This principle is a key flaw in the idea: "A complex domain can not be efficiently expressed as a single universal model and language, and must therefore be separated into Bounded Contexts (ie. an internally consistent language and model) by the system designers". We humans understand how everything we experience is part of a wider world. So if you are unable to know how things relate, as all things do, you do not yet understand the domain.
I don’t think it’s accurate to say that DDD advocates ignoring how things relate. It’s more like finding the higher level concepts that do relate to each other. You don’t need to know about internal combustion to write traffic rules. In fact, it’s better if you don’t know about that because your rules shouldn’t need to change when the mechanism that moves the car changes.
Separating things into bounded contexts sounds like specifically not connecting them, which means not understanding how they connect or even trying. And its these bounded things that seem to cause the biggest issues later on when their actual connections come to the surface as bugs in the system.
