2007
DOI: 10.1007/s11098-007-9105-0
|View full text |Cite
|
Sign up to set email alerts
|

Causation and contrast classes

Abstract: I argue that causation is a contrastive relation: c-rather-than-C* causes e-rather-than-E*, where C* and E* are contrast classes associated respectively with actual events c and e. I explain why this is an improvement on the traditional binary view, and develop a detailed definition. It turns out that causation is only well defined in 'uniform' cases, where either all or none of the members of C* are related appropriately to members of E*.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

1
44
0

Year Published

2009
2009
2017
2017

Publication Types

Select...
4
4

Relationship

2
6

Authors

Journals

citations
Cited by 56 publications
(45 citation statements)
references
References 14 publications
1
44
0
Order By: Relevance
“…Moreover the restriction to binary variables can naturally be lifted by introducing a second layer of contrastivity, beyond the presence of the range function in the signature, into the structure of token causation itself (see generally Hitchcock 1996;Maslen 2004;Schaffer 2005;Northcott 2008):…”
mentioning
confidence: 99%
“…Moreover the restriction to binary variables can naturally be lifted by introducing a second layer of contrastivity, beyond the presence of the range function in the signature, into the structure of token causation itself (see generally Hitchcock 1996;Maslen 2004;Schaffer 2005;Northcott 2008):…”
mentioning
confidence: 99%
“…In particular, relativization to contrasts introduces a pragmatic element that has traditionally been thought characteristic only of explanation. On the other hand, there is a growing segment of the literature that does endorse an explicitly contrastive view of causation itself (Northcott 2008b, Schaffer 2005, Maslen 2004. 18 This view is also endorsed by the contemporary Bayes net and causal modeling literatures (Pearl 2000, Spirtes et al 2000, as well as, as already noted, by Woodward (2003).…”
Section: ) Relation To Previous Literaturementioning
confidence: 99%
“…This initial step involves identifying the topic of the question by bringing it into relief against a contrast class, in a sense defining the explanandum by highlighting a set of other possible, but non-occurring, events. The notion of contrast class is not necessarily intuitive, but it has been successfully utilized in related domains such as the role of focal sets in constraining computations of covariation when assessing causal induction (Cheng and Novick 1992: see also Northcott 2008). The contrast class defines the question that the explanation must answer by pointing to other possible, contrastive questions that could be answered.…”
Section: Essentialist Theories Of Explanationmentioning
confidence: 99%