Composition Forum 31, Spring 2015
http://compositionforum.com/issue/31/

Linguistic Attention in Rhetorical Genre Studies and First Year Writing

Laura Aull

Abstract: Since Carolyn Miller’s Genre as Social Action, North American Rhetorical Genre Studies (RGS) has facilitated analysis of how typified rhetorical actions constitute the contexts and communities in which writers write. In first-year writing (FYW) specifically, RGS approaches have focused on macro-level textual constructs, like the audience and evidence expectations of different genres, and have bolstered valuable attention to genre awareness and transfer. In its attention to context and macro-level features, however, RGS has focused less on recurring linguistic patterns in written genres, which has contributed to two gaps in genre-based approaches to FYW: few large-scale analyses of first-year written genres, and little attention to language patterns in genre-based FYW pedagogy and research. This article aims to interrogate these gaps and offer a way beyond them, in three parts. First, it historicizes the institutional separation of U.S. rhetoric-composition and linguistics. Second, it outlines recent pedagogical genre research in RGS and English for Academic Purposes (EAP), which together offer valuable insights for approaches to FYW. Finally, it delineates selected observations from a context-informed corpus linguistic analysis of 19,463 FYW argumentative essays that draws on both RGS and EAP genre traditions. The analysis highlights rhetorical cues of the essay prompts (often absent in EAP corpus linguistic research) alongside shared linguistic patterns (often absent in RGS studies). The analysis likewise outlines overall patterns that distinguish FYW from published academic writing. The article closes with implications for pedagogy, research, and assessment.

Introduction: Rhetorical Genre Studies and the Emphasis on Whole Texts and Social Contexts

Thirty years of rhetorical and linguistic genre studies have transformed how we understand academic writing. They have cast academic genres not as value-neutral conduits for information but as typified rhetorical actions that constitute the often-tacit expectations of academic contexts and communities. Applications of rhetorical genre studies (RGS) to first-year writing (FYW) have specifically expanded discussions about FYW contexts, awareness, and transfer, with particular implications for students who are new to university genres.

At the start of RGS was Carolyn Miller’s Genre as Social Action, which theorizes genre not only as a “pattern of forms” (165) but as sociorhetorical realizations of cultural and context-specific demands. Miller’s emphasis on macro-level meaning-making aimed to expand genre conceptions focused on formal features (Bawarshi The Genre Function), and that emphasis was in keeping with historic differences between linguistic and rhetoric-composition studies.{1} At the same time, macro-level constructs are not the exclusive focus of Genre as Social Action. Following her review of Halliday, Miller frames her consideration of exigency in a way that casts rhetorical situation as both a linguistic and non-linguistic sign system: “If rhetorical situation is not material and objective, but a social construct, or semiotic structure, how are we to understand exigence, which is at the core of situation?” (C. Miller 157; emphasis mine). Miller also underscores meso-level rhetorical actions, or the “intermediate forms or strategies” (C. Miller 161) between the sentence and the whole text. And her insistence that “a genre is not just a pattern of forms” (165) does not preclude patterns of form but, rather, does not limit genres to them.

What has been most widely adopted in RGS-based approaches to FYW (which have a pedagogical impetus unlike Miller’s article but have been heavily influenced by it) has been the conceptualization of genre vis-à-vis whole-text enactments in contexts rather than patterns of form across contexts. It follows that most RGS-based approaches to FYW today emphasize constructs like discourse community and purpose as well as context-rich methods like ethnography and qualitative analysis (Bawarshi Invention; Downs and Wardle; Beaufort).{2} As Gere et al. recently argued, meso-level forms and “micro-level linguistic resources” have received much less attention in RGS to date (Gere, Aull, Lancaster, et al. 612). RGS has accordingly offered valuable ways of thinking about the contexts and social actions of FYW, but it has focused less on patterns of form that contribute to meaning-making in FYW genres. This RGS emphasis is consistent with the “erasure of the sentence” (Connors The Erasure of the Sentence 122) and “erasure of language” (MacDonald) noted in rhetoric-composition studies.

In the time that RGS has developed, so too have genre studies in applied linguistic English for Academic Purposes (EAP), which are less visible in North American composition studies and primarily focus on English language learners (ELLs). In contrast with RGS, EAP genre studies tend to focus on discourse features across contexts and individuals, examining how aggregate patterns reflect and constitute particular university and disciplinary genres (e.g., Swales Genre Analysis; Hyland Disciplinary Discourses). EAP genre studies have therefore elucidated linguistic patterns in levels and types of academic writing but have focused less on how such patterns relate to sociorhetorical details of particular writing tasks. Pedagogical RGS and EAP genre studies are not wholly separate—both emphasize the theoretical importance of both social context and discourse patterns—but the methods and conceptualizations in RGS have been “contextually-grounded,” while those in EAP have been “linguistic” (Dudley-Evans). For instance, Beaufort’s RGS-based study cited above examined an individual writer’s development across several years and university courses in order to examine how he uses genre knowledge to write in different writing contexts, while the cited EAP-based study by John Swales examined many instances of research article introductions in order to identify frequently-used moves and discourse.

As such, despite the parallel RGS and EAP instructional goals of demystifying academic writing for novice learners, their corresponding approaches have remained largely separate. In EAP, this separation helps explain less attention to the influence of particular rhetorical scenes on aggregate patterns. In RGS-based studies of FYW, the separation is reflected in two ways: few large-scale analyses of FYW genres, and relatively little emphasis on linguistic features in genre-based FYW pedagogy and research.

In three parts, this article examines these gaps in FYW and offers one way past them. The first section briefly outlines the historic separation of North American rhetoric-composition and linguistics in order to contextualize the methodological and ontological distinctions between RGS and EAP genre approaches today, in hopes that a historical awareness of the separations might help us “reinterpret and reintegrate” these existing approaches for the teaching of writing (Faigley 537). The subsequent section notes recent research on academic writing in RGS and EAP genre studies with a special focus on how insights from the two traditions together might enhance studies of FYW genres. The final section offers selected observations from a context-informed corpus linguistic analysis of over 19,000 evidenced-based argumentative essays written by incoming first-year writers that draws on conceptualizations of academic genres in both RGS and EAP traditions. The analysis highlights FYW patterns that appear to be influenced more by certain prompt cues than others and also points to linguistic patterns salient in FYW compared to published academic writing. These exploratory findings benefit from contextual attention usually absent in EAP genre research and from linguistic attention often absent in RGS research, and they thereby expose ways that parts of a rhetorical scene—in this case, cues in a FYW assignment—underpin the form (and) performance of a genre. The article closes with implications for teaching, research, and assessment.

By necessity, the three discussions in this article are brief and do not engage the full scope of associated issues. Instead, they strive to contextualize separate approaches in genre studies of academic writing and to foreground possibilities for studies of FYW that attend to both the macro-level and micro-level aspects of typified rhetorical actions.

A Brief History of Distinctions in Rhetorical and Linguistic Studies of English

Several developments over the past 200 years have contributed to separate linguistic and rhetorical approaches to written academic genres.{3} We might organize these chronologically into three overlapping periods of institutional history: (1) 1800-1910: the formation of U.S. English departments which ultimately privileged literary studies and shifted oral and written rhetorical studies to written composition (while orality went more often to speech or communication departments); (2) 1910 to 1950: the development of composition as “remedial” writing and the establishment of linguistics as a scientific field; (3) 1950 to today: the temporary linguistic debates in rhetoric-composition and the formal establishment of two fields concerned with academic writing, rhetoric-composition and applied linguistics, as largely separate from one another.

There are at least three, discipline-based explanations for how U.S. English studies emerged in the late 19th century. Literary scholar William Riley Parker suggested that the field of rhetoric, previously encompassing oratory, elocution, and all forms of written composition, began to “disperse itself” into the separate disciplines of oratory and written composition as English departments formed (349). Rhetoric-compositionist James Berlin argued that 19th-century English studies were more pointedly “installed to teach freshman composition” to prepare college graduates for professsions in which they would need to write (Rhetoric and Poetics 528). Linguist Colette Moore labeled philology as the 19th-century “anchor” for the discipline of English and modern language study, an anchor that shifted with the rise of literary studies at the turn of the century (159-60).

Shared across these accounts is the conclusion that late-19th-century English was less and less the study of philology and rhetoric and increasingly the study of literature with some written composition instruction. Two events at this time that have been interpreted as signaling this shift include the establishment of the Modern Language Association in 1883 and the introduction of remedial writing instruction at Harvard, which was followed by similar courses at most U.S. colleges.

These developments persisted in the twentieth century and affected the separate formation of rhetoric-composition and linguistics. Fueled by initiatives like the World War II GI Bill, many new students, who were thought in need of compulsory writing courses, enrolled in postsecondary schools. The resulting teaching crisis (and its many analogues since) is something scholars like Maxine Hairston (82) and David Bartholomae (11) suggest underpins the entire history of rhetoric-composition studies as a field. The mandatory FYW classes became economically valuable for English departments, as guaranteed courses and as teaching assignments for literary graduate students.{4} Scholarly and professional outlets for rhetoric-composition didn’t come until later, with the 1949 Conference on College Composition and Communication and compilations like Braddock et al. (1963) thereafter. In the meantime, linguistics was emerging as its own discipline that, though originally associated with rhetorical and philological study, had little reason to attach itself to what was becoming rhetoric-composition. Following the work of Ferdinand de Saussure, U.S. linguistics had a theoretical and scientific emphasis, which was reflected in its scholarly journal Language and its professional organization LSA by the 1920s.

By 1935, linguistics and what would become rhetoric-composition were moving in increasingly divergent directions, as the former pursued structural description of form and the latter confronted new teaching demands and a popular textbook market that “remained completely divorced from the burgeoning science of language” (Connors Composition-Rhetoric 162-63). With some exceptions, these separations persisted until the 1950s and 60s, when several debates emerged in rhetoric-composition about linguistically-informed instructional approaches (Berlin Rhetoric and Reality 112-13). Evidence of these overlaps include Charles Fries’ call for linguistic training in the teaching of writing, Francis Christensen’s sentence-combining exercises, and the Students' Right to Their Own Language (SRTOL) declaration by College Composition and Communication in 1974. SRTOL was specifically influenced by contemporaneous sociolinguistic research focused on the impact of social and cultural forces on language, and Barton and Stygall argue that such research “had the potential to enhance teachers' knowledge about the language they taught” in writing courses (2).

But many rhetoric-compositionists at the time had understandable concerns about linguistic approaches to writing. Not only did Research in Written Composition in 1963 state that formal grammar instruction provided a “negligible” or even “harmful effect” on student writing (Braddock, Lloyd-Jones and Schoer 37-38), some feared that adopting structural grammar might distance rhetoric-composition studies even further from their intellectual heritage (Crowley). The added dominance of Noam Chomsky’s theory of the ideal speaker-listener under ideal circumstances{5}—versus an actual writer with challenging needs and circumstances—helps explain why a new period of “suspicious withdrawal” characterized the relationship between linguistics and rhetoric-composition in the late twentieth-century (Barton and Stygall 5), even as applied linguistics formed.

Applied linguistics developed as a field dedicated to using linguistic findings and approaches to solve practical challenges like the needs of ELL students.{6} Though also shaped by pedagogical needs, rhetoric-composition in the late 20th-century included three strands that made connections with applied linguistics unlikely. One was the late 20th-century process movement, which tended to privilege the individual composing process (e.g., in Elbow’s expressivist model) and to attend less to the more collective forces that shape writing and the teaching of writing (Faigley; Barton and Stygall 2). Another was the “basic writing” movement led by Mina Shaughnessy, which initially maintained a focus on correctness and standard English that was resisted in sociolinguistic research of the time (e.g., Smitherman). A third was the “tacit language policy of unidirectional English monolingualism” in the field (Horner and Trimbur 594-95), which implied that writing by non-native speakers was outside the purview of rhetoric-composition (cf. Matsuda; Tardy).

Implications of these historical disciplinary divisions include the continued treatment of FYW as divorced from a longer rhetorical and linguistic tradition (Bartholomae; Kitzhaber Freshman English: A Prognosis; Purdy; S. Miller) and FYW research that rarely includes systematic analysis of smaller units of meaning like words and phrases. Some rhetoric-composition scholars have responded by calling for renewed emphasis on language: reexamination of our role as teachers of language (MacDonald), genre-based approaches that “keep generic form and generic context united” (Devitt Writing Genres 200), and a reintegration of language with writing studies (Aull and Lancaster; Devitt Written Language in Use; Gere; Aull, Lancaster, et al.). But given the focus of FYW research to date, there are not many descriptive, large-scale analyses of FYW, even as there is evidence that FYW genres are characterized by recurring linguistic features (Coffin and Hewings; Dudley-Evans; Aull and Lancaster).{7} As explained by Swales (Integrated), analysis across contexts through methods like corpus analysis is often at odds with the more top–down, process-based approach common in genre-based text analysis, where the starting point is a focus on larger units of text rather than word-level, lexical-grammatical trends.

Yet attention to both approaches can coexist in what I call context-informed corpus linguistic analysis of FYW, which examines the rhetorical cues of writing tasks alongside corpus linguistic patterns—in other words, an approach that explores examples of a FYW genre as realizations of particular rhetorical scenes and as patterns across them.{8} The next two sections strive to frame and illustrate this approach. First, by way of framing, I outline some relevant studies in RGS and EAP.

Genre Studies in Rhetoric-Composition

At the heart of RGS-based pedagogy today is a focus on genres as sites of typified social action. Genre in RGS has been a useful concept because it “emphasizes social and historical aspects of rhetoric that other perspectives do not” (C. Miller 151), and genre-based approaches to FYW especially explore these macro-level contextual aspects of genres. Amy Devitt writes that “people use genres to do things in the world (social action and purpose)” and that “these ways of acting become typified through occurring under what is perceived as recurring circumstances” and relations (Devitt Integrating 698; Devitt Generalizing 615). Devitt champions attention to context as well as form, but her framework for FYW courses ultimately foregrounds rhetorical scenes, perhaps in service of stressing “the process of learning new genres rather than specific linguistic features of specific genres” (Devitt Writing Genres 197). Anis Bawarshi underscores that “genre is what it allows us to do” and that genres become “endowed with certain social status and value” (Bawarshi The Genre Function 357), and he invites FYW students to analyze genres especially through questions of evidence and audience (Bawarshi Genre). Anne Beaufort stresses the need to understand the “broader goals and activities of a discourse community” and to engage in “discourse community analysis” in writing classrooms (Beaufort 19, 199). These studies have devoted close attention to the expectations embedded in contexts and communities, from whole-text structures to the social motives held by particular audiences. Accordingly, RGS-based approaches to FYW have especially illuminated the genre performances of whole texts in the context of FYW courses and FYW student lives.

Genre Studies in Applied Linguistic EAP

Like RGS, EAP genre studies underscores both form and social function. John Swales, for example, describes genres as “communicative events” characterized by “communicative purposes” and various patterns of “structure, style, content and intended audience” (1990a, 58). Likewise, the EAP concerns about educational access resonate with the RGS focus on the social import of genres insofar as successful negotiation of a genre can prove belonging in a discourse community (cf. Hyland Genre-Based Pedagogies). In contrast to RGS, however, EAP genre studies often include corpus linguistic analysis of recurring features across many texts, as a way of exposing how formal characteristics reflect the specialized functions and values of genres associated with particular academic expectations (Flowerdew; Hardy and Römer; Hyland Stance and Engagement). For instance, Ken Hyland has used corpus linguistic analysis to examine epistemological differences across disciplines; he delineates ways that academic articles in the natural sciences foreground evidence or phenomena rather than the authors’ reasoning by using fewer first person pronouns (or “self-mentions”), while social science and humanities articles use first person more in order to foreground the authors’ reasoning and to gain credit for their perspective (Hyland Stance and Engagement). In this example, Hyland’s object of analysis and the evidence for his claims about the values of particular discourse communities are word-level patterns.{9}

Especially via corpus linguistic analysis, EAP genre studies have examined lexical and grammatical features such as these across hundreds or thousands of academic texts. But it follows that EAP corpus-based research and resources often do not foreground contextual details foregrounded in RGS, such as prompts or local contexts. Given their commitment to non-native speaker students, EAP genre studies has also especially emphasized patterns of form in published academic writing and non-native speaker writing.

A More RGS-EAP Conceptualization of FYW

As the previous section indicates, historic separations between linguistics and rhetoric-composition are borne out in contemporary RGS and EAP genre studies. Even though both view academic genres as actions that fulfill shared communicative purposes in academic communities, there are implicit conceptual differences in each approach. In paying more attention to linguistic features, EAP genre studies can implicitly suggest that academic writing is most importantly realized in linguistic features that correspond to level or discipline. With more attention to the rhetorical cues of particular tasks, RGS can imply that academic writing is most importantly realized in the rhetorical achievements of whole texts.

This study strives to illustrate that neither of these foci is necessarily more true for FYW, nor must these foci be at odds. In this case, the conceptualizations from both RGS and EAP help elucidate FYW, by highlighting why linguistic patterns across contexts are meaningful, as well as why different rhetorical cues in the same FYW genre are also meaningful. Here, by “rhetorical cues,” I mean the writing prompts that FYW students receive, which proved in the analysis below—more than other contextual details like institution, year, or writing topic—to be an important influence in linguistic differences across FYW evidenced-based argumentative essays.

Analyzing FYW linguistic patterns vis-à-vis essay prompts is, of course, only one way to draw from genre conceptualizations in both RGS and EAP genre studies, but it reveals some compelling patterns. It offers insights based on approaching the genre of the FYW evidenced-based, argumentative essay as one realized in micro-level linguistic choices as well as recurring sociorhetorical actions, like the scope of the arguments that first-year writers make.{10} Though here I offer only selected observations from a larger project of this kind, they are sufficient to highlight two possible benefits of the approach. Because the FYW corpus analyzed is uniquely large for a specialized corpus (Flowerdew), it provides a view of recurring discourse patterns that complements more common ethnographic and qualitative analysis of FYW. Simultaneously, rather than only highlighting corpus patterns, the attention to the prompt informing each FYW subcorpus highlights distinctions that carry implications for FYW research and assignment design.

The FYW Corpus and the Analysis

The essays in the FYW corpus were written by incoming college students at two institutions as part of each institution’s directed self-placement (DSP) process. The essays were written by five FYW cohorts (2009-2013) at the University of Michigan, a public, northern research university (hereafter UM), and two FYW cohorts (2012-2013) at Wake Forest University, a private, southern university with a liberal arts mission (hereafter WFU). Each cohort of incoming students wrote in response to one essay prompt, meaning that the entire FYW corpus consists of uptakes of one of seven prompts. Unlike the opinion-based or timed writing common in standardized high school assessments, the DSP tasks required that students write over a week’s time in response to source texts in order to simulate early university writing (Gere, Aull, Green, et al.; Gere, Aull, Lancaster, et al.; Toth and Aull). In this way, the FYW corpus enables analysis of a common FYW genre in compulsory writing courses in U.S. universities as well as in international undergraduate contexts (Moore and Morton).{11} The student writers in the corpus were enrolled but had not yet begun their first-year writing courses; I refer to them as FYW students because they represent many students entering and seeking help in FYW contexts.

To facilitate identification of patterns in the FYW corpus and seven subcorpora, I used a free corpus software program called AntConc (Anthony); and to contrast the FYW patterns with published academic writing, I used another free tool, the 91-million-word academic subcorpus of the Corpus of Contemporary American English (COCAA), which is the most recent academic corpus of published writing of its size that is publically accessible.{12} My analytic process involved recursive analysis of corpus linguistics and close reading of the prompts and individual essays in a “bottom up” or data-driven process, so, all of the observations below emerged from the data rather than previous assumptions or intuition-driven analysis. More information about the DSP task as well as the specific prompts, years, institutions, and subcorpora sizes are offered in endnote {13} and are discussed in more detail below.

Overall, the similarity of the seven prompts as well as the time with which all the students had to familiarize themselves with the topic imply that differences in linguistic patterns are not based on topic but rather on the developmental level of the writers (Cumming et al.; Chapman). The broad patterns of textual features confirm this idea. In contrast to COCAA, most discourse patterns are similar across the over 19,000 first-year writers in the full FYW corpus regardless of college institution, writing topic, or year, suggesting that the “social action” of a FYW evidenced-based argumentative essay is somewhat stable. This conclusion is in keeping with other research that suggests that first-year writers have more in common with one another than with successful upper-level student writers or published academic writers (Aull and Lancaster), and the conceptualization of FYW implicit in these observations is one that prioritizes shared features across contexts. At the same time, there are distinctions in the DSP prompts that make it possible to analyze which prompt cues seem to solicit certain features of academic discourse more than others.

One reason such a prompt-attentive linguistic analysis is valuable is the dearth of research on the relationship between assignment prompts and the discourse features that commonly appear in responses to them. Research has focused on the variability of writing prompts (Jeffery; Melzer), the relationship between prompts and grades (Huot), and the importance of ensuring that students understand writing prompts and have the background knowledge to respond to them (Clark; Murphy and Yancey). There is likewise research on student performance and task-type (Plakans; Tsai and Cheng) and topic (Cumming). But assignment prompts remain frequently underconceptualized by those who create them (Gere, Aull, Lancaster, et al.), and to my knowledge there is almost no research in rhetoric-composition or specifically RGS that includes analysis of textual features vis-à-vis different writing prompts. One exception is Puma’s linguistic analysis of 100 FYW essays from the same college, which showed that FY student writers who know and feel close to a specified audience are more likely to draw from a spoken register.{14} Another is Beck and Jeffery’s 2007 study of high-stakes secondary writing assessments. Beck and Jeffery focus on language use across task types—e.g., interpretation, narrative, and argument—and conclude that argument may be best for these secondary assessments because “argument serves an important function as an organizing macrostructure for the presentation of one’s interpretive position” (75).

The writing prompt is also a subject of genre-based discussions that highlight the complex nature of the writing prompt in shaping generic academic performances. The writing prompt could be considered the sort of “homely discourse” to which Miller drew attention, but it has also been cast as its own genre or “meta-genre” which assumes and generates certain student and teacher positions (Bawarshi Genre; Giltrow; Soliday), even while such expectations may be tacit or occluded (Swales Occluded Genres). A more thorough engagement with prompts as genres is beyond the scope of this article, but these discussions underscore that writing prompts are not mere “conduits to invention” (Bawarshi Genre) and that RGS has taken seriously the role of prompts in the rhetorical scenes of FYW. By drawing on EAP approaches, we can further deduce some of the ways that particular prompt cues help shape student writing, even within what appears to be the same genre of the evidenced-based argumentative FYW essay.

The main descriptive findings are the following, which I unpack below: (1) the first-year writers use 1st person pronouns more than expert writers regardless of whether personal evidence was solicited by the prompt, but even more frequently when it was; (2) first-year writers use 1st person pronouns to mark evidence in various ways, including ways that do not mirror how expert writers use them; (3) when the FYW prompt both solicited personal evidence and posed an open-ended question—rather than inviting a direct response to a source text argument—the high frequency of references to personal evidence appears to be at the expense of references to the source text or author. These findings suggest that two textual patterns related to type and use of evidence correspond to specific prompt cues and impact the performance of the FYW genre.

Prompts and References to Personal Opinions and Views

In the seven prompts in this study, there were two key distinctions. The first distinction related to what was solicited as evidence in support of students’ arguments. Four of the prompts detailed in endnote {13} (UM 09, UM 12, WFU 12, and WFU 13) indicate students should draw their evidence from the assigned articles; three prompts invite evidence from the articles and also from students’ own “experiences” (UM 11 and UM 13) or “views” (UM 10).{15}

One key way writers in the corpora signaled personal evidence was through first person pronoun phrases beginning with in my or from my. These phrases situate claims on the author’s personal experiences, opinions, or views, such as in my personal experience or in my life. Table 1 below shows the frequencies of the top 22 phrases beginning with in my or from my.{16} The figures are normalized by 1 million words in order to make the findings comparable across subcorpora of different sizes.

Table 1. Frequency of Phrases in the COCAA and FYW Corpus.

	COCAA	FY (All)	UM09	UM10	UM11	UM12	UM13	WFU12	WFU13
From my…own, life, perspective, point of view, personal, view(s), belief(s)	3.22	13.52	11.35	6.19	18.41	8.61	25.31	3.21	6.54
In my…own, opinion, life, mind, view(s), belief(s), experience(s), eyes, personal, case	20.08	97.39	79.78	99.80	152.12	69.48	129.84	65.20	40.17
TOTAL	23.30	110.91	91.13	105.99	170.54	78.08	155.15	68.41	46.71
in my (own) view	4.83	1.14	1.95	0.41	1.75	1.43	0.65	1.07	0.00

The relative frequencies of these phrases in the FYW corpus and COCAA show salient differences, but there are also distinctions across the FYW subcorpora to which I return momentarily. With rare exceptions, the frequencies of every single in my and from my cluster are much higher in the first-year writing than in the expert writing in COCAA. The two notable exceptions are in my view and my own view, which are far more frequent in COCAA and far more rare in the FY subcorpora. Instead, the frequent FYW phrases tend to refer to the author’s belief(s) and opinion. It seems to follow that when first-year writers use these personal evidence markers frequently, they imply to their readers that the scope of their argument stems from their personal experience or perspective rather than existing research or views.

In the following examples from academic articles in COCAA, the first two offer the writer’s view of an issue under discussion, while the latter signals the writer’s view of a referenced argument:

1. Context Interpretation, in my view, should always be backward-looking. Interpretation requires one to be an honest agent. Judges are interpreters of the enacting legislature. A judge should try to find what the legislators intended, but that is often impossible. (Calabresi, Guido Harvard Law Review)
2. That is precisely the problem, in my view, with what has become of drug abuse research in recent decades: It has become increasingly divorced from the practical issues of treatment and prevention. (Dupont Robert L., Journal of Drug Issues)
3. Demographic sustainability, in my view and the view of increasing numbers of other conservationists (e.g., Redford and Feinsinger 2001, Peery et al. 2003, Soul et al. 2003), should be seen only as a threshold requirement (Eric Canderson, Bioscience)

These references appear to situate the COCAA writer’s contribution amidst other arguments, and, if we take the term view rather literally, to cast it as an outward-looking perspective. These examples of in my view in COCAA in some ways contrast examples of the most common in my phrase in the FY corpus, in my opinion.

1. Many times students approached me, asking for a little help on a test. "Hey man," they would say. "Is it cool if I sit by you for the test? I really need to get my grade up." "No way dude," I would reply. "What if you get caught? I'm not getting in trouble for you.” "Nah, the teacher wouldn't do anything. She likes me." Still not a reason to break the Academic Integrity code we all agreed to, in my opinion. (UM 13)
2. In my opinion, implementing a carbon tax on today’s society will only hinder economic growth, with the recession being too recent. (WFU 13)
3. In my opinion, this seems like one of the most productive ways to come up with a good end result, which was perhaps slightly overlooked by Lehrer. (UM 12)
4. This example, in my opinion, is exactly what type of role machines should be playing in society. The machine is doing a job, which no human could possibly accomplish alone, while increasing the quality of life for thousands of individuals at the same time. (UM 11)

In these examples, (A) clearly references personal experience, but (B), (C), and (D) could be considered the author’s views. If the author had used “view” rather than “opinion,” the various statements might appear more like expert academic argumentation, insofar as this wording (or form) is one that recurs more in expert genred performances of academic argumentation. Consider too the following examples of the common FY phrase I * believe, which seems to sometimes function like expert authors’ use of in my view.

1. After reading Jonah Lehrers' article, it became apparent that debating ideas among individuals with different viewpoints and being able to give any feedback was more beneficial and effective in developing creative ideas within a group than was the method of brainstorming. Upon further research, brainstormings' cardinal rule of no criticism, however, proved to be an important aspect of any group discussion. Therefore, I believe that while discussing ideas among a team of individuals with many perspectives, it is essential to only allow constructive criticism. (UM 12)
2. Personally, I believe I learn something new everyday [sic]. I definitely believe I have learned more from experiences I have had than from information that has been fed to me. (UM 10)
3. I believe that Osborn's definition of brainstorming is flawed because of his views on criticism. I do believe in a team coming up with numerous ideas, but I also think that debate enhances creativity. (UM 12)

These examples show that first-year writers tend to use first person singular pronouns in various ways. Sometimes first-year writers use them to refer to personal opinions or beliefs in ways that more advanced academic writers refer to their views, while other times, first-year writers refer clearly to personal experience in ways that matches more personal narrative than typical academic writing. These assorted uses of the first person singular furthermore appear much more frequently in FYW than in the writing of published academic writers. Even in the case of the UM 10 prompt, in which students are specifically invited to share their view, the first-year writers do not seem to have access to the social context in which the preferred stance (of someone dispassionately surveying a field) would seem desirable.

A look across these example linguistic patterns of the first person in the FYW corpus and COCAA suggests concrete ways that first person pronouns are used in advanced academic writing to signal the author’s views vis-à-vis an example or another view. It might follow that expressions of views more clearly function, and signal to readers, that the writer is sharing an argument or perspective, whereas phrases like in my experience or in my belief signal experiential information. Such experiential forms may not fit as clearly into the “constitutive rules that tell us how to fuse form and substance to make meaning,” and, for readers or instructors, into the “regulative rules that tell us how the fusion itself is to be interpreted within its context” (C. Miller 161). Put another way, while we know that the rule “don’t use the first person” is prescriptive and simplistic, college advice like “use the first person sparingly” or “use it when it serves your purpose” may be similarly inadequate for FYW students’ awareness of the social actions that academic argumentative genres fulfill. The findings here suggest specific ways we might foster students’ awareness of the importance of selecting and connecting evidence to their own and others’ views as they craft an academic argument, even when they (like the writers they are asked to emulate) choose to use the first person.

Prompts and References to Source Texts

It is not only students’ personal references that appear influenced by prompt cues, but also the relative frequency of references to source text materials. All of the FYW prompts emphasize the accompanying source texts as the primary or exclusive source of evidence in the DSP tasks. But they differ in whether they pose an open-ended question for students. This is the other key difference in the prompts (the first being the solicitation of personal evidence or lack thereof).

Two of the seven prompts solicit an answer to an open question, while the other five prompts imply that students’ arguments should respond to a perspective presented in a source text. In the prompts for UM 09, UM 11, UM 12, WFU 12, and WFU 13, all belong in the latter category. The UM 09 prompt asks students to “argue for or against” the proposal in the source text. The UM 11 prompt asks students to take one perspective described in the source text and “take a position on it”; similarly, the UM 12 prompt asks students to “support [their] claims about the ideas that [the source author] lays out.” The WFU 12 prompt asks students to “support or challenge” the author’s argument, and the WFU 13 prompt asks students to “support and/or challenge” a source author’s proposal. In these 5 prompts, a view from a source text is the point of departure, and students are asked to create an evidenced-based argument in response to it.

The UM 10 and UM 13 prompts instead pose more of an open question to students. The UM 10 prompt asks students to write an argument in answer to the question “What role should machines play in our lives?” The UM 13 prompt asks students to “write an essay in which you take a position on why students cheat.” Both of these prompts pose a question as the point of departure for students’ own arguments, though like the other 5 prompts, they emphasize the source text; and like UM 11, they also solicit personal evidence. Frequencies of references to source texts and authors (including author and text names as well as references to an author or article) are shown in Table 2 below. As with the previous table, figures are normalized to 1 million words.{17}

Table 2. Frequency of References to Source Texts and Authors.

	UM09	UM10	UM11	UM12	UM13	WFU12	WFU13
References to article*; author*; author and article names (One per phrase)	93.671	28.073	98.411	85.148	35.339	84.391	59.597

The two patterns shared across UM 10 and UM 13 (but not UM 11) are as follows: one, they show relatively high frequencies of first person pronouns, and two, they show relatively few references to source texts and authors (40-70% fewer references to the source text than all of the other FYW subcorpora). These trends are interesting in light of the fact that the UM 10 and UM 13 prompts are the only ones characterized by both a more open-ended question and the option for students to use personal experience as evidence. That is, they suggest that, relative to the other prompts, prompts with an open-ended (versus source-directed) question which also solicit personal evidence may lead to personal evidence at the expense of source text evidence.

All of these observations carry potential implications for FYW assignment design and assessment. In this study, FYW students referred to their own experiences, beliefs, and opinions more when they were invited to do so, and they also did so, in lower frequencies, when they were not explicitly invited to do so. Likewise, the first-year writers who were responding to an open-ended question and were invited to use personal evidence—accompanied by source material they were required to consider—referenced the source texts relatively less and their experiences relatively more. This last pattern seems to point to the similarity between the UM 10 and UM 13 prompts and a common antecedent genre for student writers entering college: standardized assessments that ask students to write an opinion-based response to an open question such as whether high school students should wear uniforms (For examples of such prompts, see Jeffery; Ramineni). If we draw from insights from RGS to explain these UM 10 and UM 13 patterns, we might therefore see them as understandable responses based on familiarity and antecedent “school genres.” Without awareness of an important social action of advanced academic arguments—joining a conversation of existing views, often occupying the rhetorically disinterested stance of “viewer”—the UM 10 and UM 13 students might have seen the rhetorical cues of the prompt and opted for using the more familiar evidence of personal opinion or experience, evidence which has been acceptable in past school genre experiences in which they responded to an open question.

Put in terms of instruction and assessment, these findings imply that when we want students to write evidenced-based academic essays, it behooves us to consider what kinds of language and evidence characterizes the academic writing we think would be a successful response to any given prompt we design. The findings more specifically imply that we may want to be cautious about expressly inviting students’ experiences and opinions as key sources of evidence. This is not because there is no place for personal evidence in academic writing, of course, but rather because (a) FYW students seem likely to draw from their experiences regardless of the prompt; (b) when FYW students draw from their own experiences or views, the linguistic features they use to do so often do not match the recurring features of expert academic writing; and (c) self-reference may lead students to reference the source text less depending on the prompt. These observations suggest, then, that in designing and facilitating uptake of the occluded genre of the writing prompt, we should attend to the fusions of form and substance that are commonly valued in academic arguments.

In sum, if we believe that a key social action of the FYW evidenced-based argumentative essay genre is positing the author’s argument as one view among many (or at least facilitating FYW students’ transition into such dialogue), then we should strive to design prompts that set up a rhetorical scene consistent with that social action. In my view, descriptive, linguistic attention to examples of student and expert writing aids this process of clarifying and aligning genre-based instruction and assessment.

Concluding Remarks

A more traditional RGS analysis of the contexts and strategies of a few of the first-year writers in this study could have uncovered more about specific texts or specific students’ understandings of the prompt cues. But it would not have exposed the same recurring, empirical patterns across many FYW texts and in comparison with published academic writing. Likewise, a more common EAP corpus linguistic analysis could have uncovered more differences between the first-year writers and the COCAA writers that are related to uses of personal evidence. But it would not have highlighted the differences in phrases that correspond to specific rhetorical cues students attend to as they write.

In a sense, then, this analysis is an additional way of considering rhetorical situation as a semiotic structure, a system in which the social action of genre is always to some degree realized in linguistic action. This study, for instance, exposes recurring language patterns which intimate a potential continuum of academic genres: from standardized high school genres that solicit personal evidence in response to open-ended questions, to early university genres that draw on personal evidence and source text evidence for argumentation, to advanced academic writing that analyzes competing positions and uses personal evidence only when it is clearly evaluated and relevant (Thaiss and Zawacki 6). The findings here suggest that genres on such a spectrum—and their relation to one another as antecedent genres—are recognizable, reconstituted, and also discoverable in part through typified linguistic patterns.

Of course, there is a great deal of variety in FYW, and there are also many who disagree with conventional FYW genres, which have been critiqued as “institutional” rather than rhetorical (Crowley 8). For this reason, some rhetoric-compositionists argue that FYW genres cannot simulate disciplines and are better-suited for fostering general writing awareness (e.g., Russel; Wardle). To me, that is all the more reason to study the empirical, linguistic patterns of academic genres as students engage with them. The study outlined in this article is one example of how we might explore patterns in the writing of many students transitioning from high school writing to more advanced academic writing as they confront particular tasks, and its results suggest specific ways we might think about prompt cues and how we discuss the social actions of academic genres in terms of evidence and argumentative scope with FYW students.

This approach also carries implications for writing assessment and writing transfer, which to date have focused little on linguistic features as integral to understanding students’ genre performances. Current writing assessment criteria in the new Common Core State Standards separates “language/writing style” from “argument” and “structure,” category separations that belie the ways that recurring, intelligible patterns in academic genres shape the arguments, organization, and logic therein. The past WPA Outcomes Statement likewise labeled language as part of the “surface features” of genres, though the recently-revised version strives to alter this message. In genre transfer research, studies show that students struggle with how to approach word-level demands (Negretti 151, 60) and sentence-level features in their writing (Jarratt et al. 4), but they rarely include textual analysis in their methods.

There is much more to be learned about both linguistic and macro-level characteristics of FYW that are not explored in this article. And linguistic attention should supplement, not supplant, attention to macro-level details. Even so, this discussion and example pose some possibilities for more contextually-aware but also linguistically-attentive approaches to FYW. Such possibilities might help illuminate more of the recurrent patterns of language use that Miller noted as one level of understanding genres (163). And they might expand current approaches to better match the ways that writers “encounter, interpret, react to, and create particular texts” (151), since those endeavors are ever rooted in contexts, whole texts, and language.

Notes

1. Throughout this article, I use the term “rhetoric-composition” for two reasons. The first reason is conceptual: this article attempts to historicize several threads in English language studies and to draw from several traditions in the analysis, and so I use the term “rhetoric-composition” to evoke the history of both rhetoric and composition studies in current genre-based approaches to first-year writing. The second reason is more practical: listing “linguistics and rhetoric and composition” appears like three fields instead of two, and because I want to refer to “linguistics” and “rhetoric-composition” as two traditions, the hyphenated term seems less confusing. (Return to text.)
2. These methodological distinctions hold true even in genre-oriented discourse studies in composition, which tend to analyse a few texts by hand but which otherwise focus on micro-level textual features and form a kind of bridge between linguistics and rhetoric-composition (e.g., see Ellen L. Barton and Gail Stygall, Discourse Studies in Composition [Hampton Pr, 2002]). (Return to text.)
3. As discussed by Connors, before the 19th century and to a large extent throughout it—following the establishment of the United States until the Morrill Act of 1862 established Agricultural and Mechanical Colleges—college education was “the province of a small group of elite students” who studied both oral and written discourse (Robert J. Connors, Composition-Rhetoric: Backgrounds, Theory, and Pedagogy [Pittsburgh: U of Pittsburgh P, 1997] 9.). Courses were not offered in formalized English departments but drew from many traditions including rhetorical, literary, linguistic, and philosophical ones. As a result, the most important events in the separate development of linguistics and rhetoric-composition (as we know them today) can be said to have occurred after the mid-19th century. (Return to text.)
4. See James Berlin, Rhetorics, Poetics, and Cultures: Refiguring College English Studies (Parlor Press LLC, 2003), Connors, Composition-Rhetoric, Sharon Crowley, Composition in the University: Historical and Polemical Essays (Univ of Pittsburgh Pr, 1998), Susan Miller, Textual Carnivals : The Politics of Composition (Carbondale: Southern Illinois University Press, 1991), William Riley Parker, Where Do English Departments Come From?, College English 28.5 (1967), Albert Kitzhaber, Rhetoric in American Colleges: 1850-1900, Dissertation (University of Washington: Southern Methodist University Press (1990), 1953), Donald C. Stewart, The Status of Composition and Rhetoric in American Colleges, 1880-1902: An Mla Perspective, College English 47.7 (1985). (Return to text.)
5. Chomsky was, of course, not the first to foreground the scientific linguistic object as divorced from human social conditions; this same passage explicitly echoes the “founders of modern general linguistics” in the late 19th century, the time of initial divisions between linguistics and rhetoric-composition noted earlier. (Return to text.)
6. For example, the journal Applied Linguistics describes that it shares “research into language with relevance to real world problems” (http://applij.oxfordjournals.org/). (Return to text.)
7. Little focus on FY writing persists even as there is an increasing international interest in more advanced undergraduate genres, e.g., Australian undergraduate thesis introductions (S. Hood, Managing Attitude in Undergraduate Academic Writing: A Focus on the Introductions to Research Reports, Analysing Academic Writing: Contextualised Frameworks eds. L. Ravelli and R. Ellis [London: Continuum, 2004].); engagement patterns in Hong Kong final year-reports (Ken Hyland, Disciplinary Discourses : Social Interactions in Academic Writing, Michigan classics ed. [Ann Arbor: University of Michigan Press, 2004].); and elemental genres in Australian short response science papers (H. Drury, Short Answers in First-Year Undergraduate Science Writing. What Kind of Genres Are They?, Academic Writing in Context: Implications and Applications. Papers in Honor of Tony Dudley-Evans, ed. M. Hewings [Birmingham: University of Birmingham Press, 2001].). (Return to text.)
8. As evidenced in the review of genre research below, this idea is theoretically espoused in applied linguistics as well as rhetoric-composition. The term, then, serves not to label something that doesn’t exist but to highlight an approach that is rare but useful in studying FY writing. (Return to text.)
9. In other examples, EAP research has examined lexico-grammatical features such as verb tense, hedges, and passive voice in academic genres (Hanania & Akhtar, 1985; Salager-Meyer, 1994; Swales, 1990a; Tarone, Dwyer, Gillette, & Icke, 1981). EAP research on slightly larger units of text include structural move analyses to describe recurring rhetorical moves and sections in research articles (Swales, 1981, 1990a), master of science dissertations (Hopkins & Dudley-Evans, 1988), medical abstracts (Salager-Meyer, 1990), popularized medical research reports (Nwogu, 1991), and university lectures (Thompson, 1994). (Return to text.)
10. Though outside the scope of this discussion, it is worth nothing that the “generic” status of this kind of writing has been hotly debated, including by Wardle (Mutt genres), from which I draw in my interpretations. I view corpus-based analysis as one additional way to explore theses important questions, in part because it can show us patterns of linguistic forms that can otherwise escape our notice. But in this study, I take the frequent use of this kind of essay and its shared communicative purposes of transitioning student writers into more evidenced-based, academic argumentation as a sufficient starting point for studying patterns across many examples of it. (Return to text.)
11. Each task asked students to read 8-12 pages of expository writing from a source like the New Yorker or Atlantic and write an evidenced-based argumentative essay of approximately 800-1000 words within a week’s time. The two institutions have obtained similar U.S. News rankings (#23 and #28), and are distinct in region and institution type (one is a northern, public research institution while the other is a southern, private institution with a liberal arts ethos). The size of the corpus—and the fact it accounts for the full incoming cohorts at each school—mitigates strong outliers from these broad student demographics. Thus the corpus can be said to approximate the writing of the hundreds of thousands of students enrolling at different kinds of top-fifty U.S. universities in response to evidenced-based reading-to-write tasks but not as representative of students at the most elite U.S. institutions, at open-access institutions, or at institutions that otherwise cater to more non-traditional students. (Return to text.)
12. COCAA is searchable as a whole or can be restricted to writing in particular discipline groups, but it also includes more general academic sources like American Scholar and Futurist and therefore captures a range of published academic writing for comparison to the FY corpus (which is a general academic rather than discipline-specific corpus). This study includes the entire COCAA corpus as what I label “expert academic writing” mainly because it is comprised of 91,044,778 words from almost 100 different peer-reviewed journals across a range of disciplinary contexts and more general academic publications The Corpus of Contemporary American English (Coca), 2008-, <http://www.americancorpus.org>. 7. The COCAA100+ peer-reviewed journals are selected to cover the range of the Library of Congress classification system: a certain percentage from B (philosophy, psychology, religion), D (world history), K (education), T (technology), etc.), both overall and by number of words per year. The last update to COCAA was June 2012 (Davies, pc, May 25, 2013). At the same time, COCAA does not include articles every year from every flagship journal in every discipline, nor does it specify according to genre (e.g., argumentative essay or research report). I therefore use COCAA not as an ideal representation of published academic journal writing but rather as a reference point for comparing features across levels of academic discourse in corpora large enough to help control for idiosyncratic writers. (Return to text.)
13. At both institutions, the students were informed that their essay would be sent to the instructor of their first writing-intensive course, but the essays were not used to place students into a course; as a part of self-placement, the course selection is ultimately the students’ choice.

    FY subcorpora designated by year, school, topic	Total number of essays	Total number of word tokens	Prompt highlights
    UM 09: Read “Most Likely to Succeed” Malcolm Gladwell. Analyze Gladwell's proposal on how to select and retain teachers in the United States, and argue for or against his proposal using evidence from the article.	3,573	3,062,449	Agree or disagree with article; draw from article
    UM 10: Some argue our reliance on machines for our daily activities enhances our lives; others argue it may diminish human interactions. Based on evidence from Jerome Groopman’s article “Robots that care” and your own views, write argument in answer to “What role should machines play in our lives?”	2,651	2,413,405	Open–ended question; draw from article and own views
    UM 11: In “Mind vs. Machine,” Brian Christian surveys several perspectives on what it means to be human. For instance, at one point he writes that “being human (and being oneself) is about more than simply showing up,” and at another that the ability to be “zany, a jokester, [is] a much more ‘human’ personality type.” Read the article carefully, and pay close attention to the many perspectives it presents on what it means to be human. Then, select one of these perspectives and — drawing on evidence from the article, as well as your own experience and/or other texts you have read — take a position on it. Do you agree with it or not, and why? How does other evidence from the article complicate your position?	2,554	2,260,043	Respond to Christian’s argument; draw from article, other texts, own experience
    UM 12: Write an essay in which you both summarize and analyze Jonah Lehrer’s article “Groupthink.” Present your analysis as a persuasive essay, using evidence from the article to support your claims about the information and ideas that Lehrer lays out.	4,380	4,846,463	Respond to Lehrer’s ideas; draw from article
    UM 13: In his 2012 article Cheating Upwards, Robert Kolker offers several views on why cheating appears to be on the rise in highly competitive U.S. educational institutions. Read this article carefully and pay close attention to the various perspectives it presents on why cheating has become so prevalent. Drawing on one or more of these perspectives, write an essay in which you take a position on why students cheat. You may supplement your argument with evidence from your own experience and/or other texts you may have read. Be sure to indicate the source for any textual evidence, aside from this article, that you use.	4,056	4,523,370	Open-ended question; draw from article, other texts, own experience
    WFU 12: Read Malcolm Gladwell’s article entitled Small Change: Why the revolution will not be Tweeted. Analyze Gladwell’s argument about digital media and social revolutions. Then write your own essay in which you support or challenge his argument using evidence from the article. 1,046	933,275	Support or challenge article; draw from article
    WFU 13: Read the informational article from the National Journal entitled The Scary Truth about How Much Climate Change Is Costing You and the argumentative piece entitled Paying for it (links are provided below). In Paying for it, Elizabeth Kolbert asserts that it is time for a carbon tax as a way to help mitigate environmental problems discussed in the National Journal article. Analyze both texts and then write your own essay in which you support and/or challenge Kolbert’s argument for a carbon tax using evidence from the two articles.	1,201	1,043,089	Agree, disagree, qualify position in article; draw from articles
    TOTAL	19,461	19,082,094

    (Return to text.)
14. The tendency for developing FY writers to employ a more informal register than more advanced writing peers is also outlined more recently in a linguistically-informed genre analysis of FY writing that pays some attention to a single prompt (Anne Ruggles Gere, Laura Louise Aull, Zak Lancaster, Moises Perales Escudero and Elizabeth Vander Lei, Local Assessment: Using Genre Analysis to Validate Directed Self-Placement, College Composition and Communication 4.64 [2013]). (Return to text.)
15. The UM 11 and UM 13 prompts also describe that students can draw from other texts, but this distinction did not appear to make a difference like the other prompt parameters, particularly because most students primarily drew from the assigned article and their own experience for evidence. (Return to text.)
16. The table below contains the full list and frequency of the top from my and in my clusters in the COCAA and FY corpora.

    	COCAA	FY (All)	UM09	UM10	UM11	UM12	UM13	WFU12	WFU13
    From my own	2.28	8.58	7.78	3.71	10.08	4.51	18.33	1.07	1.87
    from my life	0.05	0.31	0.00	0.00	0.88	0.20	0.44	1.07	0.00
    from my point of view	0.31	0.99	0.97	0.41	0.88	0.82	1.53	1.07	0.93
    from my personal	0.10	2.08	2.27	1.24	3.07	1.23	3.49	0.00	0.93
    from my perspective	0.46	1.35	0.32	0.82	2.19	1.64	1.53	0.00	2.80
    from my view	0.01	0.10	0.00	0.00	0.88	0.00	0.00	0.00	0.00
    from my views	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
    from my belief	0.00	0.05	0.00	0.00	0.44	0.00	0.00	0.00	0.00
    from my beliefs	0.00	0.05	0.00	0.00	0.00	0.20	0.00	0.00	0.00
    Total from my	3.22	13.52	11.35	6.19	18.41	8.61	25.31	3.21	6.54
    In my own	3.59	16.17	6.49	7.42	22.80	9.63	36.88	4.28	0.93
    in my opinion	3.29	45.91	43.46	63.92	72.34	33.20	41.90	45.96	29.89
    in my life	3.14	8.79	4.22	12.78	21.92	7.79	6.98	4.28	0.93
    in my mind	2.27	7.44	6.16	8.66	13.59	5.74	7.86	4.28	3.74
    in my experience	1.70	8.01	8.11	1.65	4.82	5.53	18.33	3.21	0.00
    in my experiences	0.04	1.20	1.95	0.00	1.75	1.02	1.75	0.00	0.00
    in my eyes	0.25	2.70	3.57	2.47	3.51	1.64	3.49	0.00	2.80
    in my personal	0.13	3.69	1.62	1.24	4.82	2.66	8.29	1.07	0.00
    in my views	0.02	0.05	0.32	0.00	0.00	0.00	0.00	0.00	0.00
    in my view	4.75	1.14	1.95	0.41	1.75	1.43	0.65	1.07	0.00
    in my belief	0.04	0.62	0.65	0.82	2.63	0.00	0.00	0.00	1.87
    in my beliefs	0.00	0.10	0.00	0.00	0.00	0.00	0.22	1.07	0.00
    in my case	0.83	1.56	1.30	0.41	2.19	0.82	3.49	0.00	0.00
    Total in my	20.08	97.39	79.78	99.80	152.12	69.48	129.84	65.20	40.17
    TOTAL	23.30	110.91	91.13	105.99	170.54	78.08	155.15	68.41	46.71

    (Return to text.)
17. The total frequencies in the table only account for one mention per reference in each essay; e.g., the statement “The article ‘Most Likely To Succeed’” counts once; likewise, a sentence like “the article ‘Groupthink’ by Jonah Lehrer” counts once as well, since the author is referring to the article once though in three lexical items (the article, Groupthink, and Jonah Lehrer). Multiple references, as in these two examples, occurred about 18% of the time across the subcorpora. Below is the list of targeted lexical items in explicit source text references in the corpora.

    List of lexical items in explicit source text references
    Gladwell*	“Cheating upwards”
    “Most likely to succeed”	“Small Change: Why the revolution will not be tweeted”
    Groopman*	Kolbert*
    “Robots that care”	Davenport*
    Christian*	“Paying for it”
    “Mind vs./v./versus Machine”	“The Scary Truth about How Much Climate Change Is Costing You”
    Lehrer*	article*
    “Groupthink”	author*
    Kolker*

    (Return to text.)

Works Cited

Anthony, Laurence. Antconc. Vers. 3.2.4. Computer software. Laurence Anthony, University of Waseda, 2011.

Aull, Laura Louise, and Carlton Isaac Zak Lancaster. Linguistic Markers of Stance in Early and Advanced Academic Writing: A Corpus-Based Comparison. Written Communication 1.33 (2014). Print.

Bartholomae, David. What Is Composition and (If You Know What That Is) Why Do We Teach It? Composition in the Twenty-First Century : Crisis and Change. Eds. Bloom, Lynn Z., Donald A. Daiker and Edward M. White. Carbondale: Southern Illinois University Press, 1996. 11-28. Print.

Barton, Ellen L., and Gail Stygall. Discourse Studies in Composition. Hampton Pr, 2002. Print.

Bawarshi, Anis. Genre and the Invention of the Writer : Reconsidering the Place of Invention in Composition. Logan, Utah: Utah State University Press, 2003. Print.

---. Genre and the Invention of the Writer: Reconsidering the Place of Invention in Composition. Logan, Utah: Utah State University Press, 2003. Print.

---. The Genre Function. College English 62.3 (2000): 335-60. Print.

Beaufort, Anne. College Writing and Beyond: A New Framework for University Writing Instruction. Utah State University Press, 2007. Print.

Beck, Sarah W., and Jill V. Jeffery. Genres of High-Stakes Writing Assessments and the Construct of Writing Competence. Assessing Writing 12.1 (2007): 60-79. Print.

Berlin, James. Rhetoric and Poetics in the English Department: Our Nineteenth-Century Inheritance. College English 47.5 (1985): 521-33. Print.

---. Rhetoric and Reality : Writing Instruction in American Colleges, 1900-1985. Studies in Writing & Rhetoric; Conference on College Composition and Communication. Carbondale: Southern Illinois University Press, 1987. Print.

---. Rhetorics, Poetics, and Cultures: Refiguring College English Studies. Parlor Press LLC, 2003. Print.

Braddock, Richard Reed, Richard Lloyd-Jones, and Lowell Schoer. Research in Written Composition. National Council of Teachers of English Champaign, IL, 1963. Print.

Chapman, Mark. Writing Assessment: Relationship between Prompts and Linguistic Patterns. 2013. Print.

Chomsky, Noam. Aspects of the Theory of Syntax. M.I.T. Press, 1965. Print.

Christensen, Francis. A Generative Rhetoric of the Sentence. College Composition and Communication 14.3 (1963): 155-61. Print.

Clark, Irene. A Genre Approach to Writing Assignments. Composition Forum 14.2 (2005). Web. <http://compositionforum.com/issue/14.2/clark-genre-writing.php>.

Coffin, C., and A. Hewings. Ielts as Preparation for Tertiary Writing: Distinctive Interpersonal and Textual Strategies. Analysing Academic Writing: Contextualized Frameworks (2004): 153-71. Print.

Connors, Robert J. Composition-Rhetoric: Backgrounds, Theory, and Pedagogy. Pittsburgh: U of Pittsburgh P, 1997. Print.

---. The Erasure of the Sentence. College Composition and Communication 52.1 (2000): 96-128. Print.

Crowley, Sharon. Composition in the University: Historical and Polemical Essays. Univ of Pittsburgh Pr, 1998. Print.

Cumming, Alister. Writing Expertise and Second‐Language Proficiency*. Language Learning 39.1 (1989): 81-135. Print.

Cumming, Alister, et al. Differences in Written Discourse in Independent and Integrated Prototype Tasks for Next Generation Toefl. Assessing Writing 10.1 (2005): 5-43. Print.

Davies, Mark. The Corpus of Contemporary American English (Coca). 2008-. Print.

Devitt, Amy J. Generalizing About Genre: New Conceptions of an Old Concept. Relations, Locations, Positions: Composition Theory for Writing Teachers. Eds. Vandenberg, Peter, Sue Hum and Jennifer Clary-Lemon. Urbana, Illinois: NCTE, 2006. 84-102. Print.

---. Integrating Rhetorical and Literary Theories of Genre. College English 62.6 (2000): 696-718. Print.

---. Writing Genres. Rhetorical Philosophy and Theory. Carbondale: Southern Illinois University Press, 2004. Print.

---. Written Language in Use. Contours of English and English Language Studies Eds. Adams, Michael and Anne Curzan. Ann Arbor: University of Michigan Press, 2011. 298-314. Print.

Downs, Douglas, and Elizabeth Wardle. Teaching About Writing, Righting Misconceptions:(Re) Envisioning ‘First-Year Composition’ as ‘Introduction to Writing Studies.’ College Composition and Communication (2007): 552-84. Print.

Drury, H.. Short Answers in First-Year Undergraduate Science Writing. What Kind of Genres Are They? Academic Writing in Context: Implications and Applications. Papers in Honor of Tony Dudley-Evans. Ed. Hewings, M. Birmingham: University of Birmingham Press, 2001. 104-21. Print.

Dudley-Evans, Tony. The Teaching of the Academic Essay: Is a Genre Approach Possible? Genre in the Classroom: Multiple Perspectives. Ed. Johns, A.M.: L. Erlbaum, 2002. Print.

Elbow, Peter. Writing without Teachers. Oxford University Press, 1998. Print.

Faigley, Lester. Competing Theories of Process: A Critique and a Proposal. College English 48.6 (1986): 527-42. Print.

Flowerdew, Lynne. The Argument for Using English Specialized Corpora to Understand Academic and Professional Language. Discourse in the Professions. Perspectives from Corpus Linguistics (2004): 11-33. Print.

Fries, Charles Carpenter. American Linguistics and the Teaching of English. Language Learning 6.12 (1955): 1-22. Print.

---. Merrill Linguistic Readers: A Basic Program. Charles E. Merrill Books, Inc., 1966. Print.

Gere, Anne Ruggles, et al. Assessing the Validity of Directed Self-Placement at a Large University. Assessing Writing 15.3 (2010): 154-76. Print.

Gere, Anne Ruggles, et al. Local Assessment: Using Genre Analysis to Validate Directed Self-Placement. College Composition and Communication 4.64 (2013). Print.

Giltrow, Janet. Meta-Genre. The Rhetoric and Ideology of Genre. Eds. Coe, Richard M, L. Lingard and T. Teslenko. Creskill, NJ: Hampton, 2002. 187-205. Print.

Hairston, Maxine. The Winds of Change: Thomas Kuhn and the Revolution in the Teaching of Writing. College Composition and Communication 33.1 (1982): 76-88. Print.

Hardy, Jack A, and Ute Römer. Revealing Disciplinary Variation in Student Writing: A Multi-Dimensional Analysis of the Michigan Corpus of Upper-Level Student Papers (Micusp). Corpora 8.2 (2013): 183-207. Print.

Hood, S. Managing Attitude in Undergraduate Academic Writing: A Focus on the Introductions to Research Reports. Analysing Academic Writing: Contextualised Frameworks Eds. Ravelli, L. and R. Ellis. London: Continuum, 2004. 24-44. Print.

Horner, Bruce, and John Trimbur. English Only and U.S. College Composition. College Composition and Communication 53.4 (2002): 594-630. Print.

Huot, Brian. The Literature of Direct Writing Assessment: Major Concerns and Prevailing Trends. Review of Educational Research 60.2 (1990): 237-63. Print.

Hyland, Ken. Disciplinary Discourses : Social Interactions in Academic Writing. Michigan classics ed. Ann Arbor: University of Michigan Press, 2004. Print.

---. Genre-Based Pedagogies: A Social Response to Process. Journal of Second Language Writing 12.1 (2003): 17-29. Print.

---. Stance and Engagement: A Model of Interaction in Academic Discourse. Discourse Studies 7.2 (2005): 173-92. Print.

Jarratt, Susan C., et al. Pedagogical Memory: Writing, Mapping, Translating. Writing Program Administration 33.1-2 (2009): 46. Print.

Jeffery, Jill V. Constructs of Writing Proficiency in Us State and National Writing Assessments: Exploring Variability. Assessing Writing 14.1 (2009): 3-24. Print.

Kitzhaber, Albert. Freshman English: A Prognosis. CE 23 (1962): 476-83. Print.

---. Rhetoric in American Colleges: 1850-1900. Dissertation. University of Washington: Southern Methodist University Press (1990), 1953. Print.

MacDonald, Susan Peck. The Erasure of Language. College Composition and Communication (2007): 585-625. Print.

Matsuda, Paul Kei. Composition Studies and Esl Writing: A Disciplinary Division of Labor. College Composition and Communication 50.4 (1999): 699-721. Print.

Melzer, Dan. Writing Assignments across the Curriculum: A National Study of College Writing. College Composition & Communication 61.2 (2009): 378. Print.

Miller, Carolyn. Genre as Social Action. Quarterly Journal of Speech 70.2 (1984): 151-67. Print.

Miller, Susan. Textual Carnivals : The Politics of Composition. Carbondale: Southern Illinois University Press, 1991. Print.

Moore, Colette. History of the English Langauge in the English Department: Past and Present. Contours of English and English Language Studies Eds. Adams, Michael and Anne Curzan. Ann Arbor: University of Michigan Press, 2011. 157-67. Print.

Moore, Tim, and Janne Morton. Dimensions of Difference: A Comparison of University Writing and Ielts Writing. Journal of English for Academic Purposes 4.1 (2005): 43-66. Print.

Murphy, Sandra, and Kathleen Blake Yancey. Construct and Consequence: Validity in Writing Assessment. Handbook of Research on Writing: History, Society, School, Individual, Text (2008): 365-85. Print.

Negretti, Raffaella. Metacognition in Student Academic Writing. Written Communication 29.2 (2012): 142-79. Print.

Parker, William Riley. Where Do English Departments Come From? College English 28.5 (1967): 339-51. Print.

Plakans, Lia. Comparing Composing Processes in Writing-Only and Reading-to-Write Test Tasks. Assessing Writing 13.2 (2008): 111-29. Print.

Puma, Vincent D. Relationships between Writer-Audience Proximity, Register, and Quality in the Essays of First-Year College Students. (1986). Print.

Purdy, Dwight. A Polemical History of Freshman Composition in Our Time. College English 48.8 (1986): 791-96. Print.

Ramineni, Chaitanya. Validating Automated Essay Scoring for Online Writing Placement. Assessing Writing 18.1 (2013): 40-61. Print.

Russell, D. Activity Theory and Its Implications for Writing Instruction. Reconceiving Writing, Rethinking Writing Instruction (1995): 51-78. Print.

Shaughnessy, Mina. Diving In: An Introduction to Basic Writing. College Composition and Communication 27.3 (1976): 234-39. Print.

---. Errors and Expectations: A Guide for the Teacher of Basic Writing. Oxford University Press, 1979. Print.

Smitherman, G. Talkin and Testifyin: The Language of Black America. Wayne State University Press, 1977. Print.

Soliday, M. Everyday Genres: Writing Assignments across the Disciplines. SIU Press, 2011. Print.

Stewart, Donald C. The Status of Composition and Rhetoric in American Colleges, 1880-1902: An Mla Perspective. College English 47.7 (1985): 734-46. Print.

Swales, John. Genre Analysis: English in Academic and Research Settings. Cambridge Applied Linguistics Series. Cambridge [England] ; New York: Cambridge University Press, 1990. Print.

---. Integrated and Fragmented Worlds: Eap Materials and Corpus Linguistics. Academic Discourse. Ed. Flowerdew, John. London: Longman, 2002. 150-64. Print.

---. Occluded Genres in the Academy. Academic writing (1996): 45-58. Print.

Tardy, Christine M. Building Genre Knowledge. Second Language Writing. West Lafayette, Ind.: Parlor Press, 2009. Print.

Thaiss, Chris, and Terry Myers Zawacki. Engaged Writers Dynamic Disciplines. Engaged Writers Dynamic Disciplines (2006). Print.

Toth, Christie, and Laura Aull. Directed Self-Placement Questionnaire Design: Practices, Problems, Possibilities. Assessing Writing 20.0 (2014): 1-18. Print.

Tsai, Pi-chen, and Yuh-show Cheng. The Effects of Rhetorical Task Type, English Proficiency, and Writing Anxiety on Senior High School Students' English Writing Performance. English Teaching & Learning 33.3 (2009). Print.

Wardle, E. ‘Mutt Genres’ and the Goal of Fyc: Can We Help Students Write the Genres of the University? College Composition and Communication 60.4 (2009): 765-89. Print.