Administrating Language: The Language Ideological Voices of Urban School Administrators in Urban Education

Anti-Racism and School Leadership

While much of school policy is set at the state or district level, front-line building management is critical in setting up school wide practices to promote an instructionally sound and socially just education. Administrators in urban schools, specifically principals, bear a large portion of the burden in setting policies and shaping school culture. A principal’s position is powerful and includes a variety of roles, including the public face of the school, management of staff, and instructional leadership (Lortie, 2009). Beyond management abilities, a principal must have an understanding of the linguistic and cultural contexts of their school communities. Stovall (2004) noted that the position of school leader is contested as the role stands in the space between teachers, the public, and higher administration. While administrators may receive some education on cultural diversity, it is not guaranteed. Survey data collected and analyzed by Haar and Robicheau (2008) noted that a vast majority of principals and superintendents thought that institutions of higher education should provide more training on cultural diversity for administrators, although nearly two-thirds had received some training through their districts.

In 2019, Urban Education published a special issue entitled “Negotiating the politics of anti-racist leadership: The challenges of leading under the predominance of whiteness” (Welton et al., 2019). In this issue, several papers were published describing the ways in which school leaders confront White Supremacy and Whiteness in policies that affect schools. A major implication of these studies was to include explicitly anti-racist instruction in school leader education (Diem et al., 2019) and for school leaders to be self-critical (Allen & Liou, 2019). How school leaders may develop anti-racist language policies within their school, though, merits further exploration.

In New York City schools, it was found that widespread dismantling of bilingual programs following a shift to more restrictive US language policies appeared to be tied to lack of formal training in linguistics or language diversity (Menken & Solorza, 2014). If administrators in linguistically diverse schools have not been prepared well for the linguistic diversity they will encounter, then it would seem that specific education on AAL is likely also absent from administrator development. Administrators need to be equipped with the knowledge of linguistic practices and histories of African American students in order to understand how linguistic racism continues to play a role in the marginalization, discrimination, and the blatant killing of unarmed African Americans (Baugh, 2015) and how language is vital to achieving educational equality and an education system in which Black Lives Matter (see Garza, 2014).

Language Attitudes toward AAL

While there is little information on the attitudes of administrators toward AAL, a brief look at the literature on AAL over the past three decades reveals much about how teachers view AAL. A dominant approach taken up by researchers to study language attitudes has focused on the individual and group reactions to AAL language structures (Bowie & Bond, 1994; Labov, 1968; Mitchell-Kernan, 1972). A number of dissertation studies (Boyd, 1996; LeMoine, 2003; Mackey, 1999) sought to explicitly elicit teachers’ and preservice teachers’ views of AAL by drawing on the language attitude scale (LAS) by Taylor (1973). Overall, these studies reported slightly negative views toward AAL, though results were mixed. Champion et al. (2012) used the LAS to measure the attitudes of 132 preservice teachers toward AAL, revealing that 85% of the students held negative attitudes and noted a significant effect of race of respondent on the attitude.

While some studies have found that multicultural or linguistics classes had little impact on teachers’ attitudes (Abdul-Hakim, 2002; Boyd, 1996; Mackey, 1999), others reported that specific training in AAL could have an impact on teachers’ attitudes (Bowie & Bond, 1994; LeMoine, 2003). One study notably mentioned the correlation of school philosophy or school culture to teachers’ language attitudes (Blake & Cutler, 2003).

In a scalar interpretation of attitudes, the nuance of language ideological stances are lost. Within the context of robust qualitative responses, language ideological stances can be inferred to establish whether participants align with deficit, difference, resource, or language rights perspectives toward AAL. With this study extending the notion of attitudes to school philosophy and policy, a deeper understanding needs to be gathered about attitudes and practices of those who help mold the philosophies and policies of particular schools, namely school leaders. School leaders are critical in setting the policies and structuring practices in schools. They not only set school-wide policies, but they also are integral in community building, deciding curricula, hiring faculty and staff, and instructional leaders (Darling-Hammond et al., 2007). Thus, understanding the language ideologies of school leaders may yield helpful information for education and professional development practices.

Use of AAL in Schools

Numerous scholars have called for the use of AAL in classroom practice (e.g., Alim & Baugh, 2007; Baker-Bell et al., 2017, Young et al., 2018). AAL scholars continue to echo the calls made decades ago for student’s rights to their own languages (Baugh, 2015; Smitherman, 1995), and have begun to document the important role AAL plays in the construction of communication and identity across ethnic lines in vastly changing urban schools (Paris, 2012). Baker-Bell et al. (2017) show how teachers understanding of Black Language can be enhanced using contemporary topics such as the Trayvon Martin killing along with linguistic concepts that develop Black Language consciousness. Unfortunately, these practices have rarely been entrenched in principal preparation. Often speakers of AAL are assumed to have a language disorder or a deficit because of the linguistic differences (Seymour et al., 1998). The speech-language pathology literature has noted an over representation of African Americans in special education because of the linguistic differences (Baugh, 1995, 2015; Robinson & Stockman, 2009). The use of AAL is often viewed as a deficiency of the speaker in particular because of the standard language ideology (Lippi-Green, 2000a).

Language Ideologies

Language ideologies (LI) have been defined as “any sets of beliefs about language articulated by users as a rationalization or justification of perceived language structure or use” (Silverstein, 1979, p. 193). They may be explicitly stated or tacitly underlie the decisions we make regarding language. They are manifested and enacted through social practices, including educational policies as well as in classroom learning activities (Razfar & Rumenapp, 2013). LI, as research, is not focused on the structure of language as such, but rather on the way those structures are viewed and how they organize society. This can be seen in competing interpretations of the linguistic features of AAL. Different ideological stances would presume AAL is a slang, dialect, or language depending on the ideological footing. If the form is viewed, as it often is, as “incorrect” or “non-academic” in relation to a presumed standard, it is seen as a deficit of the speaker. Other interpretations of the same form of language could, on the other hand, be viewed merely as a code that differs from another code such as a dialect, a potential resource to facilitate social interaction, or an essential demonstration of a human right to have personal autonomy in linguistic choices. Each of these orientation play a role in the way our data is interpreted below. LI necessarily includes the broadest levels of context within the unit of analysis in which some ideologies may be considered hegemonic but not determinant. In any given context there are often competing ideologies (Razfar, 2012). For example, teachers are often aware of the sociopolitical implications of such practices as school reform (Gitlin & Margonis, 1995) and mandated literacy curricula (Pease-Alvarez et al., 2010) and can assert agency to resist or negotiate these political acts.

LI, as the common sense view of language, mediate policy decisions and national talk about language (Gonzáles & Melis, 2000, 2001). Lippi-Green (2000b) points out that the definition of AAL is highly contested. For instance, she explains that AAL is used as a marker of resistance to mainstream White culture or to build solidarity with fellow African Americans, yet some notable African Americans reject it as a failure to fully engage in the social and political structures of the United States. Speakers of AAL are often cast in an inferior position, no doubt reinforced by the intersection of language and race. The issue of linguistic hegemony is often a proxy for wider racial inequalities.

Society’s common sense notions of language tend to view the issue of language diversity as a neutral inhibitor to social progress or unity and veil a racialized society. Hill (2000) writes “Contemporary racist culture, however, is reproduced especially through practices which are never condemned as racist-practices, which appear (at least to many Whites) to constitute mere common sense” (p. 246). This is also seen in “standard language” ideologies in which a person or institution in authority frames a certain “form” of language as “correct” or “standard” and the myth that language policy is truly about only “standard language” and not the cultural aim for the monoglot ideal (Silverstein, 1996), of which schools and school policy is an important tool for policing.

Given the unique role of school administrators who are managers that influence local school policy and often serve in roles in establishing practices, such as the role of instructional leader (Lortie, 2009), we assumed that part of their role, whether recognized explicitly or not, would be to administrate language policy. To the extent that the administration sets the policies, programs and pedagogies likely follow. We also assumed that school leaders play a large role in setting up school culture and influencing classroom practices. School leaders are often the ones who evaluate teachers, weigh in on curriculum, and even decide professional development. Their role is uniquely situated as a policy maker and practitioner.

At the time of our initial conceptualization, the term raciolinguistics (Flores & Rosa, 2015) had not been popularized, but the literature on language and race was extensive. From a policy standpoint, it has long been noted that institutions recognizing discrete linguistic codes as merely “different,” would inevitably result in a deficit positioning of one language under another (Smitherman, 1995)—no doubt due to White Supremacist ideologies (Shedd, 1969). Indeed, Baker-Bell (2020) elaborates on this point in her framing of Antiracist Black Language Pedagogy. She outlines three different orientations to pedagogy—eradicationist, respectability, and antiracist. It is only the last set of pedagogies that “critically interrogate white linguistic hegemony and Anti-Black Linguistic Racism” (p. 28) which holds the power for transformative change in schools. Therefore, the framing of this research adopted a lens sensitive to not only linguistic difference, but how through that difference language is raced and how race, on the converse, is languaged (Alim et al., 2016). We, therefore, seek to understand how urban school administrators view AAL and the use of AAL in their schools. Administrators hold positions of power in the schools, often setting language policy both tacitly and explicitly. These questions will lead us in addressing more complex issues of multiculturalism and linguistic diversity in urban schools, specifically in regards to how school leadership is conceptualizing the issue.

Data Collection

Participants were recruited from a local doctoral program at an urban university. The Doctor of Education (EdD) program has an explicit focus on social justice and a commitment to developing administrators, thus provided an appealing population from which participants were drawn. Students in this program are administrators in urban schools. One of the research assistants video-recorded semi-structured focus group interviews at a local university campus. Questions were prompted from a list of protocol questions and follow-up with probing questions and dialog among participants. Focus group interviews lasted between 45 min and 1 hr.

Participants

Each focus group ranged from two to five participants each with a total of 15 participants included in the study. The focus groups were created as participants responded; therefore, they were unintentionally divided by race. Three focus groups included 11 participants, all identifying as White, with one identifying as White and of Hispanic descent. The other two focus groups included four participants identifying as African American, with one identifying as a Black immigrant African American. Most of the administrative experiences of all participants were centered in majority Black or Latinx schools. Online Appendix 1 shows the position and educational background of participants. All participants were enrolled in the EdD program at the time of the study. Since these students were assumed to have been exposed to topics including English Learners (ELs), race, culture, and class issues, as well as to have had experience in diverse classrooms and/or schools, the dynamics provided a unique context to conduct focus group interviews. Many of the participants had previously had classes together and were, therefore, familiar with each other’s schools and interests. Others, however, were meeting for the first time. Pseudonyms are used for participants’ confidentiality.

Data Analysis

Developing codes and analytical tools for explicit focus group data, proved to be challenging since administrators stand in the crux between policy and practice. From a policy perspective, Baker (2011) and Ruiz (1984) discuss the language as a problem, right, or resource categories. Similar to deficit, language as a problem indicates a negative viewpoint, though it may be applied as either an individual or a social problem. The categories of a right, resource, and problem had been applied to semi-structured interviews in social work (Harrison, 2007), and thus were adopted as an initial conceptual tool. We approached the data with an understanding that when White people in professional and public settings talk about language, especially at the intersection of race, they may attempt to shift toward an assumed political correctness (Jackson, 2008) or attempt to not be viewed as racist since “Whites are more likely to be penalized (primarily by other Whites) for bringing race up in a social justice context than for ignoring it” (DiAngelo, 2011, p. 63). Thus, we allowed for a category of “language as difference” to capture this metalinguistic move and then interpreted it within the sociopolitical context of educational administrators in a setting of peers.

We analyzed the focus groups to derive themes about which the administrators discussed, specifically looking at the different relationships the administrators considered (Lortie, 2009). Administrators discussed a variety of different relationships, including their relationship with students, parents, and teachers. Additionally, we decided to employ discourse analysis to examine how the participants talked about AAL its use in schools.

Once these instances were identified (111 instances), they were coded as “deficit,” “difference,” “resource,” or “right.” AAL has been noted to be the basis for negative judgments of the speaker, hence a “deficit view.” A “difference view” would indicate the recognition of structural differences between AAL and White Mainstream English (WME), but not necessarily carry a homogenous value judgment negatively or positively. Language as a “resource” view would indicate that the participant saw AAL as a positive, value-added language structure. That is, that by using AAL it afforded access to cultural or social capital. The language as a “right” view implies advocacy in which AAL should be allowed and be used since it’s a fundamental human right. Joseph and Zayoni coded using these a priori categories across the data set yielding 67% agreement. After recalibrating and redefining the categories with the definitions in Table 1, the entire set was recoded independently and 96% agreement was achieved. Table 1 displays the coding definitions and examples.

OPEN IN VIEWER

Table 1.

Coding Definitions.

View	Definition	Examples
Deficit	The statement considers the differences between AAE and SAE but also the need to “fix” or “change” it. This is different than code switching because it assumes there is something inherently wrong with the language.	If you look at let’s say there’s a student that does use VERY STRONG Ebonics, like to the point where you almost like you can’t understand him or her, right? (White administrator)
Difference	The statement considers differences between AAE and SAE as in “code switching,” but does not take a stance on whether one language is “better” or another language is “wrong”. The comment may reference negative views by society, but the agency of the speaker does not endorse the negative view.	Exactly and even in elementary school when they say a word incorrectly I correct them you know and I tell them its nothing wrong with the way you said that word when you’re talking to your friends but when you’re talking to me or when you’re talking to your teacher you need to be talking in the standard English (White administrator)
Right	The statement takes an explicit stance taken to “value” the language. Additionally, it can be considered a resource if it is being done by teachers to “help” students or by students to access some other form of social status or power	So if a child is thinking in Ebonics, or thinking in, you have five languages you may be thinking in a more EXPANSIVE way than a person who only knows one language. Which is, and I even ask kids why do you think that language is a requirement for college, because for about two or three years it wasn’t a requirement (White administrator)

It is important to note that a “difference” viewpoint may be part of a larger deficit model, thus we coded the instances recognizing that one participant may have an overall deficit view but may also express a “difference” view for part or majority of the time. Even though a statement was considered a “resource view,” the attitude of the participant could overall be deficit. For example, if the participant claims that it is good for students to be able to speak AAL to communicate with their families then the statement would be coded as a “resource view.” However, this same participant could have a derogatory attitude toward speakers of AAL in general. By coding for both constituent group and language attitude, we were able to understand if the administrators viewed the use of AAL by certain speakers differently.

After analyzing different views of AAL, we coded the names administrators gave to AAL and WME. The focus group was prompted with the term “African American Vernacular English” and the interviewer attempted to be consistent using this term, unless elaborating on the possibility of other names or responding to one of the participants who used another name.

Next, we coded for discourse markers that demonstrate a type of epistemic modality (Lyons, 1977), or discomfort around the naming practices. Building off of Lyons (1977), we defined these particles as:

“Any utterance in which the speaker explicitly qualifies his[/her] commitment to the truth of the proposition expressed by the sentence he utters, whether the qualification is made explicit in the verbal component (. . .) in the prosodic or paralinguistic component, is an epistemically modal or modalized utterance” (p. 797).

We closely analyzed the transcript around each of the instances in which AAL or WME were named. We looked for qualifications of the naming vis a vis a truth proposition. The nominalization of the language indicated a definitive stance, and any qualification would indicate some uncertainty.

Here are examples of instances of modal shifts indicating uncertainty (underlined):

*example regards WME not AAL.

Researcher Positioning

Aria was the principal investigator for this project and worked closely with a graduate assistant (Joseph) to conceptualize the study and to develop the interview prompts. Aria identifies as a non-dominant Middle Eastern “White” male although he questions whether his kind should be classified as such and prefers the more historically accurate Afro-Asiatic.

Joseph identifies as a White male. At the time of the study, he was a PhD student. Through the initial participant recruitment and data collection phases, he interacted as a peer with students in the EdD program. In general, the EdD students understood their focus and purpose was in building data based leadership in urban schools, while PhD students produced new research. We established rapport by discussing similarities and differences across the programs. In the first three focus groups all participants identified as White. This shared racial identity with Joseph seemed to allow for the conversation to flow as it would among White peers discussing issues about race, which, in professional settings, can often be euphemized so as not to appear racist (DiAngelo, 2011). In the second two focus groups, all participants self-identified as Black or African American women. Since the participants in these two focus groups also had extensive understanding of the research process and, along with Joseph, were students, participants seemed forthcoming in the discussions of race in ways that may not occur outside of a controlled research environment.

Zayoni identifies as female, LatinX and was a PhD student at the time of the study. Her background is in feministic perspectives to explore ideological stances. Her role was to assist with the analysis of the data set. Given the shared and varying experiences between her and Joseph, this allowed for fruitful discussions around coding and analysis of the data corpus.

Limitations

While this study attempted to solicit diverse perspectives of educational administrators toward AAL, the sample was selective. We recruited at one university’s EdD program with an urban mission. Participants were invited through e-mail, faculty, and short recruitment presentations. The perspectives presented here represent a sampling of progressive educators with an urban focus rather than the general population of educational administrators. This sample represents educational leaders who have experiences, affinities, and coursework that include some reflection and thinking about the topics under investigation.

Administrator Views of AAL

Administrators negotiate multiple roles in schools. Often, they are instructional leaders as well as managers. Many administrators come from the classroom themselves and consider themselves teachers at heart. Thus, as we investigated the participants’ views of AAL, there was high variation within the way AAL was perceived. All statements about AAL being used by a person were coded (N = 111) and we found that 26 (23.4%) were negative and affirmed a deficit viewpoint. Another 45 (40.5%) statements presented a view that AAL was different than WME, but not necessarily because of deficiencies of the speaker. Twenty-four (21.6%) instances were coded as a “resource” view, and the final 16 (14.4%) statements could be considered advocating the right for a speaker to use whichever language he or she wanted. Due to the scope of the data set, it was not possible to observe any major differences between Black and White administrators in regards to these general views.

Ideological stances not monolithic

The difference, deficit, resource, or right views are not monolithic, but rather may be varied in explicit statements about AAL. While a participant may have had an overall deficit view of AAL, he or she may shift viewpoints in different contexts. However, more significantly than individual views of AAL was the relationship between the view of AAL and the constituent group that may speak AAL. There was high variation within participants.

Since administrators must consider different stakeholders, namely teachers, students, and parents, we looked at how these different views changed depending on the constituent group being considered. This revealed some notable results. Table 2 displays the frequency and percentages of each instance coded as a particular constituent group in a matrix relationship with each view. Here, N = 119 because in some statements the participant may have talked about multiple constituencies, thus the coded view would count one instance for each constituent group.

OPEN IN VIEWER

Table 2.

Frequency Counts of Constituency Groups.

Constituent group	Deficit view	Difference view	Resource view	Right view	Total
Parent/Community member	7 (70%)	2 (20%)	1 (10%)	0 (0%)	10 (100%)
Teacher	8 (42%)	5 (26%)	4 (21%)	2 (11%)	19 (100%)
Student	10 (12%)	39 (47%)	19 (23%)	15 (18%)	83 (100%)
Administrator/Self	1 (14%)	2 (29%)	4 (57%)	0 (0%)	7 (100%)

Naming Practices of AAL

Variation of names for AAL and WME

A frequency count of the number of times participants named AAL and WME revealed that AAL was much more contested than WME. That is, there was more variety in the way AAL was named. Table 3 lists the names given to AAL, the frequency counts, and the percentage of use.

OPEN IN VIEWER

Table 3.

Frequency Counts of Naming Practices.

Name	Frequency	Percentage
AAL
Vernacular	32	24.8
Ebonics	24	18.7
Slang	17	13.2
Dialect	10	7.8
African American vernacular English (AAVE)	6	4.7
Black English	6	4.7
Non-standard English	6	4.7
Home language	5	3.9
Their English/Language	4	3.1
Black vernacular	3	2.3
Incorrect grammar	3	2.3
Actual language	2	1.6
Not proper	2	1.6
Accent	1	.8
Broken English	1	.8
Culture language	1	.8
Ethnic language	1	.8
Habit	1	.8
Language of the basketball court	1	.8
Language of the block	1	.8
Second language	1	.8
Wrong language	1	.8
Total:	129	100
SAE
Standard English*	42	53.8
Proper English	11	14
Correct	7	9
Other language/dialect	4	5.1
Formal language	3	3.8
King’s English	3	3.8
Power language	3	3.8
English language	2	2.5
School language	2	2.5
Prescribed language	1	1.3
Total	78	100%

The variation within the names indicates that it was more difficult to find common ground on what to call AAL.

AAL contains a number of names that specifically point to race or skin color. This is consistent with Wolfenstein’s classic statement:

Languages have skin colors. There are White nouns and verbs, White grammar and White syntax. In the absence of challenges to linguistic hegemony, indeed language is White. If you don’t speak White you will not be heard, just as when you don’t look White you will not be seen. (1993, p. 331)

However, the names for WME seem to be silent on the issue of race. With a clear majority of the names including “standard,” one can see that there is a sense of neutrality assumed with the language use. There is no coloring of the standard, and there is no affiliation to a particular group of people with the exception of two Black administrators mockingly referring to the “King’s English” three times over two focus groups.

Significantly, after “vernacular,” the next most used name for AAL specifically indexes color. One Black administrator who used the term Ebonics said “we call it Ebonics” indicating that she, as part of a corporate African American identity, approved of the term, though it is highly contested, and considered pejorative (Baugh, 2000). The third most used term, “slang” was used both as a name for AAL as well as a label applied to generational jargon. Therefore, when coding, we only considered the instances in which “slang” was used to name AAL as a deviation from WME. With 17 instances, this associates AAL as a subset of WME, and therefore makes a value judgment about the language.

Variation of naming AAL and WME

The naming practices, though significant, are complicated even more by the discursive practices of naming. That is, when studying language in interaction, the way things are said matters. In fact, the way things are said may be a more significant indicator of an ideological stance (or a contestation of an ideological stance) than the name itself. While there was no significant difference in the way the Black or White administrators named the language, there was a significant difference in the types of discursive markers used while naming AAL and WME.

The hedging and qualifying practices around AAL and WME revealed that there was more hesitation or backtracking when naming AAL than WME. Figure 1 shows that the frequency of times a participant paused, hesitated, or qualified a name of the language, while Figure 2 displays the percentage.

OPEN IN VIEWER

Figure 1.

Number of hedging or qualifying markers used around naming practices distributed by race.

OPEN IN VIEWER

Figure 2.

Percentage of names surrounded by hedging or qualifying markers.

As these graphs show, there was nearly twice as much hedging accompanying AAL than WME. This difference indicates a discomfort or uneasiness with naming AAL. This corroborates the significant variety and inconsistency of naming AAL. While the data indicates a significant difference between naming AAL and WME, we still are left to wonder why? Possible reasons will be taken up in the discussion of findings, but it is important here to note that there is a significant difference in these hedging markers based on race of the speaker (see Figure 3).

OPEN IN VIEWER

Figure 3.

Number of names surrounded by hedging or qualifying markers.

This difference suggests that White administrators were more uncomfortable with naming AAL. During the interview there was more than one time where a participant explicitly stated “I’m not racist” or cited a Black professor or teacher to minimize the possibility of White supremacy, its legacy of racism, and an assertion of White innocence.