When I began revamping the visual arts curriculum for HKUGA College in Hong Kong, I observed a problem in junior form visual art students that needed addressing. Students could describe what they saw in an artwork but couldn’t yet argue what the artwork was trying to express. The fix was a structured framework which I had previously integrated into art curriculums at the American School of Mumbai and the International School of Beijing — Form, Theme, Context, developed by Dr. Renee Sandell. In online theory assignments I designed, this was reinforced by an AI teaching assistant that asked one question, repeatedly, until it became reflexive: what visual evidence supports that? In a new school environment, with a different style of language learner, a question arose I couldn’t yet answer: does that habit survive once students leave the framework that built it?
I can answer it now — cautiously, taking care not to make too grand a claim.
When art students arrive in the senior form VA course, the analytical framework changes out from under them. The DSE mandates Ronald Feldman’s classic DAIJ method — Describe, Analyze, Interpret, Judge — for the visual arts syllabus, a structure not optional for a department. So, the three categories junior form students spent three years internalizing — Form, Theme, Context — disappear from the paperwork. The question was never whether DAIJ is a good framework. It’s whether the underlying habit — don’t assert, argue from evidence — would survive the swap, or whether three years of work would evaporate the moment the labels changed.
My answer was to build FTC language into a research guide based on DAIJ for S4 students. I’ve argued before that Feldman is good at breaking down what to do and Sandell at how to do it. These approaches complement each other conceptually and provide both new and familiar language for the students. Inside the guide is an AI agent like the junior form’s Yue. Like the junior form version, the agent doesn’t hand students content. It comments on the efficacy of the writing: whether a claim is supported, whether the leap from observation to interpretation is earned. Same job, different framework wrapped around it.
Two S4 portfolios this year gave me something tangible, rather than hope.
Fay’s project began with a DAIJ analysis of the Guerrilla Girls’ 1989 billboard — the reclining nude with a gorilla mask, the question “Do women have to be naked to get into the Met. Museum?” printed over a hard statistic. What struck me about her report wasn’t the vocabulary, though it’s confident. It was the structure of the argument. She doesn’t just note that the piece juxtaposes a classical nude with a gorilla mask; she explains what the juxtaposition does — clashes the viewer’s expectations, then uses that friction to force engagement with the statistic underneath. She places the piece in its own history, including the detail that the Public Art Fund initially rejected the proposal as “too clear” and “too aggressive,” and she uses that history to support her reading rather than just decorating it. And in the Judge section, she does something a classmate’s Frida Kahlo analysis from the same cohort doesn’t quite manage: she reflects on how the piece expanded her own sense of what an artwork is allowed to be and lets that reflection point somewhere — toward a piece that’s part dollhouse, part protest.
That’s the second layer worth naming. Working things out in her workbook, Fay’s mind-mapping session on gender, identity, and bodily autonomy — feminism, LGBTQ+, race, all sketched out with a kind of restless, over-ambitious energy I recognize from a lot of good S4 students — turns directly into a studio decision: three identical porcelain-doll figures in Cinderella dresses, two of them wearing wolf masks, staged in a claustrophobic pink backroom. The analytical habit isn’t just producing better essays. It’s producing better decisions about what to make and why. A classmate’s process shows the same transfer at a different level of fluency — her Kahlo analysis of the sun/moon split and the cracked earth becomes a mirror-and-broken-glass piece built around the same day/night, wounded/hopeful duality she wrote about.
I want to be careful here, the way I was the first time. Two portfolios are not evidence that a tool works. Both students went through three years of junior form before they touched the DAIJ assistant, so I can’t isolate what the scaffolding contributed versus what three years of any rigorous framework would have built. What I can say is narrower and more useful: the specific habit the junior form program was built to install — evidence before interpretation — is visible in both portfolios, inside a framework the school didn’t choose and inside a tool designed on purpose to carry that habit across the gap rather than let the format change erase it.
That was always the harder design problem. Not whether a framework works, but whether the thinking underneath it can survive a framework it wasn’t built for.