading tips and list, as rhetorical choices,
methods by which each error can be corrected. To be as brief
and concise as possible, the handbooks should omit most of the
grammatical terminology and concentrate chiefly on those syn-
tactic structures which are run together and fragmented in
students' actual compositions.
 57

CHAPTER IV. S~~RY AND DISCUSSION

Summary of Research Problem,

Method, and Findings
Composition instructors generally consider run-on sen-
tences and sentence fragments as errors and usually rely on
grammar handbooks for remediation. According to the research
on each incorrect sentence structure and an analysis of actual
student run-ens and fragments, do the textbooks effectively
deal with them? Does the research reflect the grammatical
structures of students' errors, and does it therefore aptly
explain them and suggest adequate remediation? To answer
those questions, first the research was reviewed. Next, stu-
dent samples of run-ens and fragments were gathered from com-
positions students wrote in a college freshman composition
course. These samples were analyzed to determine if the
research reflected student errors accurately. Finally, gram-
mar handbooks' coverage of run-ens and fragments was examined
to judge whether it deals with each problem as the research
and the samples would recommend.
The research on each emphasizes that students with in-
complete knowledge of sentence structures and punctuation will
mark sentences according to rhetorical, not grammatical, units.
Since most of the run-ons in the student sample were comma
splices, run-ons especially seem to be errors of punctuation.
 58

That assessment, though, also indicates a failure to recognize

grammatically independent sentence structures but not a fail-
ure to coordinate them. However, as run-ons were cited as
predictors of early syntactic growth, their appearance in
college compositions could indicate a student who has not
mastered the principles of coordination. Subordination, on
the other hand, is a skill students acquire at a later age
and continue to develop as they reach college age, so sentence
fragments are faulty subordination but are also evidence of
syntactic growth. Judging from the research and student
sample, though, fragments also reveal difficulty in recog~

nizing grammatically complete sentence structures and in using

appropriate punctuation to join dependent to independent gram-
matical structures.
Run-ons and fragments also occur in particular grammati-
cal contexts, according to the research and the analysis of
the student samples. Only·one study specified common struc-
tures for run-ons, whereas several named grammatical contexts
for fragments. Also, in most cases fragments were said to
occur in sentence final position. Generally, the research on
fragments was more extensive and detailed. For each, though,
the research was careful to point out rhetorical situations
in which the use of run-ons and fragments is permissible.
For neither, unfortunately, was there much mention of
 59

instructional or remedial plans to teach students to write

correct, complete sentence structures and to proofread for and
correct run-ons and fragments. When suggestions were given,
they emphasized teaching basic sentence structures, punc-
tuation usage, and methods of coordination and subordination
concurrently.
Analysis of the student samples indicated that this
approach to instruction would be successful. Generally, the
run-ons and fragments were errors of punctuation; most run-
ens were comma splices, and the fragments occurred from in-
correct use of periods and semicolons. However, the analysis
of the students' run-ons revealed patterns not supported by
the research's suggested common grammatical contexts. Analy-
sis of the st~dents' fragments, on the other hand, showed
that the grammatical patterns described in the research were
accurate, as most students' errors were sentence final depen-
dent structures. Most importantly, though, the student run-
ens and fragments seemed to occur from the causes the research
described and indicate that the resParch's recommendations for
instruction would be successful.
The handbooks' treatment of run-ons and fragments usually
did not reflect the research. Each error was treated in sec-
tions separate from those on sentence analysis, punctuation,
coordination, and subordination. Generally, the texts
 60

followed the same format in dealing with both: the errors

were defined, rules stated how students should avoid writing
run-ons and fragments, and examples showed how methods to
correct the errors worked. The coverage of run-ons and frag-
ments was often overloaded with grammatical terminology,
especially in the case of fragments, which were treated more
extensively with complicated lists of the grammatical struc-
tures that become fragments. Furthermore, the handbooks made
frequent mention of permissible uses of fragments while they
seldom mentioned permissible contexts for run-ons.

Conclusions and Implications

The grammar handbooks' sections on run-ons and fragments
do not appear to deal with either effectively, according to
the research and the analysis of the student samples. While
the texts offer acceptable explanations of how to correct
run-ons and fragments, they fail to explain why the errors
occur or how to proofread for them in compositions. In ad-
dition, the texts need to incorporate instruction on clause
analysis and punctuation into their sections on run-ons and
fragments. Those sections might also serve the students
better if they were included with the chapters on coordination
and subordination. Also, sentence combining may be a more
effective activity than the handbooks' identify-and-correct
exercises for helping students learn to deal with run-ons and
 ------ -·--·~- -~----~------

61

fragments. Since the research indicates that sentence com-

bining clarifies both linking and embedding processes and sig-
nals appropriate uses of punctuation, it may lessen the like-
lihood of run-ons and fragments occurring.
The quantity of research on both errors is not extensive
enough to allow conclusive statements about the nature or
causes of either run-ons or fragments, even though there is
more material on sentence fragments than on run-ons. There
is especially a need for a closer examination of the gram-
matical composition of run-on sentences, which in some cases
may indicate that run-ons are actually signs of incorrect
subordination rather than faulty coordination. If the re-
search is correct, college-age students should still be ac-
quiring the ability to subordinate dependent clauses. Three
observations seem to point to the possibility that some run-
ens are actually incomplete subordinate structures: (1) there
are more run-ons than fragments in the student samples from
college compositions; (2) the second independent clause of
the students' run-ons in the sample can often become a depen-
dent clause easily; (3) college students continue to subor-
dinate progressively more complicated grammatical structures,
according to the research. Furthermore, students following
the sequence of learning described by the research have
mastered coordination by college age while they are still
 62

developing their abilities to subordinate dependent grammati-

cal structures. College students, then, should not produce
run-on sentences, but since they do, perhaps some of their
run-ons are attempts to develop abilities to subordinate.
Further studies on run-ons and fragments need to investi-
gate what coordination and subordination abilities college
students possess and what grammatical knowledge they lack in
order to prescribe accurate and effective remediation proce-
dures for run-ons and fragments. The research indicates that
college students have learned coordination and are able to use
subordination, so run-ons and fragments would not seem to be
errors in applying either principle. The research and samples
suggest that each error occurs because students have no aware-
ness of complete and incomplete grammatical structures to
direct their usage of punctuation, but apparently students
also do not know the grammatical situations that require com-
mas, semicolons, or periods. Additional studies on run-ons
and fragments would enable composition teachers to provide
students with more helpful instruction.
Studies that would further focus those remedial tech-
niques need to examine thoroughly the grammatical contexts
of students' run-ons and fragments. Exposing the structural
patterns of each should reveal deficiencies in students'
knowledge of grammar. In this paper, the analysis of student
 6)

errors showed that students' use of semicolons caused frag-

ments, so instruction on the correct uses of that punctuation
mark is evidently necessary. Specifying the common grammati-
cal patterns of run-ons and fragments should additionally
provide proofreading techniques, a matter that was seldom
mentioned in the research. This paper's analysis of student
samples suggests that students might look for commas followed
by pronouns when searching for run-ons, and for semicolons
when searching for fragments. A more extensive and scientific
analysis of students' run-ons and fragments seems essential.
The findings of such an analysis would then need to be
incorporated into the handbooks' treatment ot run-ons and
fragments. Currently, as this paper points out, the handbooks'
sections on run-ons and fragments do not accurately reflect
the present research on each or what may be the common gram-
matical patterns of students' errors. The handbooks fail to
explain either as errors in sentence structure or punctuation;
neither do the texts present run-ons and fragments in connec-
tion with coordination and subordination. While the texts do
explain valid techniques for correcting the errors, they do
not present common grammatical contexts for the errors'
occurrences as proofreading helps. In fact, proofreading for
either error is largely ignored by the texts, and students
must know how to find the errors in their compositions if
 64

they are to correct them. The handbooks' remediation of run-

ens and fragments needs revision, then, for instructors and
students often depend entirely on handbooks to tell them how
to correct the errors. While instructors and students may
sense the inadequacy of the current texts' approaches, they
generally have no other references. Consequently, neither
instructors nor students learn how to understand the errors
or recognize them in compositions, and both those abilities
are essential if students are to learn not to write run-on
sentences and sentence fragments.
 65

NOTES

1 Joseph M. Williams, "The Phenomenology of Error," Col-

lege Composition and Communication, 32 (May 1981), 153·
2 John A. Higgins, "Remedial Students' Needs vs. Emphases
in Text-Workbooks," College Composition and Communication, 24
(May 1973), 189-190.
3 Higgins, pp. 189-190.
4 Kellogg W. Hunt, "Early Blooming and Late Blooming Syn-
tactic Structures," in Evaluating Writing: Describing, Mea-
suring, Judging, ed. Charles R. Cooper and Lee Odell (Urbana,
Ill.: NCTE, 1977), PP• 96-97·
Frank O'Hare, Sentence Combining: Improving Student
Writing without Formal Grammar Instruction, Research Report
no. 15 (Urbana, Ill.: NCTE, 1973), p. 24 (NCTE Stock Number
43393).
5 Hunt, pp. 92-93·
6 Hunt, pp. 97-98.
7 Hunt, pp. 98, 100.
8 Francis Christensen and Bonniejean Christensen, Notes
toward ~New Rhetoric: Nine Essays for Teachers (New York:
Harper & Row, 1978), p. 148.
9 Christensen and Christensen, p. 143.
 66

10 Christensen and Christensen, p. 143.

11 Christensen and Christensen, p. 148.
12 Mina P. Shaughnessy, Errors and Expectations (New York:
Oxford University Press, 1977) , P• 18.
13 Shaughnessy, 23·
P·
14 Shaughnessy,
P• 24.
15 Shaughnessy, pp. 5, 119·
16 Shaughnessy, P• 18.

17 Shaughnessy, pp. 22-23.

18 Philip DiStefano and Robert J. Marzano, "Basic Skills
in Composition: A New Approach," English Education, 9 (Winter
197 8)' 117.
19 DiStefano and Marzano, p. 117.
20 DiStefano and Marzano, p. 118.
21 Dona M. Kagan, "Run-On and Fragment Sentences: An
Error Analysis," Research in the Teaching of English 14 (May
1980), 134.
22 Kagan, p. 134.
23 Kagan, p. 130.
24 Kagan, pp. 134-135·
25 Shaughnessy, p. 28.
26 William F. Irmscher, Teaching Expository Writing (New
York: Holt, Rinehart and Winston, 1979), pp. 121-122.
27 Irmscher, p. 128.
28 Irmscher, pp. 121-122.
 67

29 Irmscher, p. 123·
3° Irmscher, p. 123.
31 Mohan R. Limaye, "Approaching Punctuation as a System,"
ABCA Bulletin, 46 (March 1983), 31.
Joseph M. Williams, Style: Ten Lessons in Clarity and
Grace (Glenview, Ill.: Scott, Foresman and Company, 1981),
p. 187.
32 H. Sopher, "The Problem of Punctuation," English Lan-
guage Teaching Journal, 31 (July 1977), 308.
33 Limaye, p. 31.
34 Irene Brosnahan, "A Few Good Words for the Comma
Splice," College English, 38 (October 1976), 186-187.
Sopher, p. 308.
35 Brosnah~, p. 188.
36 WJ.lliams,
. Style,. p. 64 .
Shaughnessy, P· 25·
37 Shaughnessy, P• 18.
38 Shaughnessy,
P• 24.
39 Shaughnessy, P· 25·
40 Shaughnessy, P• 25·
41 Sarah D'Eloia, "The Uses--and Limits--of Grammar," in

The Writing Teacher's Sourcebook, ed. Gary Tate and Edward P.

J. Corbett (New York: Oxford University Press, Inc., 1981),
p. 243.
42 D'Eloia, pp. 228-229.
 68

43 D'Eloia, p. 229·
44 Kagan, p. 12 8 .
45 Kagan, p. 131.
46 Kagan, pp. 131-132·
47 Kagan, p. 137·
4S Muriel Harris, "Mending the Fragmented Free Modifier,"
College Composition and Communication, 32 (May 1981), 177·
49 Harris, p. 180.
50 Harris, p. 178.
5l Harris, pp. 177-178.
52 Constance Weaver, "Welcoming Errors as Signs of
Growth," Language Arts, 59 (May 1982), 443.
53 Weaver, pp. 440-442.
54 Martin Gliserman, "An Act of Theft: Teaching Grammar,"
College English, 39 (March 1978), 794.
55 Shaughnessy, p. 68.
56 Harris, pp. 175-176.
Charles R. Kline and w. Dean Memering, "Formal Frag-
ment: The English Minor Sentence," Research in the Teaching of
English, 11 (Fall 1977), 108-109.
57 Harris, p. 176.
Kline and Memering, p. 108.
58 Harris, p. 176.
59 Harris, p. 176.
60 Kline and Memering, pp. 108-109.
 69

61 Kline and Memering, P· 98.

62 Harris, P· 176.
63 Kline and Memering, PP· 98-99.
64 D'Eloia, p. 229.
65 Irmscher,
P· 121.
66 Limaye, p. 28.
67 Irmscher, p. 120.
68 D'Eloia, pp. 239-240.
69 Shaughnessy, PP· 27, 40, 77·
70 Shaughnessy, P· 28.
71 Shaughnessy, pp. 31, 34.
72 Shaughnessy, P· 35·
73 Shaughnessy, P· JO.
74 Shaughnessy, pp. 29-30.
75 Shaughnessy, P· 30.
76 Shaughnessy, P· 78.
77 Lester Faigley, "The Influence of Generative Rhetoric
on the Syntactic Maturity and Writing Effectiveness of College
Freshmen," Research in the Teaching of English, 13 (October
1979) I 204 •
7 8 New Directions, New Connections, Proc. of the Writing
Centers Association Fifth Annual Conference, 5-6 May 1983
(West Lafayette, Ind.: Purdue University, 1983), p. 64.
79 New Directions, p. 68.
 70

80 Elaine P. Maimon and Barbara F. Nodine, "Measuring Syn-

tactic Growth: Errors and Expectations in S.entence-Combining

Practice with College Freshmen," Research in.the Teaching of
English, 12 (October 1978), 240, 24).
81 Donna Gorrell, "Controlled Composition for Basic Writ-
ers," College Composition and Communication, 32 (May 1981),
308.
82 Gorrell, pp. J08-J09.

83 Harry Shaw, The Harper Handbook of College Composition,

5th ed. (New York: Harper & Row, Publishers, Inc., 1981),
PP• 328-329.
84 Harris, p. 180.
85 Harris, p. 180.
86 Harris, p. 180.
87 James A. Heffernan and John E. Lincoln, Writing--A

College Handbook (New York: w. w. Norton & Company, Inc.,

1982), p. 257.
88 Langdon Elsbree, Nell G. Altizer and Paul V. Kelly,

The Heath Handbook of Composition, 10th ed. (Lexington, Mass.:

D. C. Heath & Company, 1981), p. 388.
89 Elsbree, Altizer and Kelly, p. 390.
90 John c. Hodges and Mary E. Whitten, Harbrace College
Handbook, 9th ed. (New York: Harcourt Brace Jovanovich, Inc.,
1982 ), p . 46.
 ~~-~---- --~--~ --- --

71

9 1 Edward P. J. Corbett, The Little English Handbook:

Choices and Conventions, 4th ed. (Glenview', Ill.: Scott, Fores-
man and Company, 1984), p. 35·
9 2 Heffernan and Lincoln, p. 296.
93 Hodges and Whitten, p. 32.
94 Hodges and Whitten, p. 32·
95 Elsbree, Altizer and Kelly, p. 382.
96 Elsbree, Altizer and Kelly, p. 383.
97 Jim w. Corder, Handbook Qt Current English, 6th ed.,
(Glenview, Ill.: Scott, Foresman and Company, 1981), p. 42.
9 8 Corder, p. 43.
 72

BIBLIOGRAPHY

Brosnahan, Irene. "A Few Good Words for the Comma Splice."
College English, 38 (October 1976), 184-188.
Christensen, Francis, and Bonniejean Christensen. Notes to-
ward ~New Rhetoric: Nine Essays for Teachers. New York:
Harper & Row, 1978.
Corbett, Edward P. J. The Little English Handbook: Choices
and Conventions. 4th ed. Glenview, Ill.: Scott, Fores-
man and Company, 1984.
Corder, Jim w. Handbook of Current English. 6th ed. Glen-
view, Ill.: Scott, Foresman and Company, 1981.
Crews, Frederick. The Random House Handbook. Jrd ed. New
York: Random House, Inc., 1980.
D'Eloia, Sarah. "The Uses--and Limits--of Grammar." In The
Writing Teacher's Sourcebook. Ed. Gary Tate, and Edward
P. J. Corbett. New York: Oxford University Press, Inc.,
1981, PP• 225-24).
DiStefano, Philip, and Robert J. Marzano. "Basic Skills in
Composition: A New Approach." English Education, 9
(Winter 1978), 117-121.
Elsbree, Langdon, Nell G. Altizer, and Paul v. Kelly. Th~

Heath Handbook of Composition. lOth ed. Lexington,

 73

Mass.: D. C. Heath & Company, 1981.

Faigley, Lester. "The Influence of Generative Rhetoric on the
Syntactic Maturity and Writing Effectiveness of College
Freshmen." Research in the Teaching of English, 13
(October 1979), 197-206.
Fowler, H. Ramsey. The Little, Brown Handbook. Boston: Lit-
tle, Brown and Company, 1983.
Gliserman, Martin. "An Act of Theft: Teaching Grammar." Col-
lege English, 39 (March 1978), 791-799·
Gorrell, Donna. "Controlled Composition for Basic Writers."
College Composition and Communication, 32 (May 1981),
308-316.
Harris, Muriel. "Mending the Fragmented Free Modifier." Col-
lege Composition and Communication, 32 (May 1981),
175-182.
Heffernan, James A., and John E. Lincoln. Writing--A College
Handbook. New York: W. W. Norton & Company, Inc., 1982.
Higgins, John A. "Remedial Students' Needs vs. Emphases in
Text-Workbooks." College Composition and Communication,
24 (May 1973), 188-192.
Hodges, John c., and Mary E. Whitten. Harbrace College Hand-
book. 9th ed. New York: Harcourt Brace Jovanovich,
Inc., 1982.
Hunt, Kellogg w. "Early Blooming and Late Blooming Syntactic
Structures." In Evaluating Writing: Describing,
 74

Measuring, Judging. Ed. Charles R. Cooper, and Lee

Odell. Urbana, Ill.: NCTE, 1977, pp. 91-104.
Irmscher, William F. Teaching Expository.Writing. New York:
Holt, Rinehart and Winston, 1979.
Kagan, Dona M. "Run-On and Fragment Sentences: An Error Anal-
ysis." Research in the Teaching of English, 14 (May
1980), 127-138.
Kline, Charles R. , and W. Dean Memering. "Formal Fragment:
The English Minor Sentence." Research in the Teaching
of English, 11 (Fall 1977), 97-110.
Limaye, Mohan R. "Approaching Punctuation as a System." ABCA
Bulletin, 46 (March 1983), 28-33·
Maimon, Elaine P., and Barbara F. Nodine. "Measuring Syntac-
tic Growth: Errors and Expectations in Sentence-Combining
Practice with College Freshmen." Research in the Teach-
ing of English, 12 (October 1978), 233-244.
New Directions, New Connections. Proc. of the Writing Centers
Association Fifth Annual Conference. 5-6 May 1983. West
Lafayette, Ind.: Purdue University, 1983.
O'Hare, Frank. Sentence Combining: Improving Student Writing
without Formal Grammar Instruction, Research Report no.
15. Urbana, Ill.: NCTE, 1973·
Shaughnessy, Mina P. Errors and Expectations. New York: Ox-
ford University Press, 1977·
 75

Shaw, Harry. The Harper Handbook of College Composition. 5th

ed. New York: Harper & Row, Publishers, Inc., 1981.
Sopher, H. "The Problem of Punctuation." English Language
Teaching Journal, 31 (July 1977), 304-313·
Weaver, Constance. "Welcoming Errors as Signs of Growth."
Language Arts, 59 (May 1982), 438-444.
Wiener, Harvey s., and Charles Bazerman. Writing Skills Hand-
book. Boston: Houghton Mifflin Company, 1983.
Williams, Joseph M. "The Phenomenology of Error." College
Composition and Communication, 32 (May 1981), 152-168.
----------. Style: Ten Lessons in Clarity and Grace. Glen-
view, Ill.: Scott, Foresman and Company, 1981.
Willson, Jr., Robert F., John M. Kierzek, and w. Walker Gibson.
The Macmillan Handbook of English. 7th ed. New York:
Macmillan Publishing Co., Inc., 1982.
 76

APPENDIX A. RUN -ON SENTENCES: STUDENT

SAMPLE WITH GRAMMATICAL DESCRIPTIONS

short-sentence+ long sentence

1. We decided Deb would ride first, he didn't like the idea
but he was out-voted two to one.
comma splice 1st sentence: complex, noun clause
2nd sentence: compound, subject is personal pronoun, head
of clause, antecedent in first sentence
2. I lay in bed for awhile, then my mom came in to wake up
my younger brother and myself.

comma splice 1st sentence: simple

2nd sentence: simple, begins with conjunctive adverb
J. The basement floor of Macy's is extraordinary, called the
Cellar it contains a restaurant, bakery, candy store, and
delicatessen.
comma splice 1st sentence: simple
2nd sentence: simple, subject is personal pronoun, ante-
cedent in first sentence, participial phrase begins the
clause

long sentence + short sentence

4. I would work the horses from the ground for the first two
or three weeks, this was to gentle them down some.
comma splice 1st sentence: simple
2nd sentence: simple, subject is demonstrative pronoun,
head of clause, antecedent in first clause
5. Then Jim would work the horse to try to teach him some
manners, this was all until the Loco Mare.
comma splice 1st sentence: simple
2nd sentence: simple, subject is demonstrative pronoun,
head of clause, antecedent in first clause
 77

6. Once in awhile I will bring back the wonderful memories

and shed a tear over his loss, it was a great one to me.
comma splice 1st sentence: simple, compound verb
2nd sentence: simple, subject is personal pronoun, head
of clause, antecedent in first clause
7• It seemed to [si~ good to be true, it was.
comma splice 1st sentence: simple
2nd sentence: simple, subject is personal pronoun, head of
clause, antecedent in first clause (same pronoun)
8. She acted perfect in the ring, no baulking or fussing she
didn't even seem nervous.
run together 1st sentence: simple
2nd sentence: simple, subject is personal pronoun, head of
clause, antecedent in first clause (same pronoun)
9· I had to help pick them up for the guy as fast as I could,
but to no avail I was still late for my class.
run together 1st sentence: complex, adverb clause
2nd sentence: simple, subject is personal pronoun, head of
clause, antecedent in first clause (same pronoun)
10. When I hit the ground she decided not t~ turn and run, she
was ready for a fight.
comma splice 1st sentence: complex, adverb clause
2nd sentence: simple, subject is personal pronoun, head of
clause, antecedent in first clause (same pronoun)
11. As I walked down the hall and saw the others stiffly move
from place to place all I could feel was a little bit
happy, after all, we won.
comma splice 1st sentence: complex, adverb clause,
adjective clause
2nd sentence: simple, subject is personal pronoun
12. She was still there in the morning, which cost me five
dollars, I'd bet against the stall.
comma splice 1st sentence: complex, adjective clause
 78

2nd sentences simple, subject is personal pronoun, head

of clause, antecedent in first clause (pronoun~)

sentences of similar length

13. That exam earned me more than just an "A" for the course,
it taught me to respect and have confidence in my abil-
ity.
comma splice 1st sentences simple
2nd sentences simple, subject is personal pronoun, head
of clause, antecedent in first clause
14. I looked up in time to see a yellow blur moving quickly
towards me, I barely had time to hit the brakes and stop
my car.
comma splice 1st sentence: simple
2nd sentence: simple, subject is personal pronoun, head
of clause, antecedent in first clause (same pronoun)
15. My chance finally carne to me, we moved out of town.
comma splice 1st sentence: simple
2nd sentence: simple, subject is personal pronoun, head
of clause
16. We kept most of the horses at my place, I had the nicest
barn, and a bigger refrigerator.
comma splice 1st sentence: simple
2nd sentence: simple, subject is personal pronoun, head
of clause
17. You're to [sic] young, we live in town.
comma splice 1st sentence: simple
2nd sentence: simple, subject is personal pronoun, head
of clause
18. But they put us to bed for our afternoon nap, then we
'heard them pull out of the drive.
comma splice 1st sentence: simple
2nd sentence: simple, subject is personal pronoun, ante-
cedent in first clause (pronoun~), conjunctive adverb
is head of clause
 79

19. God, was I ever happy, my dream had finally come true.
comma splice 1st sentence: simpl~
2nd sentence: simple
20. I wish he were here now, I loved him then and now.
comma splice 1st sentence: complex, noun clause
2nd sentence: simple, subject is personal pronoun, head
of clause, antecedent in first clause (same pronoun)
21. I thought we had it made she just stood there shaking.
run together 1st sentence: complex, noun clause
2nd sentence: simple, subject is personal pronoun, head
of clause

Summary of grammatical description of student sample run-on

sentences
first sentence second sentence
long: 9 long: J
short: J short: 9
same: 9 same: 9
simple: 14 simple: 20
compound: 0 compound: 1
complex: 7 complex: 0
subject:
personal pronoun: 17
demonstrative pronoun: 2
antecedent in
1st sentence: 14
subject head of clause: 16
conjunctive adverb
head of clause: 2
 ----------

80

APPENDIX B. SENTENCE FRAGMENTS: STUDENT

SAMPLE WITH GRAMMATICAL DESCRIPTIONS

semicolon used incorrectly

1. Another nurse asked me to lay on a table; positioning me
so that I was extremely uncomfortable and told me not to
move.
verbal phrase (second half of compound verb preceded by
participle + adverb clause)
2. He was traveling so fast; however, that his car was able
to turn our pick-up truck one hundred and eighty degrees
around.
adverb clause, however incorrectly used with semicolon as
conjunctive adverb
J. In an attempt to break my nervousness, I decided to leave
early for school; hoping the fresh air and sunshine would
help.
final free modifier: verb phrase (participle + noun
clause)
4. I wasn't too sure about which was worse; the concussion
or all the radiation I was absorbing.
final free modifier: noun phrase with compound nouns +
adjective clause

period used incorrectly

5. My adventure began when Neal, Phil and I set out for our
first attempt of this bright and blue day to experience
the sport of water skiing. A sport which requires both
skill and endurance.
final free modifier: noun phrase (noun + adjective
clause)
6. The trick is to keep the skiis pointed straight in front
of me and the rope between the skiis. Neither of which
is an easy task.
final free modifier: pronoun + adjective clause
 81

7. Once informed, I began first by putting on my life pre-

server. Then my skiis, which were like being fastened to
the floor.
second part of compound structure (noUn + adjective
clause)
8. After straightening myself in the mens [sic] room, where
I received a strange look from a guy seeing me change. I
go to class and am about five minutes late.
sentence initial verbal phrase (participial) +adjective
clause