Grammar and Literacy: 
Embedding Outside Sources in Text

Jim Kenkel (Eastern Kentucky) & Robert Yates (Central Missouri)

A paper presented at the
Sixth Annual Conference of the NCTE Assembly for the Teaching of English Grammar,
July 27 & 28, 1995.
Pennsylvania College of Technology, Williamsport, PA. 

The Problem

In the last ten years or so, work by DeBeaugrande (1984), Noguchi (1992), and Kenkel and Yates (1992) has argued that native speakers can operationalize their tacit grammatical knowledge and apply it in an effort to resolve persistent errors in written usage, such as fragments, run-ons, and subject-verb agreement. Kolln (1991) demonstrates that this tacit knowledge or "sentence sense" can be exploited by students as part of strategies to improve written style. In this paper, however, we discuss a persistent language problem in freshman writing for which our students have no pertinent sentence sense, namely, the embedding of outside sources into text -- a task for which they lack the needed language knowledge and skill. The rhetorical strategies and language structures involved in this task are relatively rare in spoken language; indeed, their presence usually indicates high levels of literacy reflecting either extensive reading or explicit instruction.

It is natural, then, that our students struggle with incorporating outside material in their text as they will have had little relevant exposure in their reading nor will they have had occasion to use outside materials in much of their writing.

The rhetorical side of the issue has already been the focus of research. Kantz (1990), for example, in an excellent discussion of student use of outside sources, describes the reading and writing strategies of the typical college underclassman enrolled in a writing from sources course. She suggests that this typical student is proficient at reading and writing narrative. Moreover, she believes that this typical student controls an impressive range of study and writing skills including reading for the main idea, choosing relevant source material for her work, and skilled summarizing and paraphrasing. However, such students cannot overcome the difficulty they experience in writing papers based on outside sources. These students, according to Kantz, do not establish enough rhetorical distance from their sources. They tend to confuse a narration of their research reading with a genuine argument. In so doing, they don’t differentiate the rhetorical stance of their sources from their own purpose. These students do not analyze their source material to find information which they can use to support their own point of view. Kantz suggests a hierarchy of the writing difficulties which these students confront.

In this view, the typical student sees outside material as a mass of information to be directly incorporated in the student’s writing. It is to be expected, then, that these papers resemble reports more than arguments.

Kantz stresses that more incisive, purposeful reading strategies are needed to resolve these writing problems. We agree, but feel that increased rhetorical awareness, while necessary, is not sufficient. In our samples of student writing, we have texts which reveal that the writer is certainly aware of (what we call) her rhetorical obligation, but is unable to meet it because she lacks control of the lexical items and syntactic structure necessary for integrating material from outside sources into her text.

We repeat that it is not surprising that the typical underclassman would encounter difficulties with these structures peculiar to higher forms of literacy. Perera (1986), in a summary of the language difficulties encountered by children as they learn to write, identifies vocabulary and sentence structures which school children do not control. Perera believes that this lack of control reveals that children do not recognize that writing is different from speaking. Perrera observes that structures which give children difficulty include certain adverbial clauses, ellipsis, and what she calls heavy subject NPs, in other words, NPs that function as subjects that have modification or are dependent clauses. These structures rarely occur in the spoken language. Rather, they occur almost exclusively in the written language. Perera speculates that only children who have been exposed to a lot of reading use such forms in their own writing. Similarly, typical first year college students have difficulties with the language needed to embed outside sources into text because they have not been exposed to large amounts of academic reading.

If we are correct that only students with extensive reading experience control advanced literacy structures and if we are correct in assuming that many of our students have not had the relevant academic reading experience, then teachers are dealing with more than merely recognizing differences between spoken and written language. In fact, we are claiming that typical students have a deficit in their knowledge of the relevant grammatical structures and that this deficit needs to be met by appropriate pedagogical intervention.

Of course, as Kantz’ work makes clear, it would be simplistic to suggest that student problems in integrating outside texts into their writing are strictly grammatical. The problem has interdependent grammatical and rhetorical elements. However, although rhetorical and language knowledge is interdependent in texts, it is useful to analyze the rhetorical and language knowledge needed to integrate cited material skillfully into text. From a rhetorical perspective, writers need to distinguish between the perspective and purpose of the cited source and their own. Lack of this recognition is reflected in student writing where outside material is directly incorporated into the text in a (quasi) plagiaristic fashion and in writing where cited material has no prefatory or post comment. However, this rhetorical awareness is not sufficient for producing appropriate text. To integrate outside material into their texts successfully, students must not only recognize the rhetorical obligation to distinguish between outside text and their own, but they also must control the appropriate grammatical structures. These include control of specific lexical items, namely, signal verbs such as "state," "argue," "emphasize," and "conclude." Of course, use of inappropriate signal verbs frequently co-occurs with errors in direct and indirect speech, which include problems with reference relations, the grammar of complementizers, and tense choice. Finally, writers must control the grammar associated with prepositional phrases such as "according to X" and "In X’s book," which are used to topicalize information. Often associated with student use of these topicalized structures are errors in subject grammar, relative clauses, and reference relations between a topicalized antecedent and its anaphor.

We emphasize that the rhetorical and language problems referred to above are non-separable in practice. Students are unlikely to develop the needed language skills if they are unaware of the rhetorical necessity of distinguishing between their texts and that of the cited material. However, we believe that recognition of the rhetorical imperative is not sufficient in itself to solve the students’ writing problem. Indeed, students who understand the rhetorical issue but who lack control of the corresponding language structures will still write non-felicitous texts. In sum, our general contention is that development of advanced literacy skills in our students involves a linguistic as well as a rhetorical component.

In the remainder of this paper, in order to understand and better respond to the complex language challenge confronting our students, we first situate student writers within a two stage developmental sequence for embedding outside sources into text. Second, we analyze examples of student writing, illustrating how types of error reflect students’ stages of knowledge and skill development. Third, we discuss possible pedagogical responses to these problems by critically reviewing several handbooks and discussing the extent to which they provide useful guidelines for integrating outside sources. Finally, we suggest a straightforward pedagogy to supplement handbooks.

DISCUSSION

We surveyed approximately 50 papers drawn from student populations from Eastern Kentucky University and Central Missouri State University enrolled in writing from sources courses. Through the significant difficulties that these writers experienced with embedding outside sources into their texts, we discerned a two stage developmental sequence reflecting students’ growing rhetorical awareness and the developing language knowledge and skill to act on that awareness. Stage one describes a stage where the writer does not recognize the rhetorical problem, namely, the need to distinguish the perspective and purpose of the source text from the perspective and purpose of the writer. This stage itself can be characterized by two writing problems. Problem A refers to student writing revealing that the student makes no distinction between her text and the source; in other words, Problem A describes outright plagiarism. In these texts, the student indicates no source, uses no signal expressions, uses no quotation marks, and includes no prefatory or post comment. Problem B of Stage One refers to texts where the writer uses quotation marks or cites the source but is still insufficiently aware of the need to distinguish between her text and the source through use of signal expressions and prefatory and post comments. We refer to such citations as "parachuted." Problem B of Stage One, then, describes a student who is beginning to develop the needed rhetorical awareness but who does not yet have sufficient knowledge and confidence in the lexical and syntactic resources needed to show the rhetorical relation.

Stage Two describes a student who has acknowledged her rhetorical obligations but who nonetheless fails to meet them. Generally speaking, the difference between a Stage One student and a Stage Two student is that in Stage Two, the student attempts to fulfill her rhetorical obligations but does so inappropriately. This stage includes four problems pertinent to embedding outside sources typical of novice writers: use of inappropriate prefatory and post comments, use of inappropriate signal words, lack of control of the grammar and conventions of direct and indirect speech, and lack of control of the grammar of topicalization.

Inappropriate prefatory or post comments are typical of students who understand the need to avoid parachuted citations but lack either an adequate understanding of their topic or sufficient control of the content and functional vocabulary needed to develop their discussion clearly. The obstacles presented by a limited knowledge base, of course, are not peculiar to the problem of integrating outside material. However, the use of appropriate signal words to introduce citations certainly is. At an earlier period in Level 2, the student will tend to draw on words common to the spoken language such as "say," "tell," and "goes." Later, she uses signal words more appropriate to writing such as "state," "oppose," and "conclude," but still may have difficulty with the grammar and conventions of direct and indirect speech requiring her to decide whether or not to use a complementizer, to choose the appropriate tense, and to control punctuation conventions. The grammar of topicalization also presents considerable problems to student writers. Use of topicalizing phrases such as "according to X" raises problems of reference and subject grammar.

The above brief description of student difficulties indicates that we are dealing with issues of literacy for which our students’ language sense cannot help them. They do not have the needed linguistic knowledge to deal with them. Therefore, a responsibility of writing from sources courses must be to offer explicit instruction on both the rhetorical and language issues involved in embedding outside sources into text. In general, the problems which student writers have in skillfully integrating outside material into their own discussions arise from their lack of familiarity with the written language. For example, the conventions of spoken discourse require neither prefatory or post comments; a small number of signal words suffices for ordinary instances of direct or indirect speech as the ordinarily rich spoken discourse context will provide the listener with information to make appropriate inferences; topicalization presents few problems to speakers and hearers as fragments are appropriate in speech and reference is more easily determined within the context of the speech interaction.

DATA ANALYSIS

The student difficulties mentioned above are more clearly illustrated by the examples presented below drawn from student writing.1 For ease of reference, we summarize the two points in the developmental process and the problems that characterize each.

Proposed Developmental Sequence for Embedding Outside Sources

Table 1

The skill level that we designate as Level 1 reflects a lack of recognition of the writer’s rhetorical obligation to her readers. We perceive two typical problems in this level. One is outright plagiarism; in other words, the writer makes no distinction between the source text and her own, as illustrated in (1) below.

What marks this example as Level 1 is that in spite of the fact that there are only two citations and no quotations marks, three of the seven sentences are exact copies, three are near exact copies, and one, the last , is so badly copied that its intended meaning is obscured. In this text, there is certainly no distinction between the stance of the source and that of the writer.

The other problem typical of Level 1 is the lack of signaling phrases to mark a distinction between the source text and the writer’s text. Another feature of this lack of rhetorical awareness is the absence of any prefatory or follow up comment to the citation. We call such citations "parachuted citations." In both (2) and (3) below, the writer signals the reader that the information comes from an outside source, but does not establish a relation between that source and her own text. It is as if the citation were "parachuted" into the text, with no explicit indication to the reader as to how it furthers the writer’s purpose.

There are writers who produce partial parachuted citations as in (4) below, where there is an introduction to the citation but no follow-up comment. The source of Level 1 writing problems is the lack of student awareness of her rhetorical need to differentiate the purpose of the source text from her own. In so doing, the writer will foreground her own argumentation while keeping the argumentative direction of her source in the background. We believe that there are student writers who fully understand this imperative but are still unable to write felicitous text due primarily to deficits in their linguistic knowledge and skill. In fact, of the four types of Level 2 problems that we identify, three are linguistic in nature. We feel that students who are unaware of the rhetorical issues are likely not to notice their linguistic difficulties, which arise from student confusion of speech and writing.

The four problems typical of Level 2 all arise from a lack of experience with academic reading and writing. Students tend to rely on signal words appropriate to speech, for instance, "say" or "goes" instead of "argue" or "contend." Frequently associated with use of inappropriate signal verbs is confusion between direct and indirect speech. Students often don’t know the difference between direct and indirect speech. Their confusion is a result of the optionality of the COMP position in sentences such as the following.

In speech, (5) is ambiguous, having the interpretation of both (5) and (7). It is the presence of COMP that renders (5) unambiguous. On the basis of sentences such as (8) and (9) below, students tend to think that the presence or absence of COMP does not affect meaning. Use of inappropriate prefatory or post comments comes from student extension of the rich context of spoken interactions to the more constrained contexts common to written communication. The lack of control of the grammar of topicalization also can be attributed to the overgeneralization of the norms of speech to writing. For example, topicalized structures such as "according to Newt, he believes" or "Newt, he believes" are common in speech and hearers do not have difficulty in determining the intended reference of the pronoun "he" because of the richness of the immediate context. However, in writing, the reference is intolerably ambiguous.

Through the examples drawn from student writing that we present below, we support our contention that even students who understand the rhetorical imperatives involved in writing from sources can still be unable to write acceptable texts due to deficits in linguistic knowledge and skill.

Examples (10) (11) and (12) illustrate Problems 2A and 2B. These show that students overgeneralize the conventions of speech into writing by using spoken language signal verbs and lack awareness of the conventions of direct and indirect speech.

Example (13) given below is interesting because even though the writer’s syntactic control is deficient, her writing is more sophisticated than that in (10), (11), and (12). This writer clearly understands function of outside sources. She uses them as support for the argument she is pursuing rather than as foregrounded material. Unfortunately, her citation is pulled in two directions by the signal expressions she employs. The fact that her language skill has lagged behind her rhetorical awareness is not cause for dismay. Instead, we feel that she is the type of student who could accelerate her development with direct instruction.

A third type of problem typical of students in Level 2 is the use of inappropriate prefatory or post comments to their citations. This type of problem can have as its cause either a non-rhetorical reading strategy (cf. Kantz 1991) or the assumption of the presence of a rich communicative context typical of speech situations, both of which result in what Flower (1979) calls "writer-based prose." In example (14) below, although the student skillfully introduces her citation, she has used the inappropriate strategy of following it up by tacking on yet another citation. Even though she recognizes her rhetorical obligation to follow up her citation, she does not realize that using a second citation is an inappropriate means of doing so. By merely sequencing citations, she has not privileged her own argumentative purpose. She has wrongly assumed that a research paper is but series of citations, proving to her audience that she read her sources. From the alternative perspective, the use of this relatively incoherent post comment reveals that the writer believes that a rich communicative context of shared assumptions typical of speech is in place, allowing her to evaluate her post-citation comment as coherent.

Probably the most persistent of the student writing problems typical of Level 2 are those associated with the grammar of topicalization, in particular errors made with structures such as "in X’s book, he" and "according to X, she states . . ." These structures are not as simple as they seem, and, in using them, student writers make errors of reference and of subject grammar.

Examples (15), (16) and (17) contain errors certainly familiar to writing teachers. In both, the writers have established distance between their own text and that of the source by topicalizing the source text, but have then mistakenly assumed that the pronoun non-ambiguously refers to the source author.

Similar problems occur with the more complex task of citing a source which cites another source, as in (18). In this case, the student writer topicalizes the main, or matrix, source but fails to signal the relation between her source, Bigelow, and the source cited by Bigelow, viz, Koning. As in (16) and (17), she is referring not to a particular author but rather to the topicalized information.

In addition to reference problems, student writers also confuse topicalized information with the grammatical subject as in (19), (20), and (21).

The confusion between topic and grammatical subject is also apparent in (22), where the writer misanalyzes the topicalized information as the head noun of a relative clause. Also frequently associated with the confusion of topic and grammatical subject are fragments, as in (23) and (24). In (23), the writer, unsure of how to integrate the topicalized information into the matrix sentence, tries to solve the problem by starting a new sentence to make her assertion. The writer of (24), more sophisticated, skillfully passivizes the verb in the topicalized clause in order to establish an appropriate head noun for the following clause. However, she stumbles over the grammar of relative clauses.

PEDAGOGICAL RESPONSE

The above examples provide abundant evidence that many student writers lack the necessary grammatical knowledge and skill to integrate outside sources into their text. However, it is interesting to note that college handbooks often required by freshmen writing courses provide little help with this problem. None of the four texts we consulted (The Allyn & Bacon Handbook (1994), The St. Martin’s Handbook 3rd ed. (1995), The Holt Handbook 4th ed. (1995) and the Simon and Schuster Handbook for Writers (1993)) discuss the syntax of "according to" or other kinds of topicalizing form such as "in the book/article/etc." All four texts advise students to integrate their sources "smoothly" into their writing. Usually this advice is accompanied by examples. The best discussion of how integration is done is in The St. Martin’s Handbook. For example, Lunsford and Connors note, "Introducing a quotation with the author’s name and a signal verb is clear and simple way of integrating the quotation into your text." By signal verbs, Lunsford and Connors mean such words as suggest, claim disagree, etc. The Allyn and Bacon Handbook contains a similar list.

The only text which offers any kind of exercise which has passages with problems in integrating sources is Handbook for Writers. The exercise has passages with a parachuted quotation, tense shifts, and an inaccurate quotation among others. The exercise has the following directions.

There are two problems in this exercise. The first problem is an example of a parachuted quotation while the second is a problem of tense. Both of these problems at least reflect the kind of difficulties we have found in our students’s writing.

Our explanation for the problems students have is that students have not had enough experience with the written language. Although this exemplary exercise has problems we have found, it still does not provide students with a grounding in the rhetorical problems they face in using written sources in their writing and, just as importantly, the necessary GRAMMATICAL forms that are available for smoothly integrating other texts into their writing.

Our pedagogical suggestions are grounded in this explanation. First, we believe that students need to become aware of how written sources are integrated into published sources. Following Campbell (1990), we strongly recommend that students must read articles not only for their information but for examples how the rhetorical function of cited sources and the grammatical forms used to integrate such sources into the text. Second, we like the lists of "attributive" or "signal" words in The St. Martin’s Handbook and The Allyn and Bacon Handbook. Finally, we believe that students need to practice solving the grammatical issues posed by integrating sources into their text. Ideally, these exercises should be based on the students’s writing. Because of the impracticality of such exercises we recommend the kind of exercise cited above.


Note
1. Some of the samples are drawn from student responses to the material on Christopher Columbus and Cinderella found in Behrens and Rosen (1994).

References

Behrens, L., and Rosen, L. (l 994). Writing and reading across the curriculum, Fifth Edition. New York: HarperCollins.

Campbell, C. (1990). Writing with other’s words. In B. Kroll (Ed.), Second language writing: Research insights for the classroom (pp. 21l - 230). Cambridge: Cambridge University Press.

DeBeaugrande, R. (1984). Forward to the basics: Getting down to grammar. College Composition and Communication, 35, 358-67.

Flower, L. (1979). "Writer-based prose: A cognitive basis for problems in writing." College English, 41, 19-37.

Kantz, M. (1990). Helping students use textual sources persuasively. College English, 52, 74-91.

Kenkel, J. & Yates, R. (1992). Grammatical competence and the teaching of grammar. In Ed Vavra (ed.), Proceedings of the 3rd annual conference of the Association for the Teaching of English Grammar (pp. 16-25). Williamsport, PA: Association for the Teaching of English Grammar.

Kolln, M. (1991). Rhetorical grammar: Grammatical choices, rhetorical effects. New York: MacMillan.

Noguchi, R. (1991). Grammar and the teaching of writing: Limits and possibilities. Urbana: NCTE.

Perera, K. (1986). "Language acquisition and writing." In P. Fletcher and M. Garnan (Eds.), Language Acquisition. 2nd Ed. (pp. 494-518). Cambridge: Cambridge University Press.