There is this argument between artistic style and visual language that is becoming fashionable. It says that style is shallow, visual language is deep. This school of thought admonishes artists to stop trying to be recognizable and just focus on being rigorous and is driven by the assumption that when you develop a visual language, the style emerges as a byproduct. The implication of this is a one-way pipeline: investigation → language → style, in that order. But most serious practices are recursive, not sequential. You create something, an accident becomes a habit, the habit becomes material you then investigate on purpose, which produces a new accident. Style and language aren’t stages, they’re feeding each other constantly. Telling artists to “investigate first, style later” is a pedagogical convenience and not how anyone’s actual studio history reads.
The argument that artists should only trust the process and not the aesthetic is not quite the bright advice that it may appear to be on the surface. It tells the stuck artist: you’re not failing, you’re investigating. But there’s no way to actually verify that from where they’re standing. It could just as easily be avoidance dressed up as rigor; Using “I’m still in the studies phase” as a permanent excuse to never commit to anything legible. It wants “investigation” to be self-evidently more legitimate than “choosing a style,” but investigation only earns that legitimacy retroactively, once it produces something. Until then it’s indistinguishable from just not knowing what you’re doing yet. The actual felt experience of “building a visual language” and the actual felt experience of “being lost” are the same experience, from inside it. The only thing that separates them is whether it eventually coheres — and that’s not something you can assess in the moment, only in hindsight.
What’s underneath all of it, however, is a genuinely good question wearing a slightly too-tidy costume: not “style vs. visual language” but “is your practice generating information, or just repeating a decision.” That’s the actual useful test, and it doesn’t require the binary to be clean. And it connects back to the first point in an uncomfortable way: if you can’t tell investigation from wandering except in hindsight, and if the process is recursive rather than staged, then there’s no moment where someone could hand you a checklist and say “you are now doing the deep thing instead of the shallow thing.” The accident-becomes-habit-becomes-material-becomes-new-accident cycle doesn’t pause for diagnosis. You’re always somewhere in the loop, and “which part of the loop am I in” isn’t actually answerable from inside a single session or even a single body of work — you’d need distance on your own practice that most people don’t have while they’re still making the thing.
The real skill isn’t “investigate instead of choosing a style” — it’s building enough reflective distance on your own loop to notice when a habit has calcified into decoration versus when it’s still actively generating new material. That’s a much harder thing to teach, because it’s not a sequence, it’s closer to a maintained relationship with your own work. The villain in this story — choosing an aesthetic, reduced to the question how do I make this look consistent — is a strawman built to make the investigating a visual language look better because consistency-seeking is itself a legitimate and sometimes rigorous formal question. A huge amount of what you’d call visual language is the discipline of consistency under constraint. Consistency under constraint is one of the oldest and hardest problems in any studio practice. It’s not a lesser question dressed up as a lazy one. Ask anyone working in a tight palette, a fixed compositional grid, a repeated figure — the discipline of holding a constraint without it going dead is itself a form of investigation. You don’t get to consistency by accident. You get there by testing where the constraint breaks, where it can flex, what happens when you push against it.
The zero-sum argument’s binary defines “aesthetic-chasing” by its worst-case version (arbitrary, decorative, unreflective) and defines “investigation” by its best-case version (rigorous, sustained, revealing). Compared fairly — rigorous consistency-seeking versus rigorous investigation — the distinction mostly disappears. What’s left is just: are you doing the hard version of your practice or the easy version. Consistency can be either. So can investigation. So the real variable was never which question a person is asking. It was never style versus language, surface versus depth, the palette versus the sustained inquiry. Those are just different rooms a decision can be tested in. The real variable is whether the decision is still being tested at all.
Is the work generating information, or is it just repeating a decision.
Because “generating information” has a testable trace even when you can’t feel it in the moment. You can look back at a body of work and ask whether piece 40 knows something piece 4 didn’t, versus piece 40 just being piece 4 again in a different palette. It doesn’t require hindsight to validate the process, only to audit it. That’s a smaller, more honest claim than “trust the investigation.”
It absorbs the recursion problem, because “generating information” doesn’t need a sequence. It can happen inside a loop — habit becomes material becomes accident becomes habit — as long as each pass through the loop changes what you know. The test isn’t which stage you’re in. It’s whether the loop is closed or open. A closed loop repeats. It takes the same gesture and returns the same gesture, canvas after canvas, and calls the returning consistency. An open loop repeats too but it lets the repetition talk back to it. The brushstroke is the same brushstroke but what it knows has changed.
This is a harder thing to build a pedagogical program around, because it cannot be taught as a sequence and it does not resolve into a checklist. It is not “do this, then that”. It is a maintained relationship with your own habits — a willingness to keep asking a decision to justify itself, long after it has become comfortable enough to stop asking. Most work that looks alive is not the product of someone who found their visual language before their style. It is the product of someone who never let a decision go unexamined long enough to calcify.
The mark of a shallow practice is not the wrong choice. It is a choice that stopped being a question.