Pirsig observes that humans instinctively divide the world into categories to make sense of it. That’s how science, logic, and much of Western philosophy works: breaking things down, labelling the parts, and arranging them in order. “The application of this knife, the division of the world into parts… is something everybody does.”

In design systems, the “knife” is our categorisation: properties, tokens, variants, roles. I spent a lot of time creating a taxonomy of semantic color tokens, e.g. trying to decide whether my semantic color groups and therefore namings should be defined first by the design element where they are expected to appear (e.g., borders, fills, foregrounds, etc) or whether they are static or interactive (is this a color for a static page background or for an interactive button’s fill that should also have states like “hover” or “pressed”).

At some point, classification became the main task. I created a taxonomy that (in my head) worked perfectly well, but when other people started to apply it in their designs, it quickly became too rigid and not suitable for their specific context. For example, how does one chooses the color token classified as “on-bg” (on-background) vs “fg” (foreground) when we always have some kind of a background color beneath. And even when the logic of categorisation and naming did make sense (suprise, it actually happenned!), it could still produce the desing flow with components that looked odd when placed next to one another.

This is when I came across Pirsig’s quote: “When analytic thought, the knife, is applied to experience, something is always killed in the process.”

That “something” became the forest as a metaphor of the whole: the actual user experience, the product’s quality, or even the team’s shared understanding.