Why have I spent so much time working on this essay? It boils down to the fact that even after reading DD and KP, I found HM's argument *confusing*. It was hard to *confidently* argue against. But *why*? Why was it so hard to refute? What was throwing me off? What was leading me to feel such a persistent sense of friction and anxiety? What made this argument so slippery and difficult to pin down? Answering *this question* is the purpose of this essay. It is to *explain* why I felt confused. To do so requires refuting HM's argument. But it requires doing so in a way that makes it clear *why* it felt semi-compelling in the first place. This piece was never just *only* about identifying flaws. It was about explaining why those flaws felt so disorienting. It was because this was a bad argument that pretended to be a good one. There are two main threads I'd like to highlight when building a cohesive narrative (and providing a window for me reader to look through): 1. HM is not seeking good explanations 2. Instead, HM is possibility pumping (which looks like first principles thinking) It is both (1) and (2) *together* that made this hard to refute. (1) on it's own would likely be easy to identify (think of a myth, a very bad explanation). But possibility pumping *disguises this*—it makes it look as similar to FPT and seeking good explanations as possible. Here are my supporting examples: 1. Contradiction (in definition of simulation) 1. Sometimes, his definition of simulation depends on an internal observer; sometimes, it doesn’t. That’s not just vague—it’s contradictory. And if a key term in your argument isn’t even internally consistent, then you’re not working with a good explanation. You’ve got something that fails on its own terms. So how did he end up here? To me, the only plausible reason is that he wasn’t actually trying to find a good explanation. He was chasing possibilities, not truth. 2. Accept all possible decodings 2. This is another clear case of possibility pumping. The idea that we should accept _all_ possible decodings is, again, a terrible explanation. And what’s more, it intersects nicely with a Gödelesque angle—because Gödel’s work, in contrast, _was_ a good explanation. It wasn’t about accepting all logically possible interpretations. Gödel made a very specific, hard-to-vary claim. His argument had explanatory power. That’s the key distinction: Gödel succeeded because his logic was tightly coupled to a deep explanation, not just a parade of possible mappings. Drawing out this nuance is helpful—it sharpens the contrast between real explanatory thinking. 3. Anything can be interpreted as a simulation of anything 1. If your definition of simulation leads you to conclude that _anything_ can be a simulation of _anything_, then you’ve completely hollowed out the concept. It’s lost all explanatory power. You’re not illuminating anything anymore—you’re just dissolving meaning. And again, I think the only way you get there is if you’re _not_ in the business of seeking hard-to-vary explanations. You’re just possibility-pumping, hoping to stumble across something that feels novel. TODO: where does semantic bait-and-switch go? Moravec is [Logical Possibility Pumping](Logical%20Possibility%20Pumping.md).