TEACHING ACADEMIC LITERACY
For Tristan Reynolds, learning the right kind of academic language is central to student understanding of any subject.
Developing subject-specific language
The fundamental purpose of teaching students language, whether communicative or academic, is to empower them to communicate with others more clearly, cleverly, and confidently. Understanding what language people actually use in specific fields of study is therefore one of the highest-value skills that we can give to students.
As the number of international schools grow, teachers increasingly work with multilingual students who require additional support to develop their academic use of English across the whole curriculum. This means that they need specific practice with the types of language chunks that appear in specific academic disciplines–historians don’t speak or write like chemists, and poets use an entirely different idiom altogether. For students to achieve academically in these disparate fields, they need specific language training for each discipline.
Identifying key academic language
Identifying and prioritizing key discipline-specific language can present issues. While some national and international curricula identify particular concepts and vocabulary for students to learn, these items aren’t always embedded in useful chunks of language. Individual wordlists, complied by teachers or schools may be idiosyncratic and may miss word groupings now being commonly used in school textbooks, research papers and websites.
However, borrowing some tools from computational linguistics can help educators prioritize the most important chunks of language their students will need by identifying the most frequently-used collocations, phrases, or even sentence structures in a discipline. Teachers, school leaders, and school systems can draw on relevant bodies or corpora of language to help them develop students’ use of academic English.
Using frequency analysis
Developing these resources involves the use of frequency analysis, in which a reader (usually a computer) notes the most common chunks of language and their relative frequency within the relevant body of language. Frequency analysis can produce lists of the most and least commonly used language chunks. This means that teachers, school leaders, and curriculum coordinators will be able to identify the particular language chunks that appear most frequently within a corpus drawn from a specific discipline (i.e., from articles published in journals of chemistry, or from literary memoirs, or from mathematical proofs). This approach allows teachers to prioritize more common chunks of language for their discipline, which is particularly important for secondary teachers.
This technique helps educators prioritize useful language chunks by discipline. For example, the collocation ‘conception of’ has a frequency rate in the MICUSP database that varies quite widely by discipline. In English, it has a frequency of 0.57 per 10,000 words; in education, a frequency of only 0.06 per 10,000 words; in philosophy, a frequency of 1.23 per 10,000 words. It doesn’t appear at all in the database’s corpus of biology papers. This suggests that, for TOK (Theory of Knowledge) teachers discussing how philosophers develop different conceptions of a theory could be important and learning the meaning useful.
Resources for language-building
Educators have access to a number of different bodies of contemporary language usage. The MICUSP database is quite small, but there are others such as the Corpus of Contemporary American English (COCA English), which indexes many more types of text, including television scripts, online articles & comments, and contemporary literature. Educators can use the database most suited to their particular context, or cross-reference between corpora for the weighted results, as summarised below:
Publicly-Available Corpora of English | |
This is the largest commonly-used corpus of ‘all’ American English. This makes it a good choice for assessing important language chunks for public communication that students do (for example, in public speeches, project-based learning experiences, extracurricular competitions, etc.). | |
This is a large corpus of historical American English (going back to the 1820s). This makes it a useful tool for figuring out which language chunks will be needed for students to read historical documents from the relatively recent past, though it’s of limited use when looking at sources from the Early Modern, Medieval, or Classical periods. | |
The Hansard Corpus is a collection of speeches given in the British Parliament from the early 1800s to the early 2000s. It can be useful for augmenting other corpora (like the COHA) to figure out which chunks of language students will need to understand historical texts from the past two hundred years. | |
The TIME corpus is a collection of articles from over nearly a century of that magazine’s publication. This makes it useful for figuring out which language chunks students will need to understand popular writing in a variety of genres. | |
The Wikipedia Corpus offers the chance to look at specifically online English, which is helpful for supporting students who want to engage in online discourse across the curriculum. | |
The result is a collection of resources that can help guide teachers in supporting the development of subject-specific literacy. All teachers have to teach components of literacy. Whatever their subject discipline, taking an approach that identifies the highest-frequency chunks of language in a discipline can get students writing earlier in a course, more effectively in a course, and more confidently across all their courses.
Tristan Reynolds is a curricular and instructional leader, developing frameworks, practices, and documentation that support educators integrating Project-Based Learning and Education for Sustainable Development across a school’s curriculum.
FEATURE IMAGE: by Getty Images For Unsplash+
Support Images: by Getty Images For Unsplash+ & M. Faisal For Unsplash+
