Abstract
This work presents a defense of causal contrastivism based on causal contexualism. As argued, our proposal on causal contextualism is compatible with both causal contrastivism and causal binarism, including explanations of why and in which sense secondary counterfactuals are relevant.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
This can be explained considering that Steglich-Petersen is particularly focused on factual circumstances rather than other types of conditions regarding what is taken to be possible or relevant and what not. In other words, his binary account is compatible with contextualism; it is just based on a too strict kind of context.
We would not say that every binary causal claim should be formulated as a contrastive causal claim. We would agree with contrastivism that binary causal claims alone are incomplete, but adding contrast classes to construct sentences of the form (C, C*) R (E, E*) is not the only way to complete them. Sometimes, they might be completed by background assumptions associated with probabilistic inequalities; sometimes, they might be better completed by the set of conditions of a causal model; sometimes, by the history of the events involved, together with the relevant regularities; and other times they might be completed by a physicalist set of conditions and conservation laws. There is no absolute, context-independent fact of the matter about which one of these methods is the best.
This is in line with assessment relativism a la MacFarlane (2014): What matters is the context of assessment of truth, not the context of use.
When considering contextualist answers to these problems, Steglich-Petersen is inclined to think that they are trivial, in the sense that they are just based on the idea that in some cases we have more information than in others (cf. Steglich-Petersen 2012, p. 134). But since contexts can be conceived as being more than mere chunks of information, causal contextualism is not trivial.
The ascriber of some relation might not necessarily be also the assessor for the truth value of claims establishing that relation. However, we use “ascriber” for someone who both ascribes a relation and also assesses it for truth.
Why do not we just take contexts as conversational contexts? This could be tricky. First, one could feel that, in this sense, there is a set of facts of the matter according to which we may fix a context. Second, we do not exclude the idea that some contexts could be considered to be conversational contexts. In those cases, they are still epistemic items, in our sense. For instance, a couple of speakers may ascribe together a causal relation between two events, even if they disagree with regard to other ascriptions. They are ascribers, both separately and as a conversating couple of speakers.
As one may note, definition 2.2 seems almost indistinguishable from David Lewis’s (1973) classical counterfactual analysis of causation. The differences are crucial, however. While Lewis’s is an analysis of causation conceived as a metaphysical relation that is based on possible worlds semantics, our proposal on causal dependence is about an epistemic relation between contexts that are understood as epistemic items.
Two things are context-dependent in an important way, regarding counterfactual dependence in particular. One is the selection of relevant contrasts (i.e. which possibilities are relevant) and the other is the closeness relation according to which we evaluate counterfactuals (i.e. how relevant is each possibility compared to another).
Reiss (2013) offers an interesting account on how functions from conversational contexts to contrast classes may operate on the basis of commitments, intentions, physical conditions, presuppositions and laws considered by an agent. We will not discuss now in which ways his account could be compatible with (and extend) our proposal. However, two general remarks might be important. First, his definition of a context in terms of the mentioned factors could be seen as a specific example of our characterization of an epistemic context in terms of descriptive, normative and nominative expressions. Second, we do not agree with Reiss’s thesis that contrastive causal claims become redundant when the context selects a unique admissible alternative. Our proposal is perhaps less restrictive in this sense.
Causal contextualism, in a general form, does not entail causal contrastivism. Different accounts of causality may be based on context-dependence without assuming a transformation of binary causal claims into ternary or quaternary contrastive claims.
But, of course, there are other marks of relevance, such as anything that tells us something about interests, norms or standards of evaluation. For instance, an ascriber may ask about the truth of 5.8 because she could be interested in the efficiency of car locksmiths or in whether Susan trusts them. And she could as this even having scenario 2.5 in mind, where calling a locksmith was not a possibility for Susan. It might be a relevant possibility for the ascriber, though. In which sense can normativity be a key aspect of epistemic contexts will be the topic of the next section.
Causal invariantists may still want to claim that K and K’ involve different circumstances. At some point, as contextualists, we may agree with them. Anyway, our case needs the distinction between descriptions of the factual circumstances and other epistemic functions, such as norms or standards. Blurring the divide between them may lead to further problems that we cannot tackle here.
We can distinguish three different issues here: the normativity of contexts, the secondary counterfactuals, and the appropriateness conditions. Contextual normativity may determine the truth of secondary counterfactuals, but just when they come up as relevant. Many times, it might not be appropriate to consider them in the evaluation of a causal claim.
Interestingly, cases of overdetermination and redundancy in general have served as paradigmatic counterexamples against the simple, binary counterfactual account of causation.
References
Achinstein P (1975) Causation, transparency, and emphasis. Can J Philos 5(1):1–23
Dretske F (1977) Referring to events. Midwest Stud Philos 2(1):90–99
Hitchcock CR (1995) The mishap at Reichenbach fall: Singular vs. general causation. Philos Stud 78(3):257–291
Hitchcock C (1996) The role of contrast in causal and explanatory claims. Synthese 107(3):395–419
Lewis D (1973) Causation. J Philos 70(17):556–567
MacFarlane J (2014) Assessment sensitivity: Relative truth and its applications. Oxford University Press
Northcott R (2008) Causation and Contrast Classes. Philos Stud 139(1):111–123
Reiss J (2013) Contextualising Causation (Part I & Part II). Philos Compass 8(11):1066–1090
Schaffer J (2005) Contrastive causation. Philosophical Rev 114(3):327–358
Steglich-Petersen A (2012) Against the contrastive account of singular causation. Br J Philos Sci 63(1):115–143
Suppes P (1970) A probabilistic theory of causality. North-Holland Publishing Company
Weber E, De Bal I (2015) Contrastive causation in genetics and in physics. Philosophica 90:93–113
Woodward J (2003) Making things happen: A theory of causal explanation. Oxford University Press
Acknowledgements
This work has been supported by the Chilean Agency of Research and Development (projects #11180624 and #1211323).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Céspedes, E., Baumann, P. Contextual Causal Dependence and Causal Contrastivism. Axiomathes 32 (Suppl 3), 671–684 (2022). https://doi.org/10.1007/s10516-022-09622-6
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10516-022-09622-6