It’s been a while since I got some thoughts down about the current state of CALL and TELL in the era of AI. Here are a few of them, by no means exhaustive. Interested to hear others’ opinions on these considerations as usual.

What is the role of CALL (computer assisted language learning) or TELL (technology enhanced language learning)?

It’s to “assist” or “enhance” the learning of languages.

This has normally involved judiciously providing students with or depriving them of technological affordances such that they are better able to learn the target language.

My working definition of “learn a language” has been “to use and understand the language in a variety of situations without assistance”.

So technological assistance is a kind of scaffolding that will eventually be removed so that the learner no longer requires it and can use and understand the language without it.

When assessing whether a student has indeed “learned” the language, we often use technology while tightly controlling students’ access to it (consider standard assessments such as the TOEFL or IELTS).

The Internet, smartphones, and GenAI have challenged the assumption that the technological scaffolding will ultimately be removed, because most students now carry a device everywhere that can understand, translate and generate almost any language. In addition, Big Tech companies such as Google and Microsoft are baking AI into most of their existing communication services, both synchronous and asynchronous. There are a plethora of browser plugins to translate, transcribe or correct written and spoken webpage content. AI affordances affect not only “computer as a tutor” models of CALL/TELL but also “computer as a tool” models (e.g. automatic captioning and translation in Zoom meetings).

While we have traditionally separated the four skills of language learning into reading, speaking, listening, and writing, AI is now collapsing those distinctions, since any written text can instantly become a spoken one, and any spoken one a written one. Controlling students’ access to these technologies is difficult if not futile. Accessibility presents another curveball. You can’t “judiciously deprive” a hearing-impaired student of captions on a video, for example, even if that might be a legitimate exercise for non-hearing-impaired students.

It’s not hard to imagine a future where “smart glasses” become as ubiquitous as smartphones, heralding a new era of augmented reality, where information we cannot see or control is displayed before students’ eyes. Again, Big Tech companies like Meta and Google are already working on and pushing these products.

The struggle to understand the implications of all of the above for language teaching and learning approaches, methodologies, practices, and policies is a daily one; made harder by the fact that the ground seems to be shifting under our feet. Updating materials, syllabuses, and curriculums takes time, and technology moves much faster than educational bureaucracy.

Further Reading:

• What is Computer-As-Tutor vs. Computer-As-Tool?
• VLC Introduces Cutting-Edge AI Subtitling and Translation Capabilities
• Enabling and configuring translated captions (Zoom)
• AI glasses (Meta)

My Recent Papers:

• The Effect of AI-Powered Technologies on the Motivation to Learn Foreign Languages: A State-of-the-Art Review and Meta-Analysis:
• Using GPT-3 as a More Knowledgeable Other to Provide Feedback on English Language Learners’ Spoken Transcripts