= 27d567bc2d48d9f40e9ccc20ea370b15)
komponisto comments on A sense of logic - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (269)
An understandable point of view, but see here.
I actually intended to comment in a manipulationist/Pearlian spirit so I'm not sure what we disagree about. You may have to be more particular than linking to his entire book. To be clear, I'm not saying causality requires a human intervention or advocating any kind of agency theory of causality. I'm saying, like Pearl, that causal explanations tell us what would happen given intervention on a variable. Anthropologically, our conception of causality likely arose as we learned to make things happen, i.e. intervene. But it's difficult to know what 'intervention' means in domains far from human manipulation. Inventing a human-like mind that can do that kind of intervention feels like a kind of invention that would let people feel more comfortable talking about first causes.
Ah, okay, thanks for the clarification. I did in fact interpret you as proposing an agency theory of causality; I think the following helped to mislead me in that direction:
I think I actually wanted to refer to the preface of Pearl's book, in particular this part:
My implication would have been that causality is not a merely human folk-concept that makes sense only in the context of "agents"; but if you weren't saying otherwise, this is of course moot.
So I should clarify a bit more. We can understand causality as the following: A causes B iff an intervention on A alters the value of B. The concept of 'intervention' is doing the work here. Agency theories reduce 'intervention' to possible actions of free agents. I prefer Pearl's approach which fails to reduce the concept of intervention to non-causal concepts but a) doesn't obviously fail to actually describe our concept of cause the way the agency approach does and b) is a lot more illuminating. However, that doesn't mean our concept of intervention doesn't have any element of agency to it or that the two are entirely distinct. At the least, it seems plausible our evolved understanding of causality is interconnected with our evolved concept of agency, even if causality can be discussed mathematically in isolation.
An intervention on the fine structure constant is more mysterious to me than an intervention on the velocity of a pool ball- and it isn't necessarily just one being within my current capabilities and the other not. And for obvious reasons understanding a first cause under a manipulationist approach is really sketchy.