This approach is rooted in the theory that language is not “representative” of the world, but significantly constitutive of it. In this view, language is not so

This approach is rooted in the theory that language is not “representative” of the world, but significantly constitutive of it. In this view, language is not something that mechanically “points to” things, but is the system through which consciousness develops and allows us to reason about sense-impressions in a way that is uniquely human (or at least sapient). Pragmatically, sense-impressions are organized into “things we can talk about” and, having named them, we now use those names and concepts to organize the world we perceive. The flip side is that we’re less likely to develop new names for and new reason-abouts things if they comfortably fit within our existing nominal categories, unless we have good practical reason to do so. Thus a visual designer will have a great many names for gradations in color that a non-designer might not be able to reason about (“It’s all just blue, right?”), because he or she has a practical need to distinguish colors and think about them at a much finer level of granularity. This also has the interesting derivation that societies can only reason about what they can talk about, which has profound implications for how inter-group communicative discourse works — from politics to intra-corporate coordination to literature. If you’re interested in this, I’d recommend any synoptic overview of Pragmatic and New Pragmatic thought, and the linguistic work of Searle.