Name the function, not the grammar
Micro-Feedback in the Language Classroom
Using metalinguistic terminology when giving students feedback might be the surest way to gatekeep student communication.
“You used the wrong subjunctive trigger.” “You used the imperfect you should have used the preterite.”
These ARE pieces of feedback that one could give their student, but does this feedback actually do anything for students? Or when they hear the feedback do they get lost in the metalinguistic jargon because now not only do they have to try to understand and produce their L2, they have to try to reconcile their ability to communicate with their ability to recognize the names, uses, and forms of metalinguistic items. Functionally naming the language we use in class aims to inform students of what they are trying to say rather than how they said it wrong. Instead of “Use the preterite here instead of imperfect,” I might say “You used the form that describes background conditions/actions, you want to use the form that shows that an action happened once.” It’s the same feedback, but a different perspective. It shifts the vibe from “you did something wrong” to “you communicated something, let’s work to make it more clear.”
Every language form is used for SOMETHING: hypothesizing, describing, comparing and contrasting, narrating. The list goes on. So naming language forms functionally starts with considering what is the actual purpose of communication? Another way to think about it: If my student doesn’t know the term “subjunctive” how can I explain its use for our present context to make it make sense. For something like the subjunctive, I may call it “the form of desire,” “the form of recommendation,” “the form of uncertainty,” or something similar; depending on the communicative goal for the class/unit.
Dynamic Assessment treats formative assessment, summative assessment, and instruction as intrinsically linked: every instructional move doubles as a assessment, and every assessmen
t doubles as a teaching moment. The mechanism behind that link is mediation: the teacher intervenes in the learner’s process, not just at the end of it, and the intervention itself reveals what the learner can do with support, not just what they can already do alone. That’s the whole point of working in the ZPD — you’re not measuring a fixed score, you’re testing how much scaffolding closes the gap, in real time, mid-task.
Mediation only works if it lands in language the learner can act on immediately. That’s where functional naming stops being a nicety and becomes load-bearing. If the mediating move requires the student to first decode a grammar term before they can even locate what you’re talking about, the mediation has already failed — you’ve inserted a vocabulary lesson into a moment that was supposed to be a communicative one. Functional naming keeps the feedback in the same register as the task itself, so the student can take the mediation and immediately try again, which is the entire mechanism Micro-Feedback depends on.
Mediation itself comes in two grains. You can mediate the same communicative function across the whole class — everyone gets fed the same frame, the same job, and feedback tracks how each student executes that one function. Or you can individualize: different students get different mediation depending on what they specifically need in that moment. Neither is more correct; they’re just different tools for different goals, and which one you reach for says something about whether you’re checking a function class-wide or meeting individual gaps.
In my classroom, mini-feedback happens two ways. One is form-focused mini-stories, where students label the language themselves with functional naming — that’s a blog post in its own right, with its own mechanics worth unpacking separately. The other is sentence stems, which is where I want to stay for now.
On Fridays we watch Tierra Incógnita. I want the viewing to mostly be about enjoying a Spanish TV show, but students need a task to anchor their attention, so I bring out whiteboards and sentence stems — this is the uniform-function version of mediation: everyone’s tracking the same job, just with their own content.
One example: as we watched an episode, I had each student pick a character to follow. On the board: “X quiere que Y…” One sentence. Students’ first job is just to notice — which character wants something from someone else? They’ve got the frame; they just need to find a moment that fits it.
As I walk around the room checking whiteboards, I’m only giving feedback on the subjunctive word in the second half of the sentence. So when a student writes “Eric quiere que sus padres aparecen,” I tell them: Great job! To make it clearer, we’re going to use aparezcan — because that’s the hope form. Eric doesn’t know if his parents will return. We use the hope form when we want someone else to do something, but we don’t know if they’ll do it.
That one fixed location, the subjunctive verb, every time, is what makes the whole loop work. I know exactly where to look on every whiteboard, which means I can celebrate what a student’s already doing independently, gauge how much support they need to close the gap, and move to the next student without losing the thread of either the show or the feedback.
This is what not gatekeeping actually looks like, down at the level of a single verb, mid-task, on a Friday afternoon. "Use aparezcan, that's the hope form" takes the same half-second as "use the present subjunctive”, it's not slower, it's not softer, it's not letting anything slide. It's just feedback that a kid can act on without first passing a vocabulary quiz to understand what I said.