Timeline for Is a finite and acyclic causal model of the world logically consistent?
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| yesterday | comment | added | dt128 | In the proof of Lemma B, this consideration has been precisely taken into account: the inclusion of the "nothing" and the "nothing effect" into the proof was for this reason. The proof of the initial version was much simpler. In a finite acyclic causal chain, there is both a starting node and an ending node. However, the ending node cannot be the cause of anything, so it has no effect on anything. Therefore, it does not exist (this was the reasoning gap). But the ending node does exist, which leads to a contradiction. | |
| yesterday | comment | added | mudskipper | The main point is really that if you assume there is a state of the world (or many states) that is an "effect" but not a "cause", then this by itself does not make the concept of causality inconsistent. If the universe ends in a big crunch, this would by definition not an observable effect, but it is a conceivable effect. And whether or not the universe will end like that is still a question of empirical theory. | |
| yesterday | comment | added | dt128 | There seems to be a misunderstanding about the uniqueness of the starting node and the connectivity of the graph. This text only assumes the graph is finite and without loops, which results in having some starting nodes (not necessarily one) and some ending nodes. Lemma B states that this exact model of causality is logically inconsistent. | |
| 2 days ago | comment | added | mudskipper | Yes, I saw that. But that graph is a pure fiction. There are no a priori reasons to assume a finite graph with a unique root, and no loops (basically a finite tree) as in anyway representing physical reality. If you consider effective causes, then there is also no reason to for instance (1) deny that infinite regression is impossible (2) assume that there are no loops (3) assume that every node is not connected to every other node. A fully connected finite or infinite graph would perhaps be a strange model of effective causality, but still not an inconsistent one. | |
| 2 days ago | comment | added | dt128 | the discussion is about a graph; the term "chain/graph of causality" is explicitly mentioned in the text. The conditions of this chain/graph have also moved from the traditional philosophical form to its mathematical representation. As for Hume, he did not address the logical conception of a chain/graph of causality. Specifically, he stated that the estimation of such a relationship—that is, the estimation of a causal relationship between two entities—always involves a degree of error and cannot be a matter of pure logic. | |
| Dec 9 at 15:50 | history | answered | mudskipper | CC BY-SA 4.0 |