“If We Don’t Teach Them, Who Will?”: Standard Language Ideology in the University English Classroom

Ho'omana Nathan Horton

Imagine my surprise when, as a new graduate student in an English department, and a native user of “seen” for “saw,” I entered the department office and saw the meme in Figure 3.1 posted on the wall.

The presence of such sentiments in the front office of a university English department reflects the fact that university composition instructors may often find

FIGURE 3.1 Meme posted in a university English department front office (someecards.com).

themselves in a tight ideological and pedagogical spot. On the one hand, many are vocal advocates of diversity, actively working against discrimination both in their department and throughout the university and society. On the other hand, many may also view themselves as the primary defenders of and, perhaps accurately, as the primary providers of standard edited American English (SEAE) for their students (Davila 2012, 2016). The way they handle this tension in their grading and teaching is particularly relevant in addressing linguistic discrimination on campus for two reasons. First, at most universities, every undergraduate student is required to take an English composition class, creating a unique opportunity for instructors to provide explicit linguistic instruction to nearly the entire student body, typically at an early stage of their university studies. Second, because the broader populace often looks to English teachers as authorities on “proper” or “correct” English, these instructors “are in a unique position to change ideology surrounding language and possibly begin to break cycles of linguistic hegemony and standard language ideology” (Dunstan & Jaeger 2015, 789).

Given their unique position, the question of how English instructors view and address the presence of linguistic variation in their classrooms is an important one and has implications for instructors in all courses where writing is assessed. This chapter examines these issues and demonstrates the prevalence and deep embeddedness of standard language ideology both in instructors’ rating of an essay containing morphosyntactic features of African American English (AAE) and their attitudes toward student writing with “non-standard”¹ features. Specifically, I present original research which draws on survey data to investigate whether the presence of certain morphosyntactic (i.e. grammatical) features of African American English impact the rating of student essays at a large public research university in the South Central United States. I also analyze these instructors’ overt beliefs about the use of features from any stigmatized variety of English. I conclude the chapter with suggestions for addressing faculty awareness and treatment of linguistic diversity in writing assessment, and directions for future research into linguistic discrimination in student writing at the university level.

Standard Language Ideology

Standard language ideology (SLI) refers to the notion that there is one standard or correct version of a target language, which warrants greater prestige as a matter of “common sense” (for extensive discussion of SLI, see Lippi-Green 2012; Milroy & Milroy 2012). At this juncture, it is crucial to delineate between notions of a standard variety and a prestige variety, although the two are conflated almost universally by non-linguists and not uncommonly by linguists, as demonstrated by Milroy (2001). Milroy’s “non-ideological” definition of the term ’standard’ refers specifically to “the imposition of uniformity upon a class of objects” (531), in this case, a language variety. By this definition, it is not unreasonable to consider “academic English” a standard variety. “Prestige,” on the other hand, refers not to the language variety itself, but to the social value attributed to a language variety, although this can be considered a proxy for the social valuation of the users of a given language variety. In the context of this study, the most standardized variety, standard edited American English (SEAE), is typically seen as the most prestigious variety, especially in an academic context. However, it is essential to acknowledge that this variety is not necessarily awarded more prestige because of its level of standardization (a fact about the language system), but because of social facts about the users of that language system, and the individuals who have varying levels of access to or command of that system. While the standardization of SEAE may make an effective tool for clear communication in many academic contexts, this does not mean that it is more systematic or rule-governed than any other variety. Further, SEAE is not a superior form of communication in all contexts, academic or otherwise.

"Non-standard" Varieties in Education

The place of “non-standard” varieties in primary and secondary education in the United States received particular attention in the Oakland Ebonics controversy in the mid-1990s. In December 1996, the Unified Oakland School Board published a resolution legitimizing students’ use of their own variety of English in the classroom (in this case, AAE) and, further, calling for African-American students to receive instruction in their own variety whenever possible (for excellent overviews and analyses of the resolution and surrounding controversy, see Croghan 2000; Wolfram 1998). As those who study language variation would expect, the resolution received substantial pushback from all corners. Wolfram (1998) points out the incredibly unlikely public agreement of such politically divergent figures as conservative political commentator Kush Limbaugh and civil rights activist Jesse Jackson on the inappropriateness of AAE in public education, arguing that “such alliances are, however, simply a testament to the depth of the entrenched language ideology” (111). To many, if not most, in the public (Lukac 2018; Wolfram 2009) and in the realm of education, particularly English education (McBee Orzulak 2013; Weaver 2019), it is simply common sense that standard language is what ought to be used in the classroom and anything else is clearly inappropriate.

Despite substantial investigation of the implications for primary and secondary students who use “non-standard” varieties of English in the classroom, the treatment and experiences of university students who use “non-standard” varieties of English has received comparatively less attention, although research in this area has increased in the past decade. Dunstan and Jaeger (2015) and Holliday, Squires, and Lan (2019) have examined students’ on-campus experiences related to their “non-standard” language use and found that the standard language ideology held by faculty, support staff, and other students has a substantial impact on the academic experiences of university students who use stigmatized varieties, and that the students themselves are acutely aware of this stigmatization. While it is crucial to acknowledge the real-world effects of linguistic discrimination on students’ academic experience, for the purposes of the present study, we will focus on literature which addresses instructor treatment of student writing.

Johnson and VanBrackle (2012), employing a quantitative design, investigated the effect of “non-standard” features in student writing on the Regents’ Writing Exam, a holistically-scored one-hour writing assessment which sophomores in the University System of Georgia (in the U.S.) must pass in order to continue their university studies. The authors selected a Standard English essay which had been rated highly in a previous year and modified the essay by incorporating eight features of AAE (e.g. absence of third-person -s, zero copula), eight common ESL errors (e.g. subject-verb agreement, missing prepositions), and eight common SAE errors (e.g. pronoun number agreement, comma splices). This produced three essays in which the content and structure remains constant, with variation only in the morphosyntactic features mentioned above. The findings of their study revealed substantial linguistic discrimination against features of AAE; essays containing AAE features were 4.2 times more likely to receive a failing grade from raters than an essay with ESL errors, and 9.1 times more likely to fail than an essay with SAE errors (Johnson & VanBrackle 2012, 44).

These findings illustrate that students don’t just perceive their use of “nonstandard” features to negatively impact their academic experience as demonstrated in Dunstan and Jaeger (2015) and Holliday et al. (2019); use of these features in written work may, in fact, measurably and negatively impact students’ grades. Although all of the raters in their study were anonymous, Johnson and VanBrackle indicate that “the raters for the USG [University System of Georgia] Regents’ Writing Exam are required to have an M.A. in English or a graduate degree with 18 hours of graduate work in English,” and that “the majority of raters are part-time writing instructors at colleges and universities, but there are also full-time English faculty who participate” (15). These credentials and broad demographic information make it likely that several of these instructors are university English teachers, and many of the instructors in the present study have similar credentials.

Davila (2012, 2016) takes a qualitative approach to the analysis of instructor treatment of university student writing, forgoing any investigation of how these instructors rate student essays numerically and focusing instead on what features of student writing “index,” or suggest about particular author identities. Through interviews with instructors, she found that “instructors relied on their perceptions of written standardness for indexical clues” (Davila 2012, 198). Crucially, she demonstrates the bi-directionality of these instructors’ interpretation of marked linguistic features. That is, instructors use these features to index the author’s identity and draw upon the envisioned identity in their further evaluations of student writing. For example, when instructors identify linguistic features in the paper which led them to perceive a writer as White, they were often more willing to overlook “non-standard” features, viewing them more as silly “mistakes,” or to simply ignore them and focus more on the content of the paper. On the other hand, when a writer was perceived as being African-American, “non-standard” features were more likely to be described by the instructor as “errors” (Davila 2012, 197). These findings are essential to our understanding of “standardness” in student writing, as well as the way that university English instructors may view and evaluate student writing which contains “non-standard” features. Davila’s study does not attempt to answer is how these instructors’ attitudes play out in actual ratings of student writing, and that is the question pursued here.

Methods

Data collection in the present study involved an online survey distributed once in the spring of 2017 and once in the fall of 2018 to instructors of first-year undergraduate composition courses in the English department at a large public research university in the South Central United States, South Central State University (SCSU). This survey consisted of three sections. The first section requested demographic information including the residence history, teaching position, and teaching experience of the instructor. The second section presented a rating task in which participants were asked to holistically rate student essays (adopted from Johnson & VanBrackle 2012) on a scale of 0-4, the standard university grading scale. The final section explored participant attitudes toward the use of stigmatized varieties of English by asking instructors to comment on the “appropriateness” of students’ use of “non-standard” features in their writing and whether the use of such features should impact student grades. The attitudes questionnaire used Likert-scale ratings, but also included optional, open-ended follow-up questions (e.g. “Explain why or why not.”) to examine instructors’ reasoning behind the attitudes that they hold.

Essay Rating Task

The first task displayed one of two essays, adopted from Johnson and VanBrackle (2012), which are identical in content and vary only in that one essay includes morphosyntactic features of African American English (AAE). Recall that this essay, originally written in SEAE, was selected by Johnson and VanBrackle because it had received a “high pass” rating from a number of raters. The full text of the original and modified essays can be found in Johnson and VanBrackle (2012). To create the modified essay, Johnson and VanBrackle substituted eight morphosyntactic features of AAE into the original. These particular features are attested in AAE writing as well as speech (Smitherman & Smitherman-Donaldson 1986; Whiteman 1981) and were also selected on the basis that these “are well-known features of AAVE and rarely occur in the speech or writing of white speakers.” (Johnson & VanBrackle 2012, 37).

The modified essay in this study employed Johnson and VanBrackle’s modifications with one exception. One made by them (“There be”, invariant be), was deemed by the researcher and another experienced sociolinguist to be unnatural, at least in the context where it was used. The instance of invariant he was substituted for existential it, a well-known feature of AAE, e.g. “It’s a man in the other room.” Otherwise, the essay used in this study replicates that of Johnson and VanBrackle. The features used in the present study, along with the context in which they appear, are shown in Table 3.1.

Participants were informed that the essay was a sample of undergraduate writing which was written during a timed writing assignment and were asked to rate the essay holistically on a scale of 0—4, with 0 being “Poor” and 4 being “Excellent.” This scale was selected to correspond with the traditional university grading scale, in which 0 is an F and 4 is an A, and the prompt encouraged respondents to consider this fact when making their rating.

The final section of the survey aims to explore participants’ attitudes toward the use of stigmatized varieties. First, in order to understand the weight that instructors place on spelling, punctuation, and morphosyntactic features generally, participants were asked how important these elements are in student writing and what the highest possible grade would be (on a scale of A—F) for an essay which was high-quality in terms of content and organization but contained noticeable spelling, punctuation, and grammar errors. Next, participants were asked whether they feel that features of “non-standard” varieties should be marked as errors in student writing. Finally, instructors were asked about the perceived “appropriateness” of students’ use of “non-standard” features in their academic writing.

TABLE 3.1 AAE modifications to essays presented to instructors

Participants

Participants in this study (n = 34) were all current instructors in the English department at SCSU. The majority were graduate teaching assistants (30/34), although one assistant professor, two visiting assistant professors, and one lecturer also responded. All teach English composition courses for undergraduate students whose native language is English.³ 15 respondents identified as female, and 19 as male; nearly all identified as White (n = 29), three as American Indian or Alaskan Native, one as Black or African-American, and one as Hispanic or Latino. 18 indicated that they have been teaching at the university level for 0—3 years, 15 that they have taught for 4-10 years, and only one (a visiting assistant professor) that they have taught for more than ten years. Although the department does not maintain or provide demographic data about the body of first-year composition instructors as a whole, from my experience with this department, the sample in this study is representative.

Findings and Discussion

The findings of this study reveal the effects of standard language ideology on student writing, both in instructor ratings of student writing and in instructors’ conscious beliefs about the presence of “non-standard” features in student writing. First, I briefly describe the disparate treatment of the two essays, which vary only in the presence or absence of AAE morphosyntactic features. I then present the respondents’ explicit answers to questions regarding the appropriateness of the use of AAE features and compare those conscious attitudes to their actual rating of student writing.

Instructor Rating of Student Writing

Of the 34 instructors who completed the survey, 16 rated the unaltered (SEAE) essay and 18 the essay with features of AAE. A frequency distribution of the ratings of both essays is shown in Figure 3.2.

While the distributions of the scores are both relatively normal, the essays containing AAE features are generally rated lower, corroborating Johnson and VanBrackle’s (2012) finding that a holistically scored essay containing morphosyntactic features of AAE will receive lower ratings than an essay which does not contain these features. These results can be compared to instructors’ responses to the question of what the highest possible grade would be for a “high quality essay in terms of content and organization,” which contained “a noticeable amount of spelling, punctuation and grammar errors,” shown in Table 3.2.

Based on responses to this question, we would expect that if instructors perceive features of AAE as errors in spelling, punctuation, and grammar, the essay containing these features should receive mostly 3 (B) or 4 (A) ratings. Instead, the

FIGURE 3.2 Instructor ratings of student writing.

TABLE 3.2 Responses to “If a student wrote a high quality essay in terms of content and organization, but had a noticeable amount of spelling, punctuation and grammar errors, what’s the highest grade that student could receive?”

essay containing AAE features received mostly 2 (C) ratings, suggesting that either instructors are interpreting these features as errors and are harsher on such errors than they claim, or that these features are being treated more negatively than instructors might treat spelling, punctuation, and grammar errors.

Also of note here is that only the essay containing AAE features received a rating of 1 (a D in letter grade terms), although the AAE essay also received two ratings of 4 (A) while the unaltered essay received only one A rating. Both instructors who gave the AAE essay a rating of 4 (A) were graduate teaching assistants. One identified as “Hispanic/Latino” and indicated 0—3 years of teaching experience. Unfortunately, this respondent did not offer any elaboration on the attitude questions. The other respondent, who identified as “White” and indicated 4—10 years of teaching experience (including at several other universities around the country), did comment on the treatment of stigmatized features in student writing:

Besides the NOTE declaring that students have the RIGHT to use their own language (and this was a statement released in the 1970s) it is clear to anyonepaying attention that formal, academic English is a way systemic racism has sought to deligitamize [sic] minorities. Students work best with language when they are using their own language. Ownership is important.

As will be explored further in the next section, this level of support for the use of stigmatized varieties in student writing was rare in this sample, and only this instructor invokes the National Council of English Teachers’ (1974) statement on students’ right to their own language.

Instructor Attitudes toward Stigmatized Varieties in Student Writing

While the instructors in this study generally rated the essay with AAE features lower, their conscious beliefs about the use of stigmatized varieties in student writing would largely seem to predict the opposite; many indicate that student use of “nonstandard” features should not be marked as errors and is appropriate in student writing. However, a more detailed look at respondents’ justifications for these ratings reveals, that while some instructors do struggle with questions of standardness in student writing and where ideas of standardness come from, most instructors view “nonstandard” features as equal to errors in tenns of grading. Specifically, most respondents claim to be more concerned with the content and structure of a students’ writing than they are with what one respondent, referring to “non-standard” features, calls “surface level errors.” The attitude of these instructors toward “non-standard” features bears great resemblance to Davila’s (2012) findings that some English instructors perceive such features to be “silly mistakes” in White student writing, but “errors” in nonwhite writing (197).

Treatment of "Non-standard" Features as Errors

When asked whether features of stigmatized varieties should be marked as errors in student writing, most respondents indicated that they should not (22 of 34 respondents said either “probably” or “definitely” not), with five respondents selecting “probably yes” and only one respondent selecting “definitely yes.” A table of responses regarding the treatment of features of stigmatized varieties as errors in student writing is shown in Table 3.3.

TABLE 3.3 Responses to “Should features of non-standard varieties (e.g. African American English, Chicano English, Southern English, etc.) of English be marked as errors in student writing?”

Definitely yes Probably yes I’m not sure Probably not Definitely not

Because these rating questions were followed by an opportunity for instructors to “explain why or why not,” we can evaluate instructors’ reasoning behind these responses. Many respondents who indicated that “non-standard” features should probably not or definitely not be marked as errors suggest that such features are indeed errors, but that they prioritize content over mechanics in student writing. That is, these respondents consider “non-standard” features to be errors, but are willing to overlook those “errors” if the paper’s content is sound. For example, one respondent, a TA with 4-10 years of teaching experience, who gave the “non-standard” essay a rating of 2 (C) indicated that “unless these issues are part of the class’s content, these surface level errors should not factor into the evaluation of the students overall grade.”

Despite most instructors’ view of “non-standard” features as errors, a few instructors who indicated that these features should not be marked as errors acknowledged the problematic nature of referring to “non-standard” features as “errors,” and questioned their role as an instructor in providing students with knowledge of academic (standard) expectations. One respondent in particular, a TA with 4-10 years of experience, who also rated the “non-standard” essay a 2 (C), notes the potential for instructors to empower students’ use of their own language:

I’m struggling with the word “errors” here. The non-standard features should be marked as non-standard so that the student will know what white middle-class audiences will expect, but some work will hopefully have been done to help students understand how power structures work through writing and to legitimize non-standard dialects. In that context, the paper can be marked so that the student knows better how to conform or resist.

This instructor’s statement echoes Davila’s (2016, 18) suggestion that English instructors are in a unique position to support students whose language is generally stigmatized by framing conversations about “standardness” in a way that does not evaluate students’ language, but instead equips them to understand the power structures surrounding expectations of “standardness.”

Notice that, despite a supportive attitude, this instructor still gave a relatively low rating of 2 (C) to the essay containing non-standard features. One possible explanation for this is that instructors simply may not be aware of what other varieties of English look like and can only label them as errors. This suggestion is borne out in the data here as well, as some instructors who feel that “non-standard” features should not be treated as errors explicitly state that they don’t know what the features of other varieties are and, therefore, don’t know how to talk to students about them. For example, one respondent, a TA with 0—3 years of experience who gave the “non-standard” essay variant a rating of 3 (B) and indicated that “non-standard” features should definitely not be treated as errors claims that “I don’t think they should but I wouldn’t necessarily have the knowledge to distinguish between an error and a non-standard use.” In addition to similar responses from several instructors in this study, at an earlier presentation of this research, I was surprised by the number of linguists who indicated that they had never heard of some of the AAE features used in this study, most notably expletive it (e.g. “It’s many reasons”).

I now turn to the responses of a few instructors who feel that “non-standard” features should probably or definitely be marked as errors in student writing. These responses lean heavily on notions of standard language ideology, and on the belief that university English instructors have a responsibility to teach students the inappropriateness of their own language variety in academia. One respondent, a TA with 0-3 years of teaching experience who rated the “non-standard” essay a 2 (C) and indicated that features of “non-standard” varieties should probably be marked as errors, highlights the perceived responsibility of English instructors to provide students with SEAE: “We tend to use a standardized English. You never know who the student will be communicating with in the future. If we don’t teach them, who will?” While this response does not overtly express negative attitudes toward the use of students’ own variety of English in their writing, it does draw on the standard language ideology assumption that the “standardized English” that “we” use is superior, or at least, “inevitable” (Davila 2016, 133) and will be used by anyone important that “the student will be communicating with in the future.”

The most overtly negative example of an instructor’s attitudes toward students’ use of “non-standard” varieties of English comes from the sole respondent who indicated that “non-standard” features should definitely be marked as errors. This instructor is a TA who identified as White and male, with 0—3 years of teaching experience, and gave the standard essay a rating of 2 (C):

Unless the assignment is free-form and creative, it needs to adhere to the normal conventions. No one on the hiring committee of a big law or accounting firm is going to appreciate it as expressing the applicant’s individuality.

Two things stand out in this response. First, the instructor’s use of “conventions,” as it appears in this context, is addressed directly by Davila (2016), who argues that this term is often a code for “standard” (and therefore good) grammar and mechanics. Second, the instructor apparently views “non-standard” language use not as a right of the student, or as an effective form of communication, but as an overtly non-conformist expression of individuality (as suggested in the term “free-form”). This instructor apparently takes as a given the students’ prior access to standard English and the “conventions” of such genres as are valued by “the hiring committee of a big law or accounting firm,” thereby “positioning the standard as superior and accessible” (Davila 2016, 17). Through the standard language ideology framework evoked by this instructor, where access to SEAE

is assumed, the students’ own varieties are viewed (and treated in rating) not as legitimate methods of communication in academia, but as errors at best, and as openly transgressive at worst.

Appropriateness of "Non-standard" Features in Student Writing

When asked about the appropriateness of the use of “non-standard” varieties in student writing, instructors responded even more positively overall, and a table of instructors’ responses regarding the appropriateness of the use of a stigmatized variety in student writing is shown in Table 3.4.

Only one respondent indicated that the use of a student’s own variety in their writing would be inappropriate to any degree, and selected “Extremely inappropriate.” This was the same instructor described at the end of the previous section, who also indicated that “non-standard” features should definitely be treated as errors. In their follow-up answer, this instructor reiterated the inappropriateness of any variety but the standard and compared students who use a “non-standard” variety to literary choices or second language learners:

Unless you are writing in that style or genre, standard English is the order of the day. Those PhD L2’s know it, and that’s why they spend lots of time it the Wei ting [sic] Center trying to standardize their English.

It is also worth noting that the wording of this question gives examples of “nonstandard” varieties: “(e.g. African American English, Chicano English, Southern English, etc.).” This instructor’s comparison of students who are first language speakers of “non-standard” varieties of English to second language learners of English further highlights the instructor’s view of “non-standard” varieties as deficient, at least at the academic level, and as something to be remedied.

This instructor’s response, however, was an outlier, and most instructors indicated more positive views about students’ use of a “non-standard” variety. Despite these more positive selections, instructors primarily referred to their previous answer (e.g. “See my answer to the previous question”) when responding to the follow-up question “please explain why or why not.” This suggests that many instructors conflate whether or not a feature should be treated as an error in student

TABLE 3.4 Responses to “How appropriate is it for students who speak a non-standard variety of English to use features of their variety in their academic writing?”

Extremely Somewhat Neither appro- Somewhat Extremely

inappropriate inappropriate priate nor appropriate appropriate

inappropriate

15

writing with its appropriateness, a concept which I explore further below. Instructors who gave new answers to this follow-up question displayed relatively positive views about students’ use of their own language. One indicated in her answer for the “errors” question above that she was typically more “lenient” with students who use “non-standard” varieties. This instructor is a TA with 0-3 years of teaching experience, and rated the essay with AAE features a 2 (C):

It would be racist to say that those students should not use features of their variety in academic writing, and though I have no problem with it in my own classes, I know people outside of academia are not so open-minded about receiving communication that includes a non-standard variety of English. It’s something I would never discourage, and something I struggle with.

Notice that although this respondent explicitly references that suppression of an individual’s variety “would be racist,” they also gave the same grade (C) to the essay containing AAE features as the instructor who indicated that the use of “non-standard” features in student writing was “extremely inappropriate.” Furthermore, the instructor who takes a more positive approach largely draws upon the same tenets of SLI displayed by the instructor above who indicated that students’ use of their own variety is extremely inappropriate. In particular, the concern that “people outside of academia are not so open-minded about receiving communication that includes a non-standard variety of English” and the fact this is something the instructor “struggle[s] with,” highlights the tension that this instructor feels between seeking to avoid discrimination and feeling pressure to provide students with access to SEAE.

The fact that more instructors felt comfortable marking features of “non-standard” varieties as errors than felt comfortable describing students’ use of their own variety as “inappropriate” may provide some insight into the lower ratings given to the essay containing AAE features; while most instructors are aware of the problematic nature of restricting students’ use of their own variety, they may feel that their “duty” as a composition instructor requires them to view the student writing as error-filled, and to rate or grade the writing accordingly.

Conclusions and Implications

The findings of this study suggest that although university English instructors are, overall, consciously well-intentioned about linguistic diversity, they are often nevertheless strongly influenced by standard language ideology which may manifest itself in grade reductions for students who use “non-standard” varieties of English. There are several potential reasons for the lower ratings of the essay containing features of AAE. First, it is possible that these instructors recognized features of AAE and downgraded that essay intentionally but made positive statements about stigmatized varieties under the pressure of political correctness.⁴

However, given these instructors’ generally positive comments about students’ use of their own variety and the efforts of many of these instructors to promote and protect diversity on campus, I find this explanation unlikely, at least for this group of instructors.

A second possibility is that these instructors are aware, perhaps subconsciously, of the presence of these features as indicative of a “non-standard” variety of English, and they assign a lower grade accordingly. This hypothesis appears to be supported by the fact that the modified essay (but not the “standard” essay) received lower ratings than instructors claim they would give an essay which is high quality in terms of content and organization but contains spelling, punctuation, or grammar errors. This is also supported by the number of instructors who suggest that “non-standard” features should be treated as errors in student writing. Indeed, even some instructors who indicate that such features should definitely not be treated as errors nevertheless indicate that they are errors, rather than features of a legitimate variety of English. A third possibility, which may coexist with the second, is that these instructors simply don’t have what one respondent described as “the knowledge to distinguish between an error and a non-standard use.”

Implications for Practice

This study corroborates findings of recent studies (e.g. Davila 2012, 2016; Weaver 2019) in showing that many university English instructors, even those who hold positive attitudes toward students’ use of their own language variety, maintain ideals of standard language ideology, and draw on elements of what Davila terms “Standard Language Discourse” when discussing the treatment of student essays with “non-standard” features. However, similar to the findings of Weaver’s (2019) study, many of the instructors in the present study appear to struggle with these issues, finding themselves in a position to support students’ use of their own variety, but feeling a pressure, or perhaps a sense of duty to enforce students’ use of the standard language variety.

In order to begin proactively addressing these concerns, Dunstan and Jaeger (2015) suggest that it is essential to begin “educating faculty, staff, and administrators on language diversity'” (796), an endeavor which Stephany Dunstan, Walt Wolfram, and others have spearheaded with substantial success at the University of North Carolina (see Dunstan, Wolfram, Jaeger, & Crandall 2015). Given the fact that almost all of the English instructors who participated in this study held strong beliefs about the empowerment and encouragement of students’ use of their own variety, this population appears particularly ready to receive and apply education about linguistic diversity. These instructors can be early, powerful advocates for students who use stigmatized varieties and can also help to dispel myths about language variation across campus and among the general public, who often look to English instructors as the final authority on language matters.

However, Weaver’s (2019) study demonstrates that the deep embeddedness of standard language ideology, particularly in an English education setting, means that this important work is not as simple as offering one-time professional development workshops. While the instructor in her study was able to more clearly articulate her thoughts on standard language after language awareness workshops, she, much like many of the instructors in the present study, nevertheless continued to struggle with multiple ideologies about the place of standard language in the composition classroom. Therefore, Weaver argues, writing pedagogy education and other professional development work which focuses on linguistic variation should be implemented rigorously and regularly in order to best support faculty, overcoming “the deeply embedded ideology ofSLI” (Weaver 2019, 50).

Implications for Future Research

The findings of this study, which align closely with the few other studies on the matter, highlight the need for continued research into the effects of stigmatized varieties on instructors’ treatment of student writing at the university level. Although discrimination on a linguistic basis at the university level is clearly an issue, to my knowledge, Johnson and VanBrackle (2012) is the only other study which quantitatively investigates the effects of “non-standard” features on instructor rating of university student writing. Davila’s work investigates the ideologies underlying university English instructor interpretation of students’ voice, and, importantly, highlights the role that standard language ideology plays in that interpretation. However, she did not ask instructors to actually rate the essays they evaluated, and thus was unable to investigate the effects these ideologies might have on ratings. The present study attempts to begin filling this gap in the research on standard language ideology at the university level by focusing on a specific population of instructors, and by investigating the relationship between their ratings of student writing with “non-standard” features and the ideologies these instructors express regarding the use of “non-standard” varieties in student writing.

There are several ways in which the present study could be developed and improved in future research. First, the sample size is small, and is comprised only of first-year composition instructors in the English department at one university. This limited sample does allow for a focused investigation of English instructors, a population for whom the treatment of “non-standard” varieties of English is especially important, as described above. However, students interact with and receive grades on writing assignments from instructors in a broad range of other disciplines during their university career. Future research should broaden the sample to include instructors in other departments across the university.

In addition to the limited sample size, this study takes a quasi-experimental approach to investigating instructor essays. Although the modified essay in this study includes relatively salient features of AAE and features which are unlikely to be White student errors (e.g. existential if), it is not an authentic example of AAE writing, and instructor treatment of an authentic AAE essay may very well prove different. Furthermore, this study only investigated instructor treatment of features of one stigmatized variety. There are, of course, students who speak many other stigmatized varieties at universities, including Chicano and Appalachian English. It may be the case that attitudes toward one variety or another may vary based on region, the percentage of students in a university population who use a certain variety, or other factors. Future research should incorporate other varieties, especially varieties which are more prevalent at a particular university (e.g. stigmatized regional varieties).

This study represents an important expansion of our awareness that university instructors, including those who teach language and promote diversity both in and out of the classroom, are not immune to the effects of standard language ideology and linguistic discrimination, and that this may result in meaningful effects on students who use “non-standard” varieties. Despite the prevalence of SLI in these results, this study has also found that many English instructors are aware of and hold positive beliefs about the use of one’s own variety in academic writing and have considered how to teach users of stigmatized varieties to write in academia. This is an encouraging finding and one which suggests that English instructors are prepared to be a powerful force in dispelling the standard language ideology so prevalent at the university level and more broadly in American culture, a force that sociolinguists can strengthen and support through dialect education.

Notes

• 1 In this chapter, I use the tenn “non-standard" in quotation marks to refer to varieties other than “Standard American English” (SAE). Davila (2016) suggests the use of the labels conventional and unconventional to describe particular language practices within certain contexts. Davila’s terminology is an excellent pedagogical strategy and one that I myself employ in teaching composition and linguistics courses. Indeed, classification of varieties besides SAE—which Young (2007) refers to as White English Vernacular—as “non-standard” can certainly evoke some negative evaluations or be seen as suggesting that these varieties do not share a place in education alongside SAE. However, this valuation is not at all my intention, and I use the term “non-standard" only descriptively and in contrast to SAE.
• 2 Milroy (2001) points out the problematic tendency of sociolinguists to use “the ‘standard’ or ‘prestige’ or ‘careful style’ variant as the unmarked term in labeling well-known variables; for example: copula deletion, final stop deletion” (533). Such labelling inevitably suggests that users of other varieties are in some way detracting from the unmarked, standard, or correct variety, an underlying tenet of standard language ideology. While I recognize the problematic nature of such labels in describing varieties of English, I use terms such as absence of third-person -s in this particular context to describe the African American English modifications made to the original essay, written in Standard American English.
• 3 Non-native speakers of English are enrolled in a separate but parallel freshman composition program focused on ESL students and taught primarily by graduate teaching assistants in the TESL/Linguistics graduate programs.
• 4 I thank an anonymous reviewer of this work for the suggestion of this possibility.

References

Croghan, M. (2000). History, linguistic theory, California’s CLAD, and the Oakland Public Schools Resolution on Ebonics: What are the connections? World Englishes, 19 (1), 73-87.

Davila, B. (2012). Indexicality and “standard” edited American English: Examining the link between conceptions of standardness and perceived authorial identity. Written Communication, 29 (2), 180—207.

Davila, B. (2016). The inevitability of “standard” English: Discursive constructions of standard language ideologies. Written Communication, 33 (2), 127—148.

Dunstan, S. B., & Jaeger, A. J. (2015). Dialect and influences on the academic experiences of college students. The Journal oj' Higher Education, 86 (5), 777-803.

Dunstan, S. B„ Wolfram, W., Jaeger, A. J., & Crandall, R. E. (2015). Educating the educated: Language diversity in the university backyard. American Speech, 90 (2), 266-280. doi: 10.1215/00031283-3130368.

Holliday, N., Squires, L., & Lan, L. (2019). "It’s not linguistically accessible to people like me’’: The linguistic experiences of black students at HWIs. Georgetown University Round Table, Washington, DC, March 31.

Johnson, D., & VanBrackle, L. (2012). Linguistic discrimination in writing assessment: How raters react to African American “errors,” ESL errors, and standard English errors on a state-mandated writing exam. Assessing Writing, 17 (1), 35—54. doi:10.1016/j. asw.2011.10.001.

Lippi-Green, R. (2012). English with an accent: Language, ideology and discrimination in the United States. New York: Routledge, doi: 10.4324/9780203348802.

Lukac, M. (2018). Grassroots prescriptivism: An analysis of individual speakers’ efforts at maintaining the standard language ideology. English Today, 34 (4), 5-12.

McBee Orzulak, M. J. (2013). Gatekeepers and guides: Preparing future writing teachers to negotiate standard language ideology. Teaching/Writing: The Journal of Writing Teacher Education, 2 (1), 5.

Milroy, J. (2001). Language ideologies and the consequences of standardization. Journal of Sociolinguistics, 5 (4), 530—555. doi:10.1111/1467-9481.00163.

Milroy, J., & Milroy, L. (2012). Authority in language: Investigating standard English. New York: Routledge, doi:10.4324/9780203124666.

National Council of English Teachers. (1974). Students’ right to their own language. College Composition and Communication, 25 (3), 1-18.

Smitherman, G., & Smitherman-Donaldson, G. (1986). Talkin and testifyin: The language of Black America (Vol. 51). Detroit, MI: Wayne State University' Press.

Weaver, M. M. (2019). “I still think there’s a need for proper, academic. Standard English”: Examining a teacher’s negotiation of multiple language ideologies. Linguistics and Education, 49, 41—51.

Whiteman, M. F. (1981). Dialect influence in writing. Writing: The Nature, Development, and Teaching of Written Communication, 1, 153-166.

Wolfram, W. (1998). Language ideology and dialect: Understanding the Oakland Ebonics controversy. Journal of English Linguistics, 26 (2), 108—121.

Wolfram, W. (2009). Dialect awareness, cultural literacy, and the public interest. In Eth-nolinguistic Diversity and Education (pp. 141-161). New York: Routledge.

Young, V. A. (2007). Your average nigga: Performing race, literacy, and masculinity. Detroit, MI: Wayne State University Press.