PhilSoc (30/6) - Joshua Pearson
Jerome Luxon
|
07/07/26
|
Nicholas Carroll
|
Open to all
|
|
14/07/26
|
Brian Hedden
|
Open to all
|
|
21/07/26
|
Alex Sandgren
|
Open to all
|
|
28/07/26
|
James Vlachoulis
|
Confirmation
|
Abstract: Breaking News! Colin and Frank are both in their offices. Colin flipped a fair coin and it landed heads. Meanwhile, Frank didn't scratch his nose. Assuming their offices are causally isolated, the following counterfactual seems true:
(1) Had Frank scratched his nose, Colin's coin still would have landed heads.
The intuition that (1) and counterfactuals like it are true is extremely strong and motivates a general principle:
Causal Independence Principle (CIP, spoken ‘kip’): If A and C are true, and the mechanisms settling whether A and whether C are causally independent, then: if A had been false, C would still have been true.
Many philosophers are attracted to something like CIP. Influential causal modeling approaches to counterfactuals are practically built so that CIP is true. Indeed, I find CIP extremely plausible too. Still, I'll argue it's false. I’ll start by giving various arguments against CIP which assume another principle, Duality, is true. I’ll then give even more arguments against CIP that assume Duality is false. Either way, then, things are looking bad for CIP. A running theme is that CIP requires our logic for counterfactuals to be highly revisionary. I’ll then consider why CIP seemed true if it is false. I am sympathetic to a contextualist story. CIP seems true because counterfactual truth depends on which facts context tells us to hold fixed. While context often requires keeping facts causally independent from the antecedent fixed, it need not.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Jerome Luxon
PhD Candidate
School of Philosophy
Research School of Social Sciences
Australian National University