Abstract
Academic English (AE) refers to the language used in school to help students acquire and use knowledge. This article reviews current literature to determine what is known about the nature of AE within the context of K–12 schooling. It describes how AE is conceptualized in the education research literature, how these conceptualizations are realized in instructional practices, and the implications of these conceptualizations for teacher education and professional development. The student population that is the primary focus of the article is English language learners, but the findings have implications for all students who struggle with learning AE. The article raises critical challenges in defining and operationalizing AE for instruction and suggests areas for further inquiry.
Keywords
Language is at the heart of teaching and learning. It is the medium through which concepts and skills are learned and assessed, social relationships and identities are formed, and increasingly deeper and more complex disciplinary understandings are constructed over time (Bloom, Carter, Christian, Otto, & Shuart-Faris, 2004; Gee, 1996; Schleppegrell, 2001, 2004). In addition, as academic content becomes more abstract and complex, language structures become increasingly sophisticated and challenging (Bailey, Burkett, & Freeman, 2010; Snow & Uccelli, 2009; Solomon & Rhodes, 1995). The ways in which teachers, texts, and assessments use language to convey and test disciplinary knowledge determine in large part the content students learn. This is particularly true for students in the process of learning English, or English language learners (ELLs), for whom, as Frances Christie (1985) contends, language is the “hidden curriculum” of schooling; that is, the norms and patterns of language as it is used within and across academic disciplines is never made explicit to a large segment of the student population. Christie argues that success in school is largely a matter of learning the patterns of discourse through which academic concepts and skills are developed, explored, and expanded.
Recent large-scale adoption of the Common Core State Standards (CCSS) for English Language Arts (ELA) and Mathematics is prompting a renewed focus on teaching academic English (AE) across the curriculum. Developed from a partnership of the National Governors Association Center for Best Practices and the Council of Chief State School Officers, the Standards communicate what students should know at each grade level to be college and career ready by high school graduation. Emphasized in the CCSS are the language demands of each content area; students who meet the Standards are described as those who “develop the skills in reading, writing, speaking, and listening that are the foundation for any creative and purposeful expression in language” (Standards, p. 3). As schools across the nation begin to implement the CCSS, the need for defining and understanding how to teach AE is more pertinent than before.
Although facility with AE is an important factor in students’ academic success (Bailey et al., 2010; Francis, Rivera, Lesaux, Kieffer, & Rivera, 2006), the distinguishing characteristics of AE are still unclear. By examining the construct of AE, we gain a comprehensive understanding of the challenges ELLs and other students, including speakers of nonstandard varieties of English and native speakers with little exposure to AE, face in learning this language of school. Our aim in this review of the literature on AE is to describe how AE is defined and conceptualized in the education research literature, how these conceptualizations are realized in instructional practices, and the implications of these conceptualizations for teacher education and professional development. In doing so, we hope to shed light on promising instructional practices that afford ELLs and other students with opportunities to learn the language of school. We first describe the methodology used to guide our review and then discuss findings from the literature.
Methodology
The literature was collected and analyzed as part of a larger, unpublished report commissioned by the U.S. Department of Education. For the present review, the authors have condensed, reorganized, and updated literature from the original document. The collection of documents for the literature review was guided by three areas of inquiry: (a) defining AE, (b) AE teaching practices, and (c) teacher preparation and training in AE. Although considerations of culture are important for ELLs, the focus in this review is on the academic language that teachers need to support. Within each area, the review focuses on the K–12 setting and addresses particular key questions, as shown in Table 1.
Inquiry areas and questions guiding the literature review
Note. AE = academic English; ELL = English language learner.
Documents compiled for the review include research published in professional journals or books, practitioner handbooks, policy documents, and other scholarly works. Searches were restricted to works published during or after year 2000, but earlier pieces that have made substantial contributions to our current understanding of AE are also discussed. Because of the scarcity of research on AE, eligibility for inclusion in the review was necessarily broad. Quantitative and qualitative studies as well as information on the range of expert opinions that exist on the topic of AE are included. Although relevant to the topics of interest, few of the documents reviewed represent quantitative or large-scale studies, and few studies describe methods, populations, or settings in enough detail to assess fully the strength of the findings. Literature pertaining to the teaching of AE for all students was included when curricular or instructional implications for ELLs could be made because no research at this time indicates that methods for teaching AE to ELLs should differ significantly from those used to teach other students who face difficulties in acquiring AE. It is important to note that the target of proficiency in AE is the same for all students.
Conceptualizations of Academic English
In the broadest sense, AE refers to the language used in school to help students acquire and use knowledge (Bailey & Heritage, 2008; Chamot & O’Malley, 1994; Schleppegrell, 2004). However, because researchers approach their inquiry into language and schooling from different theoretical and disciplinary orientations, conceptual frameworks vary in how they define and operationalize AE for research and instruction. In this section, we review the literature that establishes the language of school as an additional register in children’s developing language repertoire. In so doing, we discuss how the literature (a) differentiates between AE and social language, (b) explicates the linguistic features in AE across and within content areas, and (c) addresses social functions of language. The AE frameworks we review are neither mutually exclusive nor contradicting, but they attend to different types and features of AE. Although grounded in differing theoretical frameworks and often disparate target learner populations, each approach or framework illustrates how AE fits within a system of language use. They share a recognition that the language of school exists and that this language is foundational to academic access and success.
How the Literature Differentiates Between Academic and Social Language
The notion of AE as a specialized register for school developed from research conducted in the late 1970s and early 1980s that examined the linguistic challenges for children from diverse backgrounds. Cummins’s early research (1980, 1981) found that assessments of ELLs’ oral language proficiency were not sufficient measures to explain the difficulties many students have in performing academic tasks in school. He argued that the linguistic demands of these tasks contributed to ELLs’ academic difficulties and described some of the characteristics that distinguished AE from the spoken language ELLs acquired more readily. During this same period, Snow’s (1983) research, which examined the interactions between children and their caretakers, noted differences among linguistically diverse children in their use of “decontextualized” language, that is, the language used when meaning is primarily conveyed through linguistic cues that are independent of the immediate communicative context. Snow also described some of the linguistic features of this language, such as the use of more precise and abstract vocabulary. Heath’s (1983) ethnographic research on the congruence between language use in children’s home communities and language use in school demonstrated that children’s home and community experiences with language positioned them more or less favorably for literacy learning in school. Researchers from this era brought attention to the notion of AE as an additional register in children’s developing linguistic repertoire and established that a language of school exists and that this language is foundational to academic access and success.
In his seminal work, Cummins (1980, 1981) characterized most of the language students bring to school as Basic Interpersonal Communication Skills (BICS) or social language. From that foundation, students develop broader language skills, which he referred to as Cognitive Academic Language Proficiency (CALP). This work by Cummins provided one of the first paradigms for thinking about AE, serving as a catalyst for a deeper investigation and conceptualization of the language of school. In the literature, BICS was considered social language and CALP the precursor to AE, with BICS relying on contextual cues for transmitting meaning although CALP was thought to be more cognitively complex and thus less reliant on these cues. Zwiers (2008) captures the general understanding of the distinctions between social language and AE: “A conversation with a friend about a recent sports event would involve much social language, whereas listening to a lecture on globalization would be more academic” (p. 20).
Cummins’s early work has been interpreted by some as a dichotomous view of language acquisition and use (Scarcella, 2003), although Cummins (1981, 1984, 2000) stresses the simultaneous acquisition and development of both social and cognitive aspects in school-age children with the distinction being in the degree of cognitive demands and contextual support of language-use situations. Cummins (2000) makes the point that language and context are inextricable, as “illustrated in the different registers required for success in university English literature courses as compared to success as a stand-up comedian” (p. 55).
Cummins’s (2000) quadrants of cognitive involvement and context provide a “visual metaphor” (p. 85) for the registers of language students have access to within different contexts. In this framework, science experiments and standardized science exams are both considered cognitively demanding. The former, however, is also considered to be embedded in context that supports student comprehension. The cognitive demand expressed by the framework refers to the information processing necessary to understand and use specific kinds of language. Cognitive processes are further explicated in terms of academic functions such as application, analysis, synthesis, and evaluation.
In reexamining this focus on differences between social language and AE, critics have pointed out that these characterizations are not sufficient to explain the complexities of language in school settings. Furthermore, that they represent a deficit view of the language that students bring to acquiring AE in the classroom (MacSwan & Rolstad, 2003). Bailey (2007), for example, cautions against “believing that there is something inherent in social language that makes it less sophisticated or less cognitively demanding than language used in an academic context” (p. 9). Rather, she suggests that the differences should be considered as “differences in the relative frequency of complex grammatical structures, specialized vocabulary, and uncommon language functions” (p. 9).
Bailey and Heritage (2008) broaden the conceptualization of school language use by breaking AE down further into School Navigational Language (SNL) and Curriculum Content Language (CCL), where SNL is the language students use “to communicate with teachers and peers in the school setting in a very broad sense”—the language of classroom management—and CCL is “the language used in the process of teaching and learning content material” (p. 15). The authors contrast SL, SNL, and CCL based on the following features: the purposes to which these language varieties are put, their degree of formality, the context of their uses, the context of their acquisition, the predominant modalities they use (e.g., listening, speaking, reading, writing), teacher expectations for language abilities across the three varieties, and grade-level expectations. This conceptualization captures the range and variety of language acquisition situations and use for all students, native English speakers and ELLs, and provides a potential framework to help educators address the role of language in school settings.
Scarcella (2008) draws on Bailey and Heritage’s (2008) description of SNL in her discussion of AE and also introduces the notions of Foundational Knowledge of English and Essential Academic Language (EAL). She posits that these three types of English knowledge are necessary for students to learn CCL. The foundational knowledge of English refers to the basics of the English language, in essence, the basic skills and common vocabulary important for communication both outside and within the school setting (e.g., knowing how to read and write, how to produce key types of sentences, how to use verb tenses, etc.). However, Scarcella (2008) clarifies that this knowledge is “not the equivalent of playground language or social, informal language” (p. 5) and requires explicit, systematic instruction. In contrast, EAL consists of “the basic features of academic language that are used across all content areas” (p. 6) including academic words, complex sentence structures, and discourse features that provide cohesion. Scarcella (2008) draws several instructional implications from her conceptualization of AE, arguing that prerequisite to the teaching and learning of subject-specific language, ELLs should have Foundational Knowledge of English, should know SNL, and would benefit from already controlling EAL.
How the Literature Explicates Features Within and Across Content Areas
Conceptualizations of AE often have as their goal an explication of the more challenging features of language as it is used within the academic disciplines of school. In this subsection, we discuss conceptualizations that attempt to describe AE across the curriculum and within content areas, attending first to comprehensive frameworks that address multiple aspects of language and then to more focused frameworks that examine the particular linguistic features of vocabulary, grammar, and discourse. Within our discussion of work on AE discourse, we delineate research by content areas because many studies in this area occur within specific academic disciplines.
In their work on articulating the AE demands of content area classrooms, Bailey and colleagues (Bailey, Butler, Borrego, LaFramenta, & Ong, 2002; Bailey, Butler, & Sato, 2007; Bailey, Butler, Stevens, & Lord, 2007; Bailey, Huang, Shin, Farnsworth, & Butler, 2006; Butler, Bailey, Stevens, Huang, & Lord, 2004) distinguish AE from social or “everyday” English on three key levels: “the lexical or academic vocabulary level, the grammatical or syntactic level, and the discourse or organizational level” (Bailey, 2007, p. 3). Using this initial characterization of AE, Bailey and colleagues further operationalized the AE construct through an empirical approach that examined multiple sources of AE use in mainstream, upper-elementary classes (Bailey & Butler, 2002, 2007; Bailey, Butler, & Sato, 2007; Butler et al., 2004). The goal was to begin to articulate the AE target that students (native speakers and ELLs) must achieve to function well in content classes, which could serve as a baseline for the development of English language proficiency (ELP) standards and curricula materials. The effort included classroom observations, along with analyses of ESL and content standards, teacher expectations, and textbooks. Together the results of these empirical studies suggest the beginnings of a potential paradigm for discussing AE.
Scarcella (2003) also conceptualizes AE on different linguistic levels, and like Bailey and Heritage (2008), Scarcella (2003) focuses on the language and cognitive skills that should ideally be developed prior to the study of subject-specific language. As part of a conceptual framework for analyzing the AE that K–12 students must develop for success at the postsecondary level, Scarcella borrows Kern’s (2000) theoretical model of academic literacy and explicates AE from linguistic, cognitive, and sociocultural/psychological dimensions. Based on an analysis of linguistic competence rather than empirical research on AE in the classroom, the linguistic dimension of the framework consists of five components: phonological, lexical, grammatical, sociolinguistic, and discourse. Within each component, Scarcella (2003) provides examples of the linguistic features that enable students to use language for complex academic tasks. Her examples highlight the language gap for students who have not yet developed AE and emphasize the need for helping students master a more robust range of language proficiency. Scarcella’s discussion of the cognitive and sociocultural/psychological dimensions of AE expands AE conceptualizations beyond linguistic features, suggesting that proficiency requires more than linguistic competence.
Snow and Uccelli (2009) provide an inventory of social and academic uses of language that is based on the notion that “communication goals are seen as driving decisions about specifics of expression” (p. 122). They suggest organizing linguistic features of language into the following categories: interpersonal stance, information load, organization of information, lexical choices, and representational congruence (i.e., how grammar is used to depict reality) with specific vocabulary and grammar structures necessary to actualize the features. They offer a pragmatic heuristic based on context and social interaction as the core for characterizing AE that captures the specifics of lexicon, grammar, and discourse features. Like Scarcella (2008), Snow and Uccelli include cognitive domains in their discussion of AE performance.
In addition to developing overarching theories and frameworks of AE, researchers and practitioners have attempted to describe the challenges individual linguistic features of AE pose for ELLs. Specifically, research in the areas of vocabulary, grammar, and discourse functions contributes to work addressing the linguistic components of AE.
Vocabulary
Over the past 10 years, much of the AE literature has focused on the importance of vocabulary as a component of AE. At its broadest level, academic vocabulary consists of words students must comprehend in order to access the concepts associated with a particular discipline and also use in order to display their acquisition of these concepts.
Using principles from corpus linguistics (samples of authentic texts), Coxhead (2000) developed and evaluated an academic word list (AWL), providing evidence that academic texts can be characterized to some degree by specialized vocabulary. To define what counted as a word, Coxhead drew on research that suggests comprehension of inflected or derived items within a word family requires little additional processing effort if the base word is known and if learners control basic word-building processes (Bauer & Nation, 1993). From a corpus of 3.5 million words drawn from written academic texts, Coxhead compiled a list of 570 word families common to a wide range of academic texts encountered by university students. Word families were selected based on (a) specialized occurrence: not being included in the first 2,000 most frequent-occurring words in English; (b) range: occurring at least 10 times in each of the four main sections of the corpus (arts, commerce, law, science) and in 15 or more of 28 additional subject areas; and (c) frequency: occurring at least 100 times in the corpus. Although not addressing spoken language, the AWL provides a specialized academic vocabulary that broadly covers written academic texts across a number of subject areas. Originally developed for ESL university students, the AWL contains many of the words that appear in academic texts well before the postsecondary level.
Another way to conceptualize academic vocabulary is in terms of tiers or categories. A commonly accepted classification system was developed by Beck, McKeown, and Kucan (2002), and later adapted (e.g., Calderón et al., 2005). It frames academic vocabulary according to three tiers. The first tier is nonacademic, conversational vocabulary such as flower or sleep; the second consists of general academic words such as however or illustrate; and the third tier is composed of content-specific, technical vocabulary such as organism or rectangle. Stevens, Butler, and Castellon-Wellington (2000) offer a similar example of a vocabulary classification scheme that uses different labels: high-frequency words (Tier 1), nonspecialized academic words (Tier 2), and specialized academic words (Tier 3). Researchers have adopted other labels, such as brick and mortar words (Dutro & Moran, 2003) and content-obligatory and content-compatible words (Snow, Met, & Genesee, 1989) to describe the complexities of academic vocabulary development.
Grammar
Traditionally, educators and researchers have viewed grammar as a sentence-level phenomenon (Celce-Murcia, 2002). From this perspective, grammatical elements are understood as syntactic categories (e.g., noun, verb, adjective, etc.) and identified according to the roles they play in a sentence (e.g., subject, verb, object, etc.; Schleppegrell, 2004). This notion of grammar has been promoted by language educators who view grammar as a set of rules for accuracy in language use, by reference grammars and style books, and via language curricula and textbooks (Celce-Murcia, 2002; Schleppegrell, 2007). Others (Halliday, 1994; Schleppegrell, 2007; Unsworth, 1999) have also argued that such “sentence-based views of grammar” (Celce-Murcia, 2002, p. 144) are inconsistent with findings from linguistic research and have “negative consequences for the way in which grammar is described and taught” (Celce-Murcia, 2002, pp. 123–124).
A functional linguistics perspective, in contrast, views grammar as “a dynamic system of linguistic choices that students learn to use to accomplish a wide variety of social, academic, and political goals in and out of school” (Gebhard, Harman, & Seger, 2007, p. 421). A key element in this system is “the configuration of grammatical structures which is typical or expected in different kinds of socially relevant tasks and links those linguistic choices with the social purposes and situations that the ‘texts’ (spoken or written) participate in” (Schleppegrell, 2004, p. 45). Based on Halliday’s systemic functional linguistics (SFL; Halliday, 1978; Halliday & Hassan, 1985), a functional approach to grammar proposes that explanations of how language works must be grounded in functional analysis because language evolved as humans performed certain social functions, or purposes, within varied social contexts.
Working within a functional tradition, linguists have identified and described major disciplinary genres (types of texts and tasks) and the language patterns most prevalently used within these genres (Schleppegrell, 2009). One of the hallmarks of academic genres is the length and complexity of sentences. Within a given text, information is presented within a series of clauses that are often connected to one another with subordinating conjunctions, such as when, after, however, and so on. These conjunctions serve as cohesive devices that help the reader, or listener, understand the relationships between the ideas or messages contained within the various clauses. Using these, and other cohesive devices, across a series of clauses in order to derive meaning from a text is largely an unconscious activity when a student has acquired the AE of a particular register (Halliday, 1994); however, for ELLs and others who are in the process of learning AE while learning academic subject matter, unpacking meaning from these texts is not automatized and is extremely challenging (National Research Council, 2010).
Another feature contributing to the complexity of academic texts is the frequent use of long noun phrases with embedded clauses (Schleppegrell, 2009). A noun phrase is a group of words that acts as a single unit of meaning, is built around a noun, and expands, or modifies, the noun’s meaning. These “expanders” function as pre- and postmodifiers of the noun. In this sentence, “If the habitats that support them are destroyed, then the species will disappear,” the embedded clause “that support them” provides additional yet necessary information about the noun “habitat.” It provides a level of specificity that the noun alone could not convey. Understanding long noun phrases is critical to unlocking meaning in text because they often serve the purpose of identifying and further specifying complex concepts.
A third feature of academic texts is the prevalent use of nominalization. Nominalization refers to the use of a verb or adjective as a noun or noun phrase (e.g., fragmentation). A number of researchers (e.g., Moschkovich, 2002; Schleppegrell, 2004, 2009; Veel, 1999) have demonstrated that AE is characterized by a greater use of nominal forms and long noun phrases that function to condense information and convey abstract concepts and relationships in precise ways. In general, these grammatical features of AE allow the writer or speaker to accumulate knowledge from one clause to the next while maintaining coherence and cohesion across a text (Halliday, 1994). Cognitive researchers studying the relationship between syntax and comprehension have found that working memory, the ability to temporarily retain and process new information, plays a part in student comprehension of such texts. Unpacking the multiple connections among language features (e.g., vocabulary, grammar, text cohesion) poses a cognitive challenge to the ELL listener or reader (Graesser, McNamara, & Kulikowich, 2011).
Discourse
Discourse works in tandem with grammar to create meaning. Wong Fillmore and Snow (2000) highlight the need for teachers to be able to analyze the discourse structures and rhetorical devices that are prevalent in texts from their disciplines, and then to provide explicit instruction in these structures and devices within the context of subject area instruction. As with grammar, discourse is described in different ways in the literature, referring broadly to any extended piece of language beyond the sentence level and to typical verbal and written interactions within academic disciplines. The literature reviewed primarily addresses one aspect of discourse, academic language functions (AE functions). According to Bailey, Butler, Stevens, et al. (2007), an AE function refers to the language associated with academic tasks and purposes, such as asking and answering informational or clarifying questions, explaining cause and effect, and persuading.
The discourse features of AE are often conceptualized within individual school disciplines. Studies in this area examine the various purposes for which language is used within a particular academic setting. This literature has led to the beginnings of taxonomies of language functions within and across school disciplines, revealing subtle and distinct differences as well as commonalities across content areas.
Science
Much of the work on AE in science is oriented toward identifying the linguistic features that will have implications for ways teachers can assist students in acquiring the language skills necessary for success in science classes. Bailey, Butler, Stevens, et al. (2007) provide an analysis of the teacher talk observed during fourth- and fifth-grade science lessons that yielded a categorization of the type of language students were hearing as teachers were providing science instruction. The analysis helped specify subcategories under two broad categories: context of instruction and teacher communicative intent. Bailey and colleagues provide a matrix that illustrates the intersection of context and communicative intent and captures the relative frequency of instances of specific communication intent across classroom visits. From this analysis, four oral language functions were identified: explanation, description, comparison, and assessment. Explanation and description were the two most frequently occurring functions in the teachers’ speech, both when imparting content information and giving directions, and when providing support for tasks students were to carry out. Comparison was used occasionally when introducing new terms and giving instructions. These data help demonstrate the range of use of AE functions in very specific types of classroom science activities.
AE functions common to science are also described in a guidebook on making science accessible to ELLs in upper elementary through high school (Carr, Sexton, & Lagunoff, 2006). The guidebook refers to the lexicon and discourse of science. Whereas the lexicon is defined as “the set of terms scientists and science learners use to communicate about their subject matter” (p. 41), academic discourse “involves use of language to convey extended expression of thought on a topic in connected speech or writing” (p. 42). Discourse patterns that are associated with the AE functions of science include providing an explanation supported by evidence, writing a report of an experiment, along with other AE functions.
Gee (2008) used excerpts from a science journal, a popular science magazine, and a high school science textbook to illustrate different discourse features of AE that vary according to purpose and audience. The first type of text is “concerned with furthering conceptual understanding within a sub-discipline of biology. Its language is carefully developed to do this—to build evidence and marshal support for certain biological claims within particular parts of the biological community” (p. 59). In contrast, the excerpt from a popular science magazine is not about methodology, theory, claims, or arguments but rather “to inform the public and to build public support for [scientists’] work in the field at large” (pp. 59–60). The third excerpt, from a high school science textbook, is an example of an explanatory definition, a genre common in science textbooks, which is characterized by language that classifies items in relation to each other. In all three instances, the language used is with a specific communicative purpose in mind.
Mathematics
The literature on the AE demands of mathematics dispels the perception that the content of math is less language dependent than that of other content areas, focusing on the role language plays as students engage in problem solving, participate in discourse practices around mathematical topics, and construct meaning during their lessons (e.g., Adams, 2003; Moschkovich, 2002; Schleppegrell, 2007).
Linguistic challenges in mathematics and common patterns of mathematics language are highlighted in a synthesis of research by applied linguists and mathematics educators (Schleppegrell, 2007). The author points out that “each subject area has its own ways of using language to construct knowledge, and students need to be able to use language effectively to participate in those ways of knowing” (p. 140). She draws on Halliday’s (1978) notion of a mathematics register and explains that the understanding of a specific mathematics register helps us understand how language constructs knowledge differently for different content areas.
Adams (2003) discusses the unique AE demands of mathematics for elementary and secondary-level classrooms, using examples from the literature on mathematics education and language learning. She describes the language of mathematics as a language of order, composed of abstractions, numerals, and symbols that can be used “to communicate, to solve problems, to engage in recreation, and to create works of art and mechanical tools” (p. 786). The author recommends instruction that draws students’ attention to the unique features of AE found in mathematics.
Syntactic and semantic features that pose difficulties for students are described by Schleppegrell (2007). The author observes that “mathematics is highly technical with characteristic patterns of vocabulary and grammar” (p. 142) and draws a distinction between the technical vocabulary of mathematics (Tier 3 following Beck et al., 2002) and the more common (Tier 1) words that have a different meaning when used in mathematics. Schleppegrell argues that, in addition to vocabulary, attention should be paid to grammatical constructions that help establish mathematical discourse. She discusses the complexities of “long, dense noun phrases such as the volume of a rectangular prism with sides 8, 10, and 12 cm” and the challenges of “the precise and technical meanings of conjunctions that may be used in different ways in ordinary everyday language” (p. 144). She argues persuasively that although vocabulary is critical, it is not sufficient to understanding the mathematics register and that educators must expand their understanding of the language of mathematics to include the grammatical constructions as well.
Common and distinct language demands of mathematics and science are also documented in an analysis of the academic language demands in state standards for algebra and biology (DiCerbo, Anstrom, & Rivera, 2013). Among the language demands common to both disciplines are definition and organization, summarization and analysis. Language demands associated with science include the ability to make inferences or hypotheses; for mathematics, analogy and exemplification.
Rather than provide an inventory of language features to characterize the AE of mathematics, Moschkovich (2002) offers an approach to looking at classroom discussions that examines the meanings of mathematics words, resources students draw on to communicate mathematics concepts, and common discourse patterns and practices. According to this view, language is but one of the resources students use to participate in mathematical practices. They also draw on social and material resources, such as gestures, objects, or use of the native language, to access and construct meaning as they engage in learning. In addition to the notion that language is only one component of mathematical discourses, this perspective assumes that meanings are “multiple, changing, situated, and sociocultural” (p. 207).
History/social studies
A range of literacy-related linguistic features that typify AE in history/social studies classrooms have emerged from analyses of reading and writing tasks in this content area. For instance, Short (1994) based her analysis of the AE linguistic features found in history/social studies textbooks on a taxonomy of six types commonly mentioned in reviews of expository texts. In her analysis, the most frequent text structures were sequential or chronological and cause/effect. Short then explored AE used in classroom routines, categorizing the language to include semantic and syntactic features as well as functions and tasks. Examples of AE demands in history include language functions used by both students and teachers such as explain and justify, language skills tasks such as taking notes and finding the main idea, text structures such as cause and effect and generalization and example, and syntactic features such as historical present and causative signals (Short, 1994).
Additional insights into the AE demands of history are provided by a qualitative discourse analysis of 345 student writing samples by ELLs and native English speakers in Grades 8 and 11 (Schleppegrell, 2005). The author identifies writing tasks that reflect “different genres, recognizable types of texts that have particular purposes” (Schleppegrell, 2005, p. 5). Specifically, expository writing tasks identified included writing historical accounts, historical explanations, and historical arguments. In an effort to specify the AE in the content area of history, Schleppegrell describes how historical accounts draw on language features such as temporal ordering through the use of adverbs and prepositional phrases; historical explanations make use of description and rhetorical features to establish events in relation to each other or to compare and contrast different points of view; and historical arguments use modality and other language resources to construct claims or authorial judgments. According to the author, writing an effective argument requires first understanding how to write an effective account and explanation.
The AE demands of history are also examined in an action research study conducted by Zwiers (2006) with two of his history-based classrooms of 60 middle school ELLs. The author defines AE broadly, to include “highly-visible and discipline-specific terms such as photosynthesis” (p. 318), “more universal words and phrases that occur across disciplines” (p. 318), and “various types of thinking used in different content areas” (p. 319). He reviewed resources on historiography and California standards for history teaching in the middle grades to identify six dimensions of historical thinking: background knowledge, cause (why?), effect (what resulted?), level of bias in reporting of historical events, empathy, and perspective, interpretation, and application. In analyzing audio-recorded lessons and student written work for “evidence of academic thinking skills and AE acquisition” (p. 327), Zwiers (2006) found that students are more likely to think critically and use AE when the teacher’s own thinking is made visible and AE is “modeled, scaffolded, and practiced in the ways that historians think about history” (p. 330). From his research, the author emphasizes the need to teach AE in tandem with critical thinking and discipline-specific content.
Common Core State Standards for English language arts and mathematics
With states adopting and implementing more rigorous college and career standards, recent research has begun to examine the AE demands of these standards and the implications of these demands for ELLs. Sato, Lagunoff, and Yeagley (2011) applied an academic language demands taxonomy, Language for Achievement (Sato, Lagunoff, & Worth, 2008), to the High School CCSS for ELA and mathematics to identify the productive academic language functions (language necessary for speaking and writing) reflected in the standards. Findings from this study provide information on the most common productive academic language functions reflected in the ELA and mathematics standards. Interpretation, analysis, evaluation, explanation, and description were the five most common language functions reflected in the ELA standards. Symbolization and representation, explanation, argumentation, description, and comparison/contrast were the five most common language functions reflected in the CCSS for mathematics. Subsequent analyses address aspects of AE reflected in the standards, including receptive AE functions, vocabulary, language features/grammar, and text structure/organization. The authors (Sato et al., 2011) underscore the practical value of this research in helping educators understand the AE demands that facilitate ELL access to engagement with grade-level academic content.
Building on work from Sato et al. (2011), the Framework for English Language Proficiency Development Standards (Framework) articulated by the Council of Chief State School Officers (CCSSO) clarifies the language demands implicit in the CCSS and Next Generation Science Standards (NGSS) for ELA, mathematics, and science and outlines “the language practices that all ELLs must acquire in order to successfully master the CCSS and NGSS and for second language acquisition more generally” (p. 1). Specifically, the Framework identifies “key practices” for standards and then describes analytical tasks and productive and receptive language functions that are necessary to engage in these practices. Key practices in ELA include analyzing complex texts, constructing arguments from evidence, and synthesizing ideas from text. In mathematics and science discourse, the Framework draws attention to mathematical explanations, problems, and arguments and the deep connection between scientific concepts and language. The Framework was designed not only to assist state agencies in developing ELP standards but also can serve as a resource to help define the language needed to engage in the practices inherent in the CCSS.
The literature describing the linguistic features of AE demonstrates the complexity researchers face when conceptualizing AE. A consensus on the defining features of AE is beginning to emerge, as evidenced by the CCSSO’s recent Framework (2012). A consistent message is that any definition of AE must consider a variety of linguistic dimensions including, but not necessarily limited to, vocabulary, grammar, discourse, and the disciplinary contexts in which AE occurs. Given that much of the empirical research supporting the AE frameworks discussed thus far is based on analyses of classroom discourse, it can be assumed that the research shares the assumption that the language of school occurs in socially situated contexts. The next section describes the way in which literature on AE addresses social functions and contexts of language.
How the Literature Addresses the Social Function of Language
Social explanations of language began to emerge more strongly in the 1990s, offering a new perspective for AE research (Ortega, 2012). A main tenet of socially situated conceptualizations of language is that language is the primary resource for making meaning and that language development depends on opportunities for social experience and interaction. Much of our understanding within this perspective comes from the discourse analysis work of functional linguists. Their examinations of the academic texts used in school and the interactions among teachers and students around academic content have identified the major disciplinary genres and described the language patterns that construct these genres (Schleppegrell, 2009). The theory that undergirds these socially situated conceptualizations of language is M. A. K. Halliday’s systemic functional linguistics (SFL; Halliday & Hassan, 1985; Martin, 1992) briefly described previously in our discussion of the literature on AE grammar. SFL theory posits that language construes social contexts: “Functional grammar connects the linguistic and the social by offering descriptions of language form in relation to semantics, and descriptions of semantic categories in relation to the contexts of social living” (Colombi & Schleppegrell, 2002, p. 9). From this perspective, a linguistic register is the “constellation” of lexical and grammatical features that construe a particular situational context. Thus, the lexical and grammatical forms of language vary according to the context, purpose, and demands of different social situations.
In general, researchers agree that AE registers vary from social registers in the ways they develop logical reasoning, organize information, and integrate interpretation (Christie, 2002; Gee, 2005; Moschkovich, 2002; Schleppegrell, 2004; Snow & Uccelli, 2009). For example, the prevalence of linguistic features, such as nominalization and long noun phrases, help distinguish academic forms of organizing information from more social forms (Christie, 2002; Moschkovich, 2002; Schleppegrell, 2004; Snow & Uccelli, 2009).
In exploring the use of AE in the classroom, Gee (2005) defines AE as a large family of “related social languages” (p. 21). Rather than characterize social language in terms of an entire discipline (e.g., the language of biology or chemistry), Gee relates it to the specific practices and activities carried out by users within particular subdisciplines (e.g., the language activities of a microbiologist doing microbiology). It is important to note that Gee’s use of social language differs markedly from that of Cummins (1980, 1981) or Bailey and Heritage (2008).
For Gee (2005), social language refers to how language is used to establish “a socially situated identity and carry out a particular socially situated activity” (p. 20). Thus, it is at the level of subdisciplines (e.g., geometry within mathematics) that Gee identifies patterns of language use. In Gee’s words, “Different patterns or co-relations of grammatical elements . . . are associated with or map to particular social languages . . . associated with specific socially situated identities and activities” (p. 20). Gee indicates that within a subdiscipline there will be variations in language use by modality (oral vs. written language) and also distinctive uses of discourse markers that signal the intentions of authors or speakers in connected texts.
In her work with 9- and 10-year-old ELLs, Gibbons (1998) stresses the intertextual nature of classroom language or the interrelationship of the language modalities—listening, speaking, reading, writing—in learning content material. To develop curricula for students who have difficulty linking language across modalities, Gibbons argues that a model of language is needed that is discourse and text-based and not focused on grammatical structures per se, although form is an integral component of any language model. Drawing on Halliday’s (1994) SFL and Halliday and Hassan’s (1985) register theory, Gibbons establishes the importance of context in the teaching and learning of AE. She examines four texts that vary by modality (written and spoken), time (past and present), and level of formality (encyclopedic text and student writing) to illustrate how linguistic features change from situation to situation as context dependency changes. The texts show a progression of specific language features that make the texts more or less context dependent as the communicative situation shifts, highlighting the critical nature of the differences in classroom contexts.
Bunch (2006) also differentiates between types of AE used in classroom contexts in his qualitative study of student-to-student interactions in two seventh-grade social studies classrooms. The author’s findings are based on his analysis of transcripts collected as part of a larger study of “how linguistically heterogeneous groups in two classrooms used language to engage in group tasks” (p. 290). In working on these different classroom tasks, the students used what the author refers to as the language of ideas and the language of display. Bunch equates the language of ideas with the context-embedded language associated with everyday or informal conversations. The language of display, in contrast, is more formal or decontextualized. The author observed that group interactions in preparation for presentations were conducted in the language of ideas whereas presentations to the teacher and class were conducted in the language of display. The author concludes that it is necessary to reconsider the notion of AE in light of the range of AE demands within the typical middle school classroom. In terms of instructional practice, he emphasizes the benefits of attending to the development of a community of academic discourse, with students engaged in extended interactions.
Drawing on SFL theory (Halliday, 1994), Christie (2002) builds an argument for the criticality of abstraction, generalization, and argument in the development of advanced literacy in writing. Christie’s goal was “to develop some measures by which the nature of students’ control of written language must change if they are to achieve the kinds of advanced literacy that participation in a complex early 21st century will require” (p. 64). She examined seven texts, six of which came from primary- and secondary-level Australian students who were native and nonnative speakers of English; one text came from a newspaper in Melbourne. All texts (with the exception of the newspaper text which was not judged) were rated as exemplars of grade-level writing. Although the sample size is small, the detailed descriptive analyses yield initial measures of “developing success in writing” (p. 64), which include control of reference and theme and growing facility with abstractions.
AE is a dynamic concept that will continually evolve as educators and researchers better understand the use of language in school settings, and the literature attempting to conceptualize AE demonstrates the complexity researchers face when describing AE. Although varying in specific details, the common themes of context, multiple varieties of language, and integration of language features and modalities are central to most of the frameworks discussed. Although none of the conceptualizations of AE are in opposition to each other, they represent different perspectives and emphases that are important to consider when examining how to teach or develop AE and the level of AE necessary for academic success. We next discuss research on AE instruction with an understanding that explorations of how to teach AE are intertwined with frameworks for understanding AE.
Research on Academic English Instruction
Empirical research on AE instruction contributes to our understanding of the best ways to teach or develop AE effectively. As with work attempting to conceptualize and define AE, research on AE instruction represents multiple perspectives and areas of focus, all of which have the potential to provide insight into aspects of teaching and learning that support AE development (Anstrom et al., 2010). The act of engaging in “multiple views and voices produces not only more socially useful and ethically responsible knowledge but also better and more valid knowledge” (Ortega, 2012, p. 210). In this section, we first discuss research that focuses on the effectiveness of various approaches to instructing ELLs in one or more language features (e.g., vocabulary, past tense forms). We then turn to instructional research that examines the social discourse practices (e.g., student interaction patterns) through which AE can be developed across and within content areas.
Instruction in Academic Vocabulary
Recent research on academic vocabulary emphasizes the importance of instruction in general and content-specific vocabulary, including the multiple meanings of words within and across content areas, and explicit teaching of word morphology and word origin. According to multiple authors (e.g., Bailey, Butler, Stevens, et al., 2007; Beck et al., 2002; Calderón et al., 2005), general academic vocabulary (Tier 2 words) is frequently neglected in both language and content instruction. These words are used across academic disciplines (e.g., conclusion, evidence, represent) and often have “everyday” counterparts that are suggestive of but not as precise as the academic terms, for example, ending versus conclusion (Beck et al., 2002; Calderón, 2007). The literature suggests that this general academic vocabulary is critical for students to understand the concepts associated with academic content and demonstrate their understanding to others.
The importance of word study, or explicit instruction in the structure and origin of words, is also emphasized in literature on vocabulary instruction. Aspects of academic vocabulary that are important for teachers to understand include polysemy (multiple meanings) and word derivations, which involves knowledge of roots and affixes, and how these units work to change the meaning and usage of words (Calderón et al., 2005). According to Wong Fillmore and Snow (2000), students can learn words faster and more efficiently if they are instructed in the relationships between word forms, and thus teachers should know how the lexicon is structured and acquired.
Intervention research that illustrates the effectiveness of word study in developing AE is reported by Calderón (2007). The report describes results of a 6-month academic vocabulary treatment based on the Success for All model, a whole-school reform model that includes a reading, writing, and oral language development program. Study participants included 300 bilingual (English-Spanish) third-graders across eight elementary schools in El Paso, Texas. In each school, students received 90 minutes of reading instruction per day, including 30 minutes focused specifically on academic vocabulary: pronunciation, meanings, and English-Spanish cognates. The vocabulary used in the lessons was selected from school texts and classified according to a process adapted from the three-tier model developed by Beck et al. (2002). Calderón and colleagues further classified target words with respect to the following criteria: (a) concrete or abstract, (b) presence of a Spanish cognate, (c) polysemous within a certain content area, and (d) utility across content areas. The intervention had a positive effect on academic vocabulary: Posttest scores on the English versions of the Woodcock Language Proficiency Battery–Revised were significantly higher for the experimental group than those of the control group in three of four sections. Findings also indicated that even though the treatment included no instruction on Spanish language, only references to Spanish cognates, the ELL’s scores slightly improved on the Spanish test as well.
In one quasi-experimental study of Word Generation, a whole-school academic vocabulary improvement intervention, middle school (Grades 6–8) language minority students in treatment schools showed greater gains than English-only students (Snow, Lawrence, & White, 2010). For 24 weeks, students in the intervention group were introduced to five target words and then participated in a sequence of activities that embedded use of the words across the content areas. Comparison of student scores on a pre- and post-multiple-choice vocabulary test revealed that students in the intervention group learned significantly more words than students in the control group, but the effect sizes were greater for language minority students than for English-only students. In addition, improvements on the vocabulary assessments predicted performance on the state assessment in ELA. The authors suggest that improvements on the state ELA test reflect not only Word Generation’s emphasis on specific word knowledge but also the “new content, deep reading and comprehension skills, discussion, argumentation, and writing” (Snow et al., 2010, p. 341) that the curriculum taught.
Principles of quality instruction in academic vocabulary were also emphasized in a study of 23 self-contained classrooms of fifth, sixth, and seventh graders in Canadian public schools (Scott, Jamieson-Noel, & Asselin, 2003). According to the authors, precise use of vocabulary is a defining feature of AE, along with knowledge of how words are used and grammatical accuracy. Observations were conducted over 3 days in each of the classrooms; classrooms included a cross-section of learners in terms of geographic area, socioeconomic status, and ELP. As part of the study, the authors examined (a) the time spent on vocabulary instruction; (b) the format of instruction, including individual work, small group work, or whole class instruction; and (c) instructional methods. Vocabulary instruction was defined as “every instance in which word-level knowledge was a focus of instruction” (Scott et al., 2003, p. 273). Instructional quality was assessed according to the principles of supporting students in personalizing word learning, immersing students in words, providing repeated exposure to words, and encouraging students to make connections between known words and words they are learning. Few instances were observed of instruction demonstrating these principles of quality instruction, with the exception of providing multiple exposures to words. The authors conclude that “even in classrooms with rich contexts, teachers do considerable mentioning and assigning and little actual teaching” (Scott et al., 2003, p. 282). As a result, the amount and type of vocabulary instruction occurring in schools is not necessarily adequate for ELLs or other students with limited proficiency in AE. Scott and colleagues recommend a focus on both curriculum and professional development that illustrates research-based vocabulary instruction.
AE instruction in academic vocabulary was part of a larger research project focused on how secondary teachers define and teach AE (Bruna, Vann, & Escudero, 2007). The authors describe findings from a sheltered science class of ninth-grade ELLs. One of the findings reported by the authors is that, although AE was a primary instructional focus, it was defined only as academic vocabulary. Classroom observations indicated that the grammatical and discourse features of language were neither modeled nor taught, and students were not engaged in the language of science beyond the lexical level. The authors conclude from their analysis that a narrow focus on academic vocabulary ultimately limits ELLs from “the very linguistic input and output they need in order to acquire the language of science” (Bruna et al., 2007, p. 51). Moreover, by equating academic language with academic vocabulary, students were not given sufficient opportunities to produce extended discourse around academic content, potentially limiting their ability to fully understand and interact in the discipline of science.
Instruction in Grammar
Beginning in the late 1970s with the development and widespread use of more communicative language teaching methods, explicit grammar instruction was often deemphasized in ELL instruction. Current second language acquisition research has refocused attention on the importance of grammar instruction, and particularly grammar feedback, as necessary for ELLs to attain high levels of English proficiency (Ellis, 2002; Long, 2007; Nassaji & Fotos, 2004). According to their review of research on grammar instruction, Nassaji and Fotos (2004) found that instructional approaches that emphasize student noticing, continued awareness of target structures in meaningful contexts, and opportunities to produce and practice these forms are essential for student learning of more linguistically complex structures.
In keeping with Nassaji and Fotos’s (2004) findings on the importance of explicit instruction in complex grammatical forms, Snow and Uccelli (2009) illustrated the potential challenges for students when explicit instruction in grammar and AE functions is not provided. The study involved a classroom intervention to help a diverse group of seventh-grade urban students acquire general academic vocabulary. To determine the effectiveness of the intervention, students were asked to write short argumentative essays at the end of each week that included the academic vocabulary studied. No explicit instruction in constructing an argumentative essay was given. Analysis of the student essays involved comparing the linguistic features of the students’ writing with those of an adult version on the topic. This analysis uncovered a number of AE features not present in the student writing. Features missing from the student writing included, among others, lexical density (i.e., packing a lot of information into a few words); the use of low-frequency words such as enhance and transmit; modal verbs such as would or might; connectives such as whereas or if; words that signal a detached stance (e.g., some would argue . . ., others contend . . . ); and elaborate noun phrases (e.g., a primary goal of education; Snow & Uccelli, 2009).
Despite this renewed attention to grammar, Bayley’s (2009) review of the research on the relationship between explicit grammar instruction and oral language development of ELLs noted that although there were a number of studies that examined different types of grammar instruction for second language learners, most focused on college-level international students; very few studies have focused on the impact of grammar instruction on immigrant ELLs in K–12 schools in the United States. Additionally, of the studies included in his review, most were narrowly focused on only a few grammatical forms (Bayley, 2009).
One of the few studies conducted with ELLs in a U.S. school setting examined the integration of grammar and content instruction. Doughty and Varela (1998) studied past tense acquisition by 34 middle school science students. The 6-week intervention involved students in writing science lab reports followed by interviews with their teachers. The interviews were designed to elicit responses in the past tense, and teachers were instructed to respond to student errors with corrective recasts. Posttest results showed greater gains on both oral and written measures for students involved in the intervention. However, the results of a 2-month delayed posttest failed to show that gains in accuracy of use of the past tense were maintained.
Because of the paucity of research in the area of explicit grammar instruction for ELLs in U.S. schools, few conclusions about the instructional approaches briefly outlined here can be made at this time. Additional studies using varied instructional approaches and varied measures are needed to provide a research base on which informed decisions can be made. Much of the research on grammar described up to this point was premised on the notion that we can support AE development by isolating and focusing on individual grammatical forms, such as the past tense. However, as we demonstrate in the next section, grammar instruction is often intertwined with classroom discourse as a linguistic resource in the development of AE.
Classroom Discourse
Instructional research on classroom discourse focuses on how AE is developed through social interactions in classroom contexts. This area of research includes studies drawn from discourse analysis, interactional linguistics, and sociolinguistics. Although this literature is set within the context of content classrooms, the primary emphasis in the research is on how effectively teachers engage students in academic discourse and the language of particular disciplines.
The importance of providing middle and high school ELLs with extended oral and written interaction is the emphasis of Verplaetse’s (2008) review of two lines of second language acquisition literature: (a) interaction studies and (b) the work of sociocultural linguists. Findings from both types of literature indicate that certain types of classroom interaction and instructional strategies provide ELLs with the practice necessary for developing AE, whereas other practices may be detrimental to AE development. Instructional practices that support student interactions with peers, teachers, and text include modifying teacher questions and responses to include follow-up questions, instructional conversations, and nonevaluative listening; modeling responses for beginning and early intermediate ELLs, challenging intermediate ELL students to produce frequent, extended utterances; and increasing student-to-student activities such as small group and pair group tasks. Verplaetse also emphasizes the importance of written interaction, citing literature on adolescent literacy development and the educational needs of adolescent ELLs. Her review indicates the need for “frequent opportunities for students to write extended text and to write on academic subjects other than personal narratives” (Verplaetse, 2008, p. 177). She concludes from the different literature cited that it is the responsibility of the teacher to provide opportunities for extended oral and written discourse in support of the AE development of their ELL students. Evidence cited from Verplaetse’s (2008) research also suggests that many teachers shelter ELLs from the types of interactions needed for AE development.
Similar findings are reported by Zwiers (2007), who presents results from a qualitative study of AE development in middle school content classrooms. As part of the study, the author explores the concept of AE, the social and cultural factors that underlie the acquisition of AE, and the teacher–student interactions that are likely to influence its development. In the classrooms observed, discourse patterns sometimes interfered with AE development. Among the teaching strategies observed, questioning was the most prevalent, followed by the use of gestures, examples, analogies, personifying, academic idioms, and linguistic enabling. According to the author, linguistic enabling is the least effective of these strategies, limiting students’ opportunities to develop AE. Linguistic enabling refers to teachers accepting oral or written responses without challenging students to elaborate or produce AE. Open-ended questions, although requiring more academic thinking and use of AE, were typically directed at the English proficient students in the class rather than the ELLs. Implications from the study findings emphasize the need for professional development in the cognitive skills required by specific disciplines along with the language patterns associated with these differing cognitive skills, “classroom activities that reflect complex thinking and language patterns valued in academia” (Zwiers, 2007, p. 113), patterns of discourse that negatively affect language and thinking, and the nature of communication in the classroom.
The nature of communication in secondary mathematics classrooms is described in a report of a case study by Brenner (1998). Two algebra classes within a small urban school district were observed over the course of 6 weeks. In one class composed of Spanish speakers with low levels of ELP, students received direct, sheltered instruction primarily through large group interactions. In a second class composed of English-proficient and less proficient learners, students interacted primarily in small groups. Both classes were taught by first-year teachers implementing a math curriculum that focuses on supporting student learning through small group instruction. Data were analyzed for type of interaction, type of mathematical communication, and language of interaction (English or Spanish). In general, the author found mathematical communication to be less extensive in the sheltered classroom where responsibility for carrying out mathematical discussions was not fully turned over to the students. The author concludes that although the small sample size and differences between the classes does not “warrant strong conclusions for policy” (Brenner, 1998, p. 125), it suggests the need for further study of the ways in which teachers and students negotiate and develop patterns of interaction that may inhibit or support mathematical communication.
Moschkovich (2002) further explores the situated nature of AE in her examination of approaches to teaching mathematics, focusing on shifting notions of the mathematical communication that are integral to mathematical learning. The first approach, tied to a conception of mathematics instruction as carrying out computations or solving word problems, simplifies the view of language learning as a focus on the vocabulary students need to learn to carry out these operations. The second perspective, based on a situational framework of language use, focuses on meaning making or, as Moschkovich puts it, a “shift in focus from acquiring words to developing meanings for those words” (p. 195). The third perspective, and the one she suggests more closely fits current theories of teaching and learning, is a situated-sociocultural view centered on student participation in instructional discourses (e.g., negotiating meaning, explaining solutions). From a sociocultural standpoint, the key issue in AE development is students’ ability to make use of different registers in the contexts of doing what they need to do in school (Schleppegrell, 2009).
Gibbons’s work (2003, 2008) on exploring the interaction of registers in AE development and content learning with ELLs demonstrates how ELLs move from talking about what they have learned in their science experiments in “here-and-now” registers to using more presentational, scientifically precise language necessary for producing oral and written reports (Gibbons, 2003). Gibbons illustrates her construct of a mode continuum by describing how over the course of a unit on magnetism students move from simple sharing of their experiences, or recounts, to creating explanations that do not depend on shared experiences. Teachers were able to move students from recounts to scientific explanations by explicitly focusing on genre expectations and the language needed to shift registers. In doing so, teachers focused on both new scientific knowledge (magnetism) and on the language needed to express this knowledge. Specific strategies used included recasting students’ language into more academic forms, describing and discussing the linguistic features of the academic register, reminding students to use the new language as they discussed scientific content with one another, and illustrating how the written language of science facilitates the presentation of new concepts (Gibbons, 2003).
Snow (2008) discusses another area of academic discourse that involves developing thinking and speaking patterns important in the sciences by engaging elementary students in “inner state” tasks. Such tasks include asking students to discuss how someone else feels or thinks about an issue. This can be done in the context of a story the child is reading, where an adult engages the child in conversation about character perspectives in the story. Snow points out that they can discuss why there are differences in what characters know and think, as well as what is imaginary, true, or perceived to be true by each character. She explains that such skills are necessary for discussing facts, opinions, hypotheses, arguments, claims, truth, reality, and points of view in scientific discourse. Snow concludes that these activities are valuable for early-elementary ELLs because developing these tools early on helps smooth the transition to the AE of science.
A second illustration of how discourse skills can be taught comes from Schleppegrell’s (2004) description of the development of AE through student involvement in classroom sharing time. As a young child explains information and experiences unknown to the rest of the class, he or she is assuming the role of an authority on the topic. The student learns to form topic sentences and detailed, chronological descriptions. Although the teacher may pose questions in prompting the child to implement certain discourse markers and patterns, the interaction is not conversational. In fact, Schleppegrell explains that the repeated question–answer interaction between a teacher and student gives the student practice in developing a “lexically explicit” expression, and together the student and teacher create a focused, expanded discourse (p. 34). Schleppegrell further notes that such class presentations also require the student to explain decontextualized information, a skill of academic discourse.
How science language is learned through discipline-specific activities was the question behind a study on how one “science teacher socializes his students into the science register—thereby teaching students science language and meaning—which links what they do and observe with the theory being taught” (Mohan & Slater, 2006, p. 303). The study methods consisted of participant interviews and observation of teacher–student interactions in one class of 30 ninth-grade students representing a mix of native English speakers and ELLs. From the perspective of SFL, the authors examined the discourse patterns demonstrated by the teacher and students in terms of three levels of social practice: knowledge structures, question–answer relationships, andlexical cohesion. Analysis indicated that the teacher modeled AE by recasting student responses from practical experiences to theoretical understanding of science, from incorrect and informal discourse to the language of scientists, and from imprecise to more precise use of vocabulary. Mohan and Slater (2006) conclude from their study that conducting research from an SFL perspective has the potential to build our understanding of how to develop students’ ability to use the language of science and, therefore, students’ ability to understand science content.
The research on AE instruction and development presents a complex view of what can be achieved through teacher behavior. Research does not yet provide a definitive reading of what constitutes best AE instruction or even whether AE can be successfully taught. Gee (2005) has argued, for example, that meaning in language is tied to situation and experience, which cannot be replicated easily in the classroom setting. The AE development he proposes is intended to mimic academic discourse through discipline-specific practices such as the extended turn taking common to science discussions. Gee’s proposition is consistent with new recommendations that call for AE educators to focus on “indirect intervention where learners can acquire language experientially” (Understanding Language, 2012, p. iii). Arguably, such recommendations might be considered evidence of an “emerging willingness” among AE researchers to consider the contextual constraints of the social and personal (Ortega, 2012, p. 215) as they relate to conceptualizing AE, instructing AE, and developing teacher practice in AE.
Implications for Teacher Knowledge and Professional Development
We have discussed the complexity involved in conceptualizing AE and the varying perspectives on teaching or developing AE. These differing viewpoints have important implications for how teachers are educated about AE and the ways in which they apply their knowledge in the classroom. Teacher education is the purview of a broad array of institutions, including colleges of education, school districts, schools, state and federal assistance centers, as well as other non- and for-profit entities. These institutions often have conflicting approaches to educating teachers about language and language instruction that are reflective of the larger debate on what constitutes AE and how ELLs best learn this language. Despite these differing views and approaches, improving the language and literacy skills of ELLs will depend on finding ways to deepen all teachers’ knowledge of language and language development (Adger, Snow, & Christian, 2002; Bunch, 2013).
Any discussion on educating teachers about AE instruction for ELLs involves identifying the understandings and skills that teachers need to help students comprehend and produce the language of the academic disciplines. Bailey et al. (2010) refer to the need for teachers to engage in “thinking and acting linguistically” (p. 608). Wong Fillmore and Snow (2000) have described thinking linguistically as teachers’ knowing the linguistic features and discourse structures of their academic disciplines as they relate to the AE tasks required of students. Acting linguistically means that teachers are analyzing the discourse structures and rhetorical devices that are prevalent in texts from their discipline and are explicitly teaching these structures and devices within the context of subject area instruction (Wong Fillmore & Snow, 2000).
Educating teachers to think and act linguistically requires reconsidering the content of the coursework and experiences offered to teachers at both the preservice and in-service levels. Perhaps the most rigorous proposal for reforming preservice teacher education is Wong Fillmore and Snow’s (2000) call for the “systematic and intensive preparation” of teachers in educational linguistics (p. 4). They recommend that all teachers understand how the language modalities (reading, writing, speaking, and listening) function across different academic contexts and develop the linguistic knowledge and skill to help students develop fluency with each modality within the academic disciplines they teach. Teachers should also have the linguistic knowledge necessary to select instructional programs and materials within their teaching domains that support students to develop increasingly sophisticated language skills and to plan instructional activities that provide opportunities for students to use language in new and increasingly complex ways.
To prepare teachers to engage in this type of instruction, Wong Fillmore and Snow (2000) suggest that additional educational linguistics courses be required as part of the teacher preparation curriculum. Although most authors agree on the importance of the teacher knowledge base Wong Fillmore and Snow call for, they contend that the addition of significantly more coursework (seven additional courses) to the teacher preparation curriculum is not a practicable solution (Baca & Escamilla, 2002; Richardson, 2002; Walker, Ranney, & Fortune, 2005). Other authors have suggested alternative approaches to adding coursework such as having language specialists coach teacher educators on how to integrate language education into existing teacher education courses and distributing the linguistics courses throughout undergraduate, graduate, and professional development programs (Baca & Escamilla, 2002). Baca and Escamilla (2002) also point out the important role of state content standards in guiding instruction. Because content standards drive curriculum, instruction, and assessment, the integration of language standards with content standards should be considered and would have important implications for AE instruction.
A case study of a course designed to prepare elementary and secondary teachers in language-sensitive instructional practices for ELLs offers another alternative for preparing teachers to address AE (Walker et al., 2005). The University of Minnesota course was designed as a one-credit seminar to facilitate teachers viewing themselves as language teachers as well as academic subject matter teachers. What makes this course unique is the differentiation of course content by teaching context, including grade span and discipline. Although it is unclear whether course learning transferred to classroom teaching, the authors were encouraged that teachers’ exposure to the language demands of their grade level or content area resulted in enhancing understanding of the differences between social and AE (Walker et al., 2005).
Another effort to develop teacher understanding of AE focused on the use of teacher inquiry with graduate students in teacher education. A study of the approach to inquiry found that teacher research projects can influence teachers’ perceptions of their preparedness to teach AE (Merino, 2007). The teacher research included both a case study and an intervention study. Students and faculty who were surveyed regarding these projects found that, among other things, an effective feature of this approach was the “focused attention on the academic language demands of the discipline standards, combined with guided reviews of the literature to identify pedagogically sound ways to address these demands” (p. 6). The research findings led to the implementation of a Credential/MA program at the University of California-Davis that focused on preparing teachers to work with ELLs.
Although initial efforts to educate teachers on AE during their preparation coursework are promising, Tellez and Waxman (2004) point out that preservice training is not sufficient; school districts must also provide comprehensive, long-term professional development programs that extend teachers’ knowledge of and ability to support ELLs in learning AE. A research synthesis on teacher professional development found very few studies that address the effectiveness of professional development programs for teachers of ELLs (Knight & Wiseman, 2006). Furthermore, the research that does exist infrequently addresses AE directly or addresses it as one component among many. The studies selected for inclusion in this review focus on professional development on teaching AE within the academic disciplines of science, ELA, and history, as a growing body of research indicates that teachers’ perceptions of the usefulness of professional development are often dependent on the degree to which their specific discipline or teaching context is addressed in the professional development (e.g., Garet, Porter, Desimone, Birman, & Yoon, 2001). Accordingly, professional development that contextualizes learning about AE to teachers’ subject matter and grade level is perceived as more useful than professional development on AE in general (Ballantyne, Sanderman, & Levy, 2008).
In the first of a series of studies, Hart and Lee (2003) describe the results of a teacher professional development intervention aimed at helping 53 third- and fourth-grade teachers at six schools improve science and academic literacy achievement for ELLs. The professional development prepared teachers to implement two science and language instructional units at two grade levels over the course of an academic year. Four full-day workshops were devoted to science inquiry and integrating academic language and literacy in science instruction. The authors state that a “variety of language functions (e.g., describe, explain, report, draw a conclusion)” (Hart & Lee, 2003, p. 483) were part of the curriculum as were lessons on key academic vocabulary.
After the first year of professional development, the researchers concluded that teachers developed a “broader conceptualization of literacy in science instruction” (Hart & Lee, 2003, p. 489). This change from a superficial understanding of science literacy was evidenced by teacher remarks on language and literacy that were more diverse, greater in number, and more integrated with science content. Although there was no statistically significant change in observed reading and writing teaching practices, teacher use of linguistic scaffolding increased. The authors indicate that the participants did not report increased knowledge about or skill in fostering literacy possibly
because increased knowledge about literacy resulting from professional development created in teachers an awareness of the need for improvement in a domain in which they formerly felt relatively confident . . . professional development may have served primarily to make them aware of the limits of their knowledge. (Hart & Lee, 2003, p. 493)
In a subsequent study, Lee, Deaktor, Enders, and Lambert (2008) examined the impact of the professional development intervention on the integration of science inquiry and AE on student outcomes. Using project-developed science tests for students at upper elementary grade levels, the researchers studied students’ “inquiry abilities” over the 3-year period of the intervention. Significance tests of mean scores between pre- and posttests indicated statistically significant increases each year on all measures of the project-developed tests. The researchers also conducted comparisons of the project-developed test items with items from the National Assessment of Education Progress and the Trends in International Mathematics and Science Samples. These comparisons suggested overall positive performance by the students of teachers who had participated in the professional development intervention at the end of each school year. Of particular interest were the substantial gains made by low achieving, low socioeconomic status, and ESL-exited students, which exceeded those made by mainstream students, and thus provide some evidence for the effectiveness of the intervention for reducing the achievement gap between monolingual mainstream students and ELL students.
A separate but related study (Stoddart, Pinal, Latzke, & Canaday, 2002) involved a 5-week professional development intervention focused on helping 24 elementary school teachers of predominantly Latino ELLs integrate inquiry-based science and AE instruction. The authors describe AE from a functional perspective, indicating that academic subjects, such as science, have a linguistic register—norms and patterns of language use essential to the practice of the discipline (Halliday, 1978). Stoddart et al. (2002) argue that the use of language functions associated with science, such as formulating hypotheses and proposing alternative solutions, is fundamental to the process of inquiry science. The researchers developed a five-level rubric to assess teachers’ understanding of science and language integration as measured by pre- and postintervention interviews. Prior to their participation, the majority of teachers viewed themselves as well prepared to teach science or language, but not both. After their participation in the program, the majority of teachers believed they had improved in the domain in which they had initially felt less prepared. This change typically involved a shift from a restricted view of the connections between inquiry science instruction and second language development to more elaborate reasoning about the different ways the two could be integrated.
Research on professional development provides support for a functional linguistics approach to examining teaching and learning of AE. Analysis of ELL student writing has indicated that a limited focus on sentence-level grammatical errors or overall structure is not sufficient to address the kinds of errors that occur in ELL student writing. Teachers need deeper knowledge of AE features within subject area domains to address ELL writing difficulties (e.g., Schleppegrell, 2004). Based on these premises, a team of educators and researchers (Aguirre-Munoz, Parks, Benner, Amabisca, & Boscardin, 2006) created a professional development intervention that provided teachers with an alternate approach to addressing the issues occurring in ELLs’ academic writing. SFL was used as the basis for the professional development intervention because it “provided direct guidance for analysis of language use in texts containing academic language and guidance as well in the analysis of student writing in a manner that more directly fosters the development of students’ academic language development” (Aguirre-Munoz et al., 2006, p. 4). Using this approach to AE, the team developed 4 days of training modules for 12 teachers from five school districts in southern California with varying levels of teaching experiences. Because cohesion in writing is a difficulty for many ELL writers, the modules focused teachers’ attention on the linguistic features of a specific genre of text—in this case, a character study—that promoted cohesion across sentences within a particular genre of writing. Strategies for creating cohesion, including the use of nominalization, long noun phrases, and prepositional phrases, were emphasized.
Findings from this study include teachers’ reports of increased levels of understanding of ELLs’ writing development, improved analyses of student writing to inform instruction, and more specific feedback on student writing. There were also qualitative differences in teachers’ identification of problems in student essays. Whereas pretest feedback on student writing had focused on either mechanics (e.g., spelling, sentence fragments) or on global issues (e.g., organization), posttest feedback was much more specific and related to training content. This feedback included pointing out weaknesses in noun phrases, verb phrases, and connections between clauses. In general, a trend emerged in which teacher feedback moved from vague (e.g., “develop ideas more”) to specific (e.g., “expand noun phrases”). The researchers concluded that the professional development was effective in training teachers to examine student writing from a functional linguistics perspective and to generate instruction that could improve student understanding of a written character study.
An important contribution of this study (Aguirre-Munoz et al., 2006) was the effort to ground the professional development in an approach to AE that could address both teacher development in understanding AE and provide teachers with instructional tools to support ELLs’ AE development. The use of functional linguistics allowed the researchers to achieve both goals. The study showed improvements in teacher understanding of AE and in the level of specificity of the feedback teachers were able to provide on student writing. The researchers did not study teachers’ implementation of the new feedback strategies in the classroom, nor did they examine whether student writing improved after instruction using functional linguistics.
Another example of professional development grounded in a functional linguistics approach to language comes from work with middle- and secondary-level history teachers in California (Schleppegrell, 2005; Schleppegrell & Achugar, 2003). In ongoing professional development, researchers used functional linguistics to help history teachers develop a metalanguage to talk about the ways in which language constructs time, cause, and the author’s interpretation of events in history. Schleppegrell and colleagues also developed strategies teachers could use to engage students in the language patterns of their history texts. Using these tools, teachers then worked with students to deconstruct texts and explore the ways in which language creates meaning in history. An important finding from this work is that engaging students in close reading of their texts using functional linguistics enhanced students’ interaction with their texts and participation in class discussions. Both qualitative and quantitative findings indicate that the approach improved students’ reading and writing skills in history as well as their understanding of historical content (Schleppegrell, Greer, & Taylor, 2008).
Research on professional development in AE is in its infancy, but the literature reviewed suggests that high-quality professional development can provide meaningful learning experiences for teachers on AE within subject area contexts. In a recent review of research on professional development for mainstream teachers of ELLs, Bunch (2013) posits the need for teachers to develop pedagogical language knowledge, which he describes as the “knowledge of language directly related to disciplinary teaching and learning and situated in the particular (and multiple) contexts in which teaching and learning take place” (p. 307). Moving forward, Bunch calls for professional development that conceptualizes AE as action and a vehicle for engaging with content standards. An important consideration for future research is how AE is defined and operationalized within professional development interventions. It is not enough to show improvements in teacher understanding and application to practice without carefully describing how AE is being conceptualized within a particular content area and within instruction. These studies will also need to provide sufficient information on the nature of AE as it is being operationalized within the professional development interventions so applications for policy and practice can be made.
Conclusions and Research Priorities
Our primary purpose in this review has been to describe the construct of AE and the ways in which it has been addressed in instructional and teacher professional development research. According to much of the literature, AE cannot be described apart from the context in which it is used. The contexts exemplified range from school or classroom settings at different grade levels to teacher expectations for academic reading, writing, or discussions to the types of texts or discourse that are brought into play during instruction. Related to context are the different registers of AE as it is used in science, mathematics, social studies, language arts, and other content areas, as well as the language used for classroom routines and to interact socially within school settings. A consistent message from the literature is the importance of describing patterns of language use as they relate to particular situations and purposes. Finally, common to many of the approaches to conceptualizing AE are descriptions and examples that document the linguistic competence necessary for accessing content through listening, speaking, reading, and writing tasks.
Several conclusions can be drawn from this literature on AE instruction. First, the literature demonstrates that instruction may be influenced by the particular view of AE held. This view, in turn, is determined by the dialogue that occurs within and, to a limited extent, across educational communities. For example, educational communities that focus on the social functions of language emphasize instruction focused on classroom discussion of academic concepts and the ways in which students make sense of these concepts. Through this discussion, teachers support students’ AE development by helping them use language to clarify their thinking and make connections to written and oral texts.
Second, the literature shows that improving students’ AE is a concern not only for educators of ELLs but also for educators from subject areas such as science, mathematics, social studies, and language arts. Experts tend to agree that AE is more difficult to learn than other language registers; not only ELLs but also students from a broad spectrum of backgrounds have difficulty learning it. Thus, formulating and providing educational experiences that promote the learning of AE is critical for all students, including ELLs. What is not clear from the available research is how these educational experiences should vary depending on students’ English proficiency levels, literacy abilities, and prior education.
A third conclusion that can be drawn from the literature on AE instruction is that there are differences in AE across content areas that must be accounted for in instruction. Although common dimensions of AE permeate all content areas, such as complex vocabulary and long noun phrases, specific language demands are unique depending on the task and discipline. For instance, whereas students are expected to use language to substantiate technical claims in science content, they need to use language to analyze events or make arguments in social studies. In all of the content areas discussed, there is an expectation that the teacher address lexical, syntactic, and semantic features specific to the discipline. Instruction focused only on an individual language feature, such as academic vocabulary, does not fully address the challenges of AE acquisition. AE includes not only vocabulary but also complex grammatical structures and discourse patterns that contribute to cohesion and coherence in communication and that vary according to the purpose or function for which language is used. Therefore, all teachers need a deep understanding of the language demands of their discipline to ensure students can access the content. According to the literature, professional development has the potential to challenge and change teacher beliefs and practices about language when it is ongoing and directly relevant to the content teachers are teaching.
A final implication of the AE instructional literature is to examine more carefully the effect of classroom interactions on AE acquisition. The literature gives some support to the idea that opportunities for engaging students in AE means rethinking the ways teachers approach both language and content instruction. It calls for a better balance between teacher and student talk, and instruction in which modeling and scaffolding of academic discussions and questioning techniques occur. The literature also demonstrates the importance of instruction that provides opportunities for students to use AE as it is used in professional and academic communities, and to develop students’ metalinguistic awareness of AE features. Of relevance to this discussion is the possible role of literacy engagement in promoting AE. Guthrie (2004), Lindsay (2010), Cummins (2011), and others confirm the importance of literacy engagement in predicting student literacy outcomes. This research suggests that teaching AE is likely to be more effective in a context that builds student perception of their literacy competence by providing generous access to print and interactions with literacy.
These implications from the literature on conceptualizing and providing instruction in AE lead to the third area of inquiry, the issue of educating teachers. The literature recognizes the importance of identifying the specific skills and knowledge necessary to develop a deep understanding of AE and implement AE instruction. However, research on teacher professional development on AE is just beginning to document and evaluate results of varying approaches to developing teacher knowledge and skill in AE. Although there are somewhat clearer expectations for ELL teachers on the AE knowledge and skills they should possess, the literature indicates that AE language development is a responsibility not only of ELL teachers but of general education teachers as well. A major implication of the professional development literature is that more rigorous and ongoing research on professional development in the area of AE teaching and learning is needed to provide guidance to effective practice.
Based on the conclusions and implications, the authors recommend several priorities for research. A first priority is to develop a systematic approach to investigating the AE demands of teacher–student interactions, teacher talk, textbooks, and assessments. Recent efforts in this area, such as the Framework for English Language Development discussed previously (e.g., CCSSO, 2012) suggest potential to result in a commonly accepted framework of AE that could be used to guide states and districts in making instructional, professional development, and assessment decisions. Second, research should focus on identifying the specific demands AE places on ELLs and the ways in which AE is used in different school settings. A third priority is to examine the impact of different approaches to AE instruction on AE acquisition in ELLs. Finally, more rigorous documentation and evaluation of professional development programs designed to develop teacher knowledge and skill in AE is needed. For each of the potential investigations outlined here, the expectation is that research into AE will consider the influence of students’ prior educational experiences, ELP, culture, and home language. ELLs are not a homogenous group. They include students who speak one or multiple languages, students of privilege and poverty, students of diverse cultures, experiences, and circumstances. Research has “barely begun to investigate” the ways in which AE can be instructed or developed for the broad range of learners of English (Ortega, 2012, p. 220). It is clear though, that attention to their different characteristics is an essential feature of meaningful research in this area.
Footnotes
Notes
Authors
PATRICIA A. DICERBO, EdD, is a research scientist and English language learner (ELL) specialist at The George Washington University Center for Equity and Excellence in Education, a Center of the Graduate School of Education and Human Development, Washington, DC, USA; e-mail:
KRISTINA A. ANSTROM, EdD, died September 2013 from amyotrophic lateral sclerosis. She was assistant director and a senior research scientist of The George Washington University Center for Equity and Excellence in Education (GW-CEEE), a Center of the Graduate School of Education and Human Development. She was a teacher, researcher, and leader in the field of education for more than 25 years. At GW-CEEE, she directed numerous applied research and technical assistance projects relating to English language learners (ELLs). Her research emphasized inclusive curricula and learning environments for ELLs. She most recently led a multiyear study funded by the Bill and Melinda Gates Foundation to identify the academic language needed to access state biology and mathematics standards. She authored numerous reports, articles, and books, including a textbook for ELLs, Keys to Learning, that has been widely disseminated and used by teachers and teacher educators. She earned a doctorate in curriculum and instruction from The George Washington University.
LOTTIE L. BAKER, EdD, is a senior English language fellow at the U.S. State Department where she is serving as educational advisor to the Thailand Ministry of Education, English Language Institute. She is an English language learner (ELL) specialist and consults on issues such as program models, teacher professional development, and curriculum design. Previously, she worked as a senior research associate at The George Washington University Center for Equity and Excellence, where she conducted program evaluations, provided technical assistance to state education agencies on matters related to ELL policy, and led teacher professional development activities. She has been an ESL classroom teacher at the primary and community college level, and she has taught graduate-level teacher education courses in ESL methodology and second language acquisition. She holds a doctorate in curriculum and instruction from The George Washington University.
CHARLENE RIVERA, EdD, is a research professor at The George Washington University Graduate School of Education and Human Development and executive director and founder of The George Washington University Center for Equity and Excellence in Education, Washington, DC, USA; e-mail: